Dymas/abogen
1
0
Fork 0
mirror of https://github.com/denizsafak/abogen.git synced 2026-07-18 05:40:26 +02:00
Code Issues Packages Projects Releases Wiki Activity

chore: normalize line endings to LF, add .gitattributes

Browse Source 
- Add .gitattributes with text=auto eol=lf for all text files
- Renormalize all files in index to LF line endings
- Fixes massive whitespace-only diffs between main and feature branch
This commit is contained in:
Artem Akymenko
2026-07-12 16:20:42 +03:00
parent c85ea9d64f
commit d8fcfb1cce
60 changed files with 19745 additions and 19745 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/test_publish.yml
+63 -63
View File
@@ -1,63 +1,63 @@
name: Build multi-arch Docker Image
on:
# Build and push
#release:
# types: [published]
# Build only
#push: it
# branches: [main]
# TODO - enable build on pull requests if build times can be reduced
# pull_request:
workflow_dispatch:
env:
IMAGE_REPOSITORY: ghcr.io/denizsafak/abogen
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
- name: Login to Github Container Registry
# Only if we need to push an image
# if: ${{ github.event_name == 'release' && github.event.action == 'published' }}
uses: docker/login-action@v3
with:
registry: ghcr.io
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
# Setup for buildx
- name: Set up QEMU
uses: docker/setup-qemu-action@v3
- name: Set up Docker Buildx
id: buildx
uses: docker/setup-buildx-action@v3
# Debugging information
- name: Docker info
run: docker info
- name: Buildx inspect
run: docker buildx inspect
# Build and (optionally) push the image
- name: Build image
uses: docker/build-push-action@v6
with:
context: ./abogen
file: ./abogen/Dockerfile
# platforms: linux/amd64,linux/arm/v7,linux/arm64,linux/ppc64le,linux/s390x
# platforms: linux/amd64,linux/arm64
platforms: linux/amd64 # using the solution mentioned in https://github.com/denizsafak/abogen/issues/46
# Only push if we are publishing a release
# push: ${{ github.event_name == 'release' && github.event.action == 'published' }}
push: true
# Use a 'temp' tag, that won't be pushed, for non-release builds
tags: ${{ env.IMAGE_REPOSITORY }}:${{ github.event.release.tag_name || 'latest' }}
# Use a cache to reduce build times
cache-to: type=gha,mode=max
cache-from: type=gha
name: Build multi-arch Docker Image
on:
# Build and push
#release:
# types: [published]
# Build only
#push: it
# branches: [main]
# TODO - enable build on pull requests if build times can be reduced
# pull_request:
workflow_dispatch:
env:
IMAGE_REPOSITORY: ghcr.io/denizsafak/abogen
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
- name: Login to Github Container Registry
# Only if we need to push an image
# if: ${{ github.event_name == 'release' && github.event.action == 'published' }}
uses: docker/login-action@v3
with:
registry: ghcr.io
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
# Setup for buildx
- name: Set up QEMU
uses: docker/setup-qemu-action@v3
- name: Set up Docker Buildx
id: buildx
uses: docker/setup-buildx-action@v3
# Debugging information
- name: Docker info
run: docker info
- name: Buildx inspect
run: docker buildx inspect
# Build and (optionally) push the image
- name: Build image
uses: docker/build-push-action@v6
with:
context: ./abogen
file: ./abogen/Dockerfile
# platforms: linux/amd64,linux/arm/v7,linux/arm64,linux/ppc64le,linux/s390x
# platforms: linux/amd64,linux/arm64
platforms: linux/amd64 # using the solution mentioned in https://github.com/denizsafak/abogen/issues/46
# Only push if we are publishing a release
# push: ${{ github.event_name == 'release' && github.event.action == 'published' }}
push: true
# Use a 'temp' tag, that won't be pushed, for non-release builds
tags: ${{ env.IMAGE_REPOSITORY }}:${{ github.event.release.tag_name || 'latest' }}
# Use a cache to reduce build times
cache-to: type=gha,mode=max
cache-from: type=gha
abogen/assets/settings.svg
+30 -30
View File
@@ -1,31 +1,31 @@
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
<!-- Uploaded to: SVG Repo, www.svgrepo.com, Generator: SVG Repo Mixer Tools -->
<svg height="800px" width="800px" version="1.1" id="_x32_" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"
viewBox="0 0 512 512" xml:space="preserve">
<style type="text/css">
.st0{fill:#808080;}
</style>
<g>
<path class="st0" d="M502.325,307.303l-39.006-30.805c-6.215-4.908-9.665-12.429-9.668-20.348c0-0.084,0-0.168,0-0.252
c-0.014-7.936,3.44-15.478,9.667-20.396l39.007-30.806c8.933-7.055,12.093-19.185,7.737-29.701l-17.134-41.366
c-4.356-10.516-15.167-16.86-26.472-15.532l-49.366,5.8c-7.881,0.926-15.656-1.966-21.258-7.586
c-0.059-0.06-0.118-0.119-0.177-0.178c-5.597-5.602-8.476-13.36-7.552-21.225l5.799-49.363
c1.328-11.305-5.015-22.116-15.531-26.472L337.004,1.939c-10.516-4.356-22.646-1.196-29.701,7.736l-30.805,39.005
c-4.908,6.215-12.43,9.665-20.349,9.668c-0.084,0-0.168,0-0.252,0c-7.935,0.014-15.477-3.44-20.395-9.667L204.697,9.675
c-7.055-8.933-19.185-12.092-29.702-7.736L133.63,19.072c-10.516,4.356-16.86,15.167-15.532,26.473l5.799,49.366
c0.926,7.881-1.964,15.656-7.585,21.257c-0.059,0.059-0.118,0.118-0.178,0.178c-5.602,5.598-13.36,8.477-21.226,7.552
l-49.363-5.799c-11.305-1.328-22.116,5.015-26.472,15.531L1.939,174.996c-4.356,10.516-1.196,22.646,7.736,29.701l39.006,30.805
c6.215,4.908,9.665,12.429,9.668,20.348c0,0.084,0,0.167,0,0.251c0.014,7.935-3.44,15.477-9.667,20.395L9.675,307.303
c-8.933,7.055-12.092,19.185-7.736,29.701l17.134,41.365c4.356,10.516,15.168,16.86,26.472,15.532l49.366-5.799
c7.882-0.926,15.656,1.965,21.258,7.586c0.059,0.059,0.118,0.119,0.178,0.178c5.597,5.603,8.476,13.36,7.552,21.226l-5.799,49.364
c-1.328,11.305,5.015,22.116,15.532,26.472l41.366,17.134c10.516,4.356,22.646,1.196,29.701-7.736l30.804-39.005
c4.908-6.215,12.43-9.665,20.348-9.669c0.084,0,0.168,0,0.251,0c7.936-0.014,15.478,3.44,20.396,9.667l30.806,39.007
c7.055,8.933,19.185,12.093,29.701,7.736l41.366-17.134c10.516-4.356,16.86-15.168,15.532-26.472l-5.8-49.366
c-0.926-7.881,1.965-15.656,7.586-21.257c0.059-0.059,0.119-0.119,0.178-0.178c5.602-5.597,13.36-8.476,21.225-7.552l49.364,5.799
c11.305,1.328,22.117-5.015,26.472-15.531l17.134-41.365C514.418,326.488,511.258,314.358,502.325,307.303z M281.292,329.698
c-39.68,16.436-85.172-2.407-101.607-42.087c-16.436-39.68,2.407-85.171,42.087-101.608c39.68-16.436,85.172,2.407,101.608,42.088
C339.815,267.771,320.972,313.262,281.292,329.698z"/>
</g>
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
<!-- Uploaded to: SVG Repo, www.svgrepo.com, Generator: SVG Repo Mixer Tools -->
<svg height="800px" width="800px" version="1.1" id="_x32_" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"
viewBox="0 0 512 512" xml:space="preserve">
<style type="text/css">
.st0{fill:#808080;}
</style>
<g>
<path class="st0" d="M502.325,307.303l-39.006-30.805c-6.215-4.908-9.665-12.429-9.668-20.348c0-0.084,0-0.168,0-0.252
c-0.014-7.936,3.44-15.478,9.667-20.396l39.007-30.806c8.933-7.055,12.093-19.185,7.737-29.701l-17.134-41.366
c-4.356-10.516-15.167-16.86-26.472-15.532l-49.366,5.8c-7.881,0.926-15.656-1.966-21.258-7.586
c-0.059-0.06-0.118-0.119-0.177-0.178c-5.597-5.602-8.476-13.36-7.552-21.225l5.799-49.363
c1.328-11.305-5.015-22.116-15.531-26.472L337.004,1.939c-10.516-4.356-22.646-1.196-29.701,7.736l-30.805,39.005
c-4.908,6.215-12.43,9.665-20.349,9.668c-0.084,0-0.168,0-0.252,0c-7.935,0.014-15.477-3.44-20.395-9.667L204.697,9.675
c-7.055-8.933-19.185-12.092-29.702-7.736L133.63,19.072c-10.516,4.356-16.86,15.167-15.532,26.473l5.799,49.366
c0.926,7.881-1.964,15.656-7.585,21.257c-0.059,0.059-0.118,0.118-0.178,0.178c-5.602,5.598-13.36,8.477-21.226,7.552
l-49.363-5.799c-11.305-1.328-22.116,5.015-26.472,15.531L1.939,174.996c-4.356,10.516-1.196,22.646,7.736,29.701l39.006,30.805
c6.215,4.908,9.665,12.429,9.668,20.348c0,0.084,0,0.167,0,0.251c0.014,7.935-3.44,15.477-9.667,20.395L9.675,307.303
c-8.933,7.055-12.092,19.185-7.736,29.701l17.134,41.365c4.356,10.516,15.168,16.86,26.472,15.532l49.366-5.799
c7.882-0.926,15.656,1.965,21.258,7.586c0.059,0.059,0.118,0.119,0.178,0.178c5.597,5.603,8.476,13.36,7.552,21.226l-5.799,49.364
c-1.328,11.305,5.015,22.116,15.532,26.472l41.366,17.134c10.516,4.356,22.646,1.196,29.701-7.736l30.804-39.005
c4.908-6.215,12.43-9.665,20.348-9.669c0.084,0,0.168,0,0.251,0c7.936-0.014,15.478,3.44,20.396,9.667l30.806,39.007
c7.055,8.933,19.185,12.093,29.701,7.736l41.366-17.134c10.516-4.356,16.86-15.168,15.532-26.472l-5.8-49.366
c-0.926-7.881,1.965-15.656,7.586-21.257c0.059-0.059,0.119-0.119,0.178-0.178c5.602-5.597,13.36-8.476,21.225-7.552l49.364,5.799
c11.305,1.328,22.117-5.015,26.472-15.531l17.134-41.365C514.418,326.488,511.258,314.358,502.325,307.303z M281.292,329.698
c-39.68,16.436-85.172-2.407-101.607-42.087c-16.436-39.68,2.407-85.171,42.087-101.608c39.68-16.436,85.172,2.407,101.608,42.088
C339.815,267.771,320.972,313.262,281.292,329.698z"/>
</g>
</svg>
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 2.6 KiB

After

Width:  |  Height:  |  Size: 2.5 KiB

abogen/predownload_gui.py
+591 -591
View File
File diff suppressed because it is too large Load Diff
abogen/pyqt/gui.py
+4304 -4304
View File
File diff suppressed because it is too large Load Diff
abogen/pyqt/predownload_gui.py
+591 -591
View File
File diff suppressed because it is too large Load Diff
abogen/pyqt/voice_formula_gui.py
+1598 -1598
View File
File diff suppressed because it is too large Load Diff
abogen/subtitle_utils.py
+584 -584
View File
File diff suppressed because it is too large Load Diff
abogen/tts_plugin/__init__.py
+170 -170
View File
@@ -1,170 +1,170 @@
"""TTS Plugin Architecture - Public API.
This package defines the frozen Plugin API for the TTS Plugin Architecture.
All public interfaces are fully defined but contain no business logic.
Public modules:
- types: Core domain value objects (AudioFormat, Duration, VoiceSelection, etc.)
- errors: Error hierarchy (EngineError and subtypes)
- manifest: Plugin manifest types (PluginManifest, EngineManifest, etc.)
- engine: Engine and EngineSession protocols
- capabilities: Optional capability interfaces (VoiceLister, PreviewGenerator, etc.)
- host_context: HostContext dataclass
- plugin: Plugin contract (create_engine function signature)
- loader: Plugin discovery and loading
- plugin_manager: Plugin management and engine creation
- utils: Direct utility functions (get_voices, create_pipeline, etc.)
Usage:
from abogen.tts_plugin import (
# Types
AudioFormat,
Duration,
VoiceSelection,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
EngineConfig,
# Errors
EngineError,
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
# Manifest
PluginManifest,
EngineManifest,
VoiceSourceManifest,
VoiceManifest,
ParameterManifest,
AudioFormatManifest,
EnumOption,
RequirementManifest,
GpuRequirement,
ModelManifest,
# Engine
Engine,
EngineSession,
# Capabilities
VoiceLister,
PreviewGenerator,
StreamingSynthesizer,
CancelableSession,
# Host Context
HostContext,
HttpClient,
# Plugin Manager
get_plugin_manager,
reset_plugin_manager,
# Utils
get_voices,
get_default_voice,
is_plugin_registered,
resolve_voice_to_plugin,
create_pipeline,
)
"""
from abogen.tts_plugin.capabilities import (
CancelableSession,
PreviewGenerator,
StreamingSynthesizer,
VoiceLister,
)
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import (
CancelledError,
ConfigurationError,
EngineError,
InternalError,
InvalidInputError,
ModelLoadError,
ModelNotFoundError,
NetworkError,
)
from abogen.tts_plugin.host_context import HttpClient, HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
EnumOption,
GpuRequirement,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
# Plugin Manager and Utils
from abogen.tts_plugin.plugin_manager import get_plugin_manager, reset_plugin_manager
from abogen.tts_plugin.utils import (
create_pipeline,
get_default_voice,
get_voices,
is_plugin_registered,
resolve_voice_to_plugin,
)
__all__ = [
# Types
"AudioFormat",
"Duration",
"VoiceSelection",
"ParameterValues",
"SynthesisRequest",
"SynthesizedAudio",
"EngineConfig",
# Errors
"EngineError",
"ModelNotFoundError",
"ModelLoadError",
"NetworkError",
"InvalidInputError",
"ConfigurationError",
"CancelledError",
"InternalError",
# Manifest
"PluginManifest",
"EngineManifest",
"VoiceSourceManifest",
"VoiceManifest",
"ParameterManifest",
"AudioFormatManifest",
"EnumOption",
"RequirementManifest",
"GpuRequirement",
"ModelManifest",
# Engine
"Engine",
"EngineSession",
# Capabilities
"VoiceLister",
"PreviewGenerator",
"StreamingSynthesizer",
"CancelableSession",
# Host Context
"HostContext",
"HttpClient",
# Plugin Manager
"get_plugin_manager",
"reset_plugin_manager",
# Utils
"get_voices",
"get_default_voice",
"is_plugin_registered",
"resolve_voice_to_plugin",
"create_pipeline",
]
"""TTS Plugin Architecture - Public API.
This package defines the frozen Plugin API for the TTS Plugin Architecture.
All public interfaces are fully defined but contain no business logic.
Public modules:
- types: Core domain value objects (AudioFormat, Duration, VoiceSelection, etc.)
- errors: Error hierarchy (EngineError and subtypes)
- manifest: Plugin manifest types (PluginManifest, EngineManifest, etc.)
- engine: Engine and EngineSession protocols
- capabilities: Optional capability interfaces (VoiceLister, PreviewGenerator, etc.)
- host_context: HostContext dataclass
- plugin: Plugin contract (create_engine function signature)
- loader: Plugin discovery and loading
- plugin_manager: Plugin management and engine creation
- utils: Direct utility functions (get_voices, create_pipeline, etc.)
Usage:
from abogen.tts_plugin import (
# Types
AudioFormat,
Duration,
VoiceSelection,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
EngineConfig,
# Errors
EngineError,
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
# Manifest
PluginManifest,
EngineManifest,
VoiceSourceManifest,
VoiceManifest,
ParameterManifest,
AudioFormatManifest,
EnumOption,
RequirementManifest,
GpuRequirement,
ModelManifest,
# Engine
Engine,
EngineSession,
# Capabilities
VoiceLister,
PreviewGenerator,
StreamingSynthesizer,
CancelableSession,
# Host Context
HostContext,
HttpClient,
# Plugin Manager
get_plugin_manager,
reset_plugin_manager,
# Utils
get_voices,
get_default_voice,
is_plugin_registered,
resolve_voice_to_plugin,
create_pipeline,
)
"""
from abogen.tts_plugin.capabilities import (
CancelableSession,
PreviewGenerator,
StreamingSynthesizer,
VoiceLister,
)
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import (
CancelledError,
ConfigurationError,
EngineError,
InternalError,
InvalidInputError,
ModelLoadError,
ModelNotFoundError,
NetworkError,
)
from abogen.tts_plugin.host_context import HttpClient, HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
EnumOption,
GpuRequirement,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
# Plugin Manager and Utils
from abogen.tts_plugin.plugin_manager import get_plugin_manager, reset_plugin_manager
from abogen.tts_plugin.utils import (
create_pipeline,
get_default_voice,
get_voices,
is_plugin_registered,
resolve_voice_to_plugin,
)
__all__ = [
# Types
"AudioFormat",
"Duration",
"VoiceSelection",
"ParameterValues",
"SynthesisRequest",
"SynthesizedAudio",
"EngineConfig",
# Errors
"EngineError",
"ModelNotFoundError",
"ModelLoadError",
"NetworkError",
"InvalidInputError",
"ConfigurationError",
"CancelledError",
"InternalError",
# Manifest
"PluginManifest",
"EngineManifest",
"VoiceSourceManifest",
"VoiceManifest",
"ParameterManifest",
"AudioFormatManifest",
"EnumOption",
"RequirementManifest",
"GpuRequirement",
"ModelManifest",
# Engine
"Engine",
"EngineSession",
# Capabilities
"VoiceLister",
"PreviewGenerator",
"StreamingSynthesizer",
"CancelableSession",
# Host Context
"HostContext",
"HttpClient",
# Plugin Manager
"get_plugin_manager",
"reset_plugin_manager",
# Utils
"get_voices",
"get_default_voice",
"is_plugin_registered",
"resolve_voice_to_plugin",
"create_pipeline",
]
abogen/tts_plugin/capabilities.py
+103 -103
View File
@@ -1,103 +1,103 @@
"""Capability interfaces for the TTS Plugin Architecture.
This module defines optional capability interfaces that engines can implement.
Capabilities are additive; implementing new capabilities doesn't break old plugins.
"""
from __future__ import annotations
from typing import Iterator, Protocol, runtime_checkable
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import SynthesisRequest, SynthesizedAudio, VoiceSelection
@runtime_checkable
class VoiceLister(Protocol):
"""Protocol for listing available voices.
Engines that support voice listing should implement this interface.
"""
def listVoices(self, sourceId: str) -> list[VoiceManifest]:
"""List available voices for a given source.
Args:
sourceId: The voice source identifier.
Returns:
List of VoiceManifest describing available voices.
Raises:
EngineError: On failure.
"""
...
@runtime_checkable
class PreviewGenerator(Protocol):
"""Protocol for generating voice previews.
Engines that support voice preview should implement this interface.
"""
def generatePreview(self, voice: VoiceSelection, text: str) -> SynthesizedAudio:
"""Generate a preview audio for a voice.
Args:
voice: Voice selection for the preview.
text: Text to use for the preview.
Returns:
SynthesizedAudio with the preview audio data.
Raises:
EngineError: On failure.
"""
...
@runtime_checkable
class StreamingSynthesizer(Protocol):
"""Protocol for streaming synthesis.
Optional capability of EngineSession, not Engine.
Engines that support streaming synthesis should implement this interface.
"""
def synthesizeStream(self, request: SynthesisRequest) -> Iterator[bytes]:
"""Synthesize audio in streaming mode.
Args:
request: The synthesis request.
Yields:
Audio chunks as they become available.
Raises:
CancelledError: If cancel() is called during iteration.
EngineError: On synthesis failure.
"""
...
# This is a generator function; implementation will use yield
yield b"" # pragma: no cover
@runtime_checkable
class CancelableSession(Protocol):
"""Protocol for cancellation support.
Optional capability for engines that support cancellation.
cancel() causes synthesize() to raise CancelledError.
"""
def cancel(self) -> None:
"""Cancel in-progress synthesis.
After cancellation, synthesize() raises CancelledError.
The session remains usable after cancellation.
Raises:
EngineError: If called after dispose().
"""
...
"""Capability interfaces for the TTS Plugin Architecture.
This module defines optional capability interfaces that engines can implement.
Capabilities are additive; implementing new capabilities doesn't break old plugins.
"""
from __future__ import annotations
from typing import Iterator, Protocol, runtime_checkable
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import SynthesisRequest, SynthesizedAudio, VoiceSelection
@runtime_checkable
class VoiceLister(Protocol):
"""Protocol for listing available voices.
Engines that support voice listing should implement this interface.
"""
def listVoices(self, sourceId: str) -> list[VoiceManifest]:
"""List available voices for a given source.
Args:
sourceId: The voice source identifier.
Returns:
List of VoiceManifest describing available voices.
Raises:
EngineError: On failure.
"""
...
@runtime_checkable
class PreviewGenerator(Protocol):
"""Protocol for generating voice previews.
Engines that support voice preview should implement this interface.
"""
def generatePreview(self, voice: VoiceSelection, text: str) -> SynthesizedAudio:
"""Generate a preview audio for a voice.
Args:
voice: Voice selection for the preview.
text: Text to use for the preview.
Returns:
SynthesizedAudio with the preview audio data.
Raises:
EngineError: On failure.
"""
...
@runtime_checkable
class StreamingSynthesizer(Protocol):
"""Protocol for streaming synthesis.
Optional capability of EngineSession, not Engine.
Engines that support streaming synthesis should implement this interface.
"""
def synthesizeStream(self, request: SynthesisRequest) -> Iterator[bytes]:
"""Synthesize audio in streaming mode.
Args:
request: The synthesis request.
Yields:
Audio chunks as they become available.
Raises:
CancelledError: If cancel() is called during iteration.
EngineError: On synthesis failure.
"""
...
# This is a generator function; implementation will use yield
yield b"" # pragma: no cover
@runtime_checkable
class CancelableSession(Protocol):
"""Protocol for cancellation support.
Optional capability for engines that support cancellation.
cancel() causes synthesize() to raise CancelledError.
"""
def cancel(self) -> None:
"""Cancel in-progress synthesis.
After cancellation, synthesize() raises CancelledError.
The session remains usable after cancellation.
Raises:
EngineError: If called after dispose().
"""
...
abogen/tts_plugin/engine.py
+95 -95
View File
@@ -1,95 +1,95 @@
"""Engine interfaces for the TTS Plugin Architecture.
This module defines the core Engine and EngineSession protocols.
These are the primary interfaces that plugin implementations must satisfy.
"""
from __future__ import annotations
from typing import Protocol, runtime_checkable
from abogen.tts_plugin.types import SynthesisRequest, SynthesizedAudio
@runtime_checkable
class EngineSession(Protocol):
"""Protocol for a session that owns mutable execution state.
An EngineSession is created by Engine.createSession() and owns
mutable execution state isolated from other concurrent work.
It is NOT thread-safe.
Lifecycle:
1. Created by Engine.createSession()
2. Used for synthesis via synthesize()
3. Disposed via dispose()
After dispose(), all methods except dispose() raise EngineError.
"""
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
"""Synthesize audio from text.
Args:
request: The synthesis request containing text, voice, parameters, and format.
Returns:
SynthesizedAudio with the synthesized audio data.
Raises:
EngineError: On synthesis failure. Session remains usable after error.
EngineError: If called after dispose().
"""
...
def dispose(self) -> None:
"""Release session resources.
This method is idempotent and safe to call multiple times.
It never raises exceptions (catches and logs internally).
After dispose(), all methods except dispose() raise EngineError.
"""
...
@runtime_checkable
class Engine(Protocol):
"""Protocol for a TTS engine that creates sessions.
An Engine is a factory for EngineSession instances. It is stateless
and thread-safe for createSession().
Lifecycle:
1. Created via create_engine() (plugin contract)
2. Sessions created via createSession()
3. Disposed via dispose()
Thread Safety:
- createSession() is thread-safe and can be called from any thread.
- dispose() must be called after all sessions are disposed.
- Disposing engine while sessions are alive violates API contract.
"""
def createSession(self) -> EngineSession:
"""Create a new session for synthesis.
Returns:
A new EngineSession instance. Ownership transfers to caller.
Raises:
EngineError: On failure. No partially initialized session is returned.
"""
...
def dispose(self) -> None:
"""Release engine resources.
Caller must ensure all sessions created by this engine are disposed
before calling dispose(). Disposing an engine while any session is
still alive violates the API contract; behavior is undefined.
This method is idempotent and safe to call multiple times.
It never raises exceptions (catches and logs internally).
After dispose(), all methods except dispose() raise EngineError.
"""
...
"""Engine interfaces for the TTS Plugin Architecture.
This module defines the core Engine and EngineSession protocols.
These are the primary interfaces that plugin implementations must satisfy.
"""
from __future__ import annotations
from typing import Protocol, runtime_checkable
from abogen.tts_plugin.types import SynthesisRequest, SynthesizedAudio
@runtime_checkable
class EngineSession(Protocol):
"""Protocol for a session that owns mutable execution state.
An EngineSession is created by Engine.createSession() and owns
mutable execution state isolated from other concurrent work.
It is NOT thread-safe.
Lifecycle:
1. Created by Engine.createSession()
2. Used for synthesis via synthesize()
3. Disposed via dispose()
After dispose(), all methods except dispose() raise EngineError.
"""
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
"""Synthesize audio from text.
Args:
request: The synthesis request containing text, voice, parameters, and format.
Returns:
SynthesizedAudio with the synthesized audio data.
Raises:
EngineError: On synthesis failure. Session remains usable after error.
EngineError: If called after dispose().
"""
...
def dispose(self) -> None:
"""Release session resources.
This method is idempotent and safe to call multiple times.
It never raises exceptions (catches and logs internally).
After dispose(), all methods except dispose() raise EngineError.
"""
...
@runtime_checkable
class Engine(Protocol):
"""Protocol for a TTS engine that creates sessions.
An Engine is a factory for EngineSession instances. It is stateless
and thread-safe for createSession().
Lifecycle:
1. Created via create_engine() (plugin contract)
2. Sessions created via createSession()
3. Disposed via dispose()
Thread Safety:
- createSession() is thread-safe and can be called from any thread.
- dispose() must be called after all sessions are disposed.
- Disposing engine while sessions are alive violates API contract.
"""
def createSession(self) -> EngineSession:
"""Create a new session for synthesis.
Returns:
A new EngineSession instance. Ownership transfers to caller.
Raises:
EngineError: On failure. No partially initialized session is returned.
"""
...
def dispose(self) -> None:
"""Release engine resources.
Caller must ensure all sessions created by this engine are disposed
before calling dispose(). Disposing an engine while any session is
still alive violates the API contract; behavior is undefined.
This method is idempotent and safe to call multiple times.
It never raises exceptions (catches and logs internally).
After dispose(), all methods except dispose() raise EngineError.
"""
...
abogen/tts_plugin/errors.py
+62 -62
View File
@@ -1,62 +1,62 @@
"""Error hierarchy for the TTS Plugin Architecture.
This module defines typed exceptions that engines raise.
Engines should never raise raw exceptions; they must use EngineError or its subtypes.
"""
from __future__ import annotations
class EngineError(Exception):
"""Base exception for all engine errors.
All engine operations that can fail should raise EngineError or one of its subtypes.
After dispose(), all methods except dispose() raise EngineError.
"""
pass
class ModelNotFoundError(EngineError):
"""Raised when a required model is not found."""
pass
class ModelLoadError(EngineError):
"""Raised when a model fails to load."""
pass
class NetworkError(EngineError):
"""Raised when a network operation fails."""
pass
class InvalidInputError(EngineError):
"""Raised when invalid input is provided to the engine."""
pass
class ConfigurationError(EngineError):
"""Raised when there is a configuration error."""
pass
class CancelledError(EngineError):
"""Raised when an operation is cancelled.
This is raised by synthesize() when cancel() is called during synthesis.
"""
pass
class InternalError(EngineError):
"""Raised when an internal engine error occurs."""
pass
"""Error hierarchy for the TTS Plugin Architecture.
This module defines typed exceptions that engines raise.
Engines should never raise raw exceptions; they must use EngineError or its subtypes.
"""
from __future__ import annotations
class EngineError(Exception):
"""Base exception for all engine errors.
All engine operations that can fail should raise EngineError or one of its subtypes.
After dispose(), all methods except dispose() raise EngineError.
"""
pass
class ModelNotFoundError(EngineError):
"""Raised when a required model is not found."""
pass
class ModelLoadError(EngineError):
"""Raised when a model fails to load."""
pass
class NetworkError(EngineError):
"""Raised when a network operation fails."""
pass
class InvalidInputError(EngineError):
"""Raised when invalid input is provided to the engine."""
pass
class ConfigurationError(EngineError):
"""Raised when there is a configuration error."""
pass
class CancelledError(EngineError):
"""Raised when an operation is cancelled.
This is raised by synthesize() when cancel() is called during synthesis.
"""
pass
class InternalError(EngineError):
"""Raised when an internal engine error occurs."""
pass
abogen/tts_plugin/host_context.py
+46 -46
View File
@@ -1,46 +1,46 @@
"""Host context for the TTS Plugin Architecture.
This module defines the HostContext dataclass that provides minimal
host services to plugins. It is the only interface through which
plugins can access host functionality.
"""
from __future__ import annotations
import logging
from dataclasses import dataclass
from pathlib import Path
from typing import Protocol, runtime_checkable
@runtime_checkable
class HttpClient(Protocol):
"""Protocol for HTTP client provided by host.
Plugins can use this for network requests (e.g., API-based engines).
"""
def get(self, url: str, **kwargs: object) -> object:
"""Perform an HTTP GET request."""
...
def post(self, url: str, **kwargs: object) -> object:
"""Perform an HTTP POST request."""
...
@dataclass(frozen=True)
class HostContext:
"""Minimal host context provided to plugins.
Contains only essential host services. No business logic.
Attributes:
config_dir: Directory for API keys, preferences, and configuration.
logger: Logger for plugin logging.
http_client: HTTP client for network requests.
"""
config_dir: Path
logger: logging.Logger
http_client: HttpClient
"""Host context for the TTS Plugin Architecture.
This module defines the HostContext dataclass that provides minimal
host services to plugins. It is the only interface through which
plugins can access host functionality.
"""
from __future__ import annotations
import logging
from dataclasses import dataclass
from pathlib import Path
from typing import Protocol, runtime_checkable
@runtime_checkable
class HttpClient(Protocol):
"""Protocol for HTTP client provided by host.
Plugins can use this for network requests (e.g., API-based engines).
"""
def get(self, url: str, **kwargs: object) -> object:
"""Perform an HTTP GET request."""
...
def post(self, url: str, **kwargs: object) -> object:
"""Perform an HTTP POST request."""
...
@dataclass(frozen=True)
class HostContext:
"""Minimal host context provided to plugins.
Contains only essential host services. No business logic.
Attributes:
config_dir: Directory for API keys, preferences, and configuration.
logger: Logger for plugin logging.
http_client: HTTP client for network requests.
"""
config_dir: Path
logger: logging.Logger
http_client: HttpClient
abogen/tts_plugin/loader.py
+365 -365
View File
@@ -1,365 +1,365 @@
"""Plugin loader infrastructure for the TTS Plugin Architecture.
This module provides functionality to discover, import, validate, and load
TTS plugins. It handles both valid and invalid plugins, providing diagnostic
messages for errors.
The loader does NOT:
- Create Engine instances (that's the plugin's create_engine() responsibility)
- Manage plugin lifecycle (that's the Plugin Manager's responsibility)
- Implement any TTS engine functionality
"""
from __future__ import annotations
import importlib
import re
import sys
import types
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any, Callable
from abogen.tts_plugin.manifest import ModelManifest, PluginManifest
# Host API version for compatibility checking
HOST_API_VERSION = "1.0"
@dataclass(frozen=True)
class PluginLoadError:
"""Diagnostic information for a failed plugin load.
Attributes:
plugin_id: Plugin identifier if available, otherwise directory name.
path: Path to the plugin directory.
errors: List of error messages describing what went wrong.
"""
plugin_id: str
path: Path
errors: tuple[str, ...] = field(default_factory=tuple)
@dataclass(frozen=True)
class PluginLoadResult:
"""Result of loading a plugin.
Attributes:
success: Whether the plugin loaded successfully.
manifest: The plugin manifest if successful.
model_requirements: Model requirements if successful.
create_engine: The create_engine function if successful.
module: The plugin module if successful.
error: Error information if failed.
"""
success: bool
manifest: PluginManifest | None = None
model_requirements: tuple[ModelManifest, ...] | None = None
create_engine: Callable[..., Any] | None = None
module: types.ModuleType | None = None
error: PluginLoadError | None = None
def _parse_api_version(version: str) -> tuple[int, int] | None:
"""Parse an api_version string into (major, minor) tuple.
Args:
version: Version string in format "MAJOR.MINOR".
Returns:
Tuple of (major, minor) or None if invalid format.
"""
match = re.match(r"^(\d+)\.(\d+)$", version)
if match:
return int(match.group(1)), int(match.group(2))
return None
def _check_api_version_compatibility(plugin_version: str) -> str | None:
"""Check if plugin api_version is compatible with host.
Architecture spec:
- Format: semver (MAJOR.MINOR)
- Compatibility: Host rejects plugin if major version differs
- Minor version: backward compatible, Host accepts higher minor
Args:
plugin_version: Plugin's api_version string.
Returns:
Error message if incompatible, None if compatible.
"""
plugin_ver = _parse_api_version(plugin_version)
if plugin_ver is None:
return f"Invalid api_version format: '{plugin_version}'. Expected format: MAJOR.MINOR"
host_ver = _parse_api_version(HOST_API_VERSION)
if host_ver is None:
return f"Invalid host api_version format: '{HOST_API_VERSION}'"
if plugin_ver[0] != host_ver[0]:
return (
f"api_version major mismatch: plugin={plugin_ver[0]}, host={host_ver[0]}. "
f"Major version must match for compatibility."
)
return None
def _validate_manifest(module: types.ModuleType, plugin_dir: Path) -> list[str]:
"""Validate that a plugin module has required exports.
Args:
module: The imported plugin module.
plugin_dir: Path to the plugin directory.
Returns:
List of error messages (empty if valid).
"""
errors: list[str] = []
# Check PLUGIN_MANIFEST
manifest = getattr(module, "PLUGIN_MANIFEST", None)
if manifest is None:
errors.append("Missing PLUGIN_MANIFEST export")
elif not isinstance(manifest, PluginManifest):
errors.append(
f"PLUGIN_MANIFEST must be a PluginManifest instance, "
f"got {type(manifest).__name__}"
)
# Check MODEL_REQUIREMENTS
model_reqs = getattr(module, "MODEL_REQUIREMENTS", None)
if model_reqs is None:
errors.append("Missing MODEL_REQUIREMENTS export")
elif not isinstance(model_reqs, list):
errors.append(
f"MODEL_REQUIREMENTS must be a list, got {type(model_reqs).__name__}"
)
else:
for i, req in enumerate(model_reqs):
if not isinstance(req, ModelManifest):
errors.append(
f"MODEL_REQUIREMENTS[{i}] must be a ModelManifest instance, "
f"got {type(req).__name__}"
)
# Check create_engine
create_engine = getattr(module, "create_engine", None)
if create_engine is None:
errors.append("Missing create_engine export")
elif not callable(create_engine):
errors.append(
f"create_engine must be callable, got {type(create_engine).__name__}"
)
return errors
def _validate_capabilities(manifest: PluginManifest) -> list[str]:
"""Validate plugin capabilities.
Args:
manifest: The plugin manifest to validate.
Returns:
List of error messages (empty if valid).
"""
errors: list[str] = []
# Known capabilities (can be extended)
known_capabilities = frozenset({
"voice_list",
"preview",
"voice_clone",
"voice_blend",
"streaming",
"cancel",
})
for cap in manifest.capabilities:
if cap not in known_capabilities:
errors.append(f"Unknown capability: '{cap}'")
return errors
def _validate_api_version(manifest: PluginManifest) -> list[str]:
"""Validate api_version compatibility.
Args:
manifest: The plugin manifest to validate.
Returns:
List of error messages (empty if valid).
"""
errors: list[str] = []
error = _check_api_version_compatibility(manifest.api_version)
if error:
errors.append(error)
return errors
def load_plugin_from_dir(plugin_dir: Path) -> PluginLoadResult:
"""Load and validate a plugin from a directory.
The plugin directory must contain an __init__.py that exports:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Callable
Args:
plugin_dir: Path to the plugin directory.
Returns:
PluginLoadResult with success status and either plugin data or error info.
"""
plugin_id = plugin_dir.name
errors: list[str] = []
# Check if directory exists
if not plugin_dir.exists():
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=(f"Plugin directory does not exist: {plugin_dir}",),
),
)
# Check for __init__.py
init_file = plugin_dir / "__init__.py"
if not init_file.exists():
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=("Missing __init__.py in plugin directory",),
),
)
# Import the module
module_name = f"abogen.tts_plugin._loaded.{plugin_id}"
try:
# Remove from cache if already imported (for testing)
if module_name in sys.modules:
del sys.modules[module_name]
spec = importlib.util.spec_from_file_location(
module_name, init_file, submodule_search_locations=[]
)
if spec is None or spec.loader is None:
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=(f"Failed to create module spec for {init_file}",),
),
)
module = importlib.util.module_from_spec(spec)
sys.modules[module_name] = module
spec.loader.exec_module(module)
except Exception as e:
# Clean up module from sys.modules on import failure
if module_name in sys.modules:
del sys.modules[module_name]
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=(f"Failed to import plugin module: {e}",),
),
)
# Validate manifest
manifest_errors = _validate_manifest(module, plugin_dir)
errors.extend(manifest_errors)
# If manifest is valid, perform additional validation
manifest = getattr(module, "PLUGIN_MANIFEST", None)
if isinstance(manifest, PluginManifest):
# Validate api_version
api_errors = _validate_api_version(manifest)
errors.extend(api_errors)
# Validate capabilities
cap_errors = _validate_capabilities(manifest)
errors.extend(cap_errors)
# Use manifest id if available
plugin_id = manifest.id
# Check if any errors occurred
if errors:
# Clean up module from sys.modules
if module_name in sys.modules:
del sys.modules[module_name]
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=tuple(errors),
),
)
# Get MODEL_REQUIREMENTS
model_requirements = tuple(getattr(module, "MODEL_REQUIREMENTS", []))
create_engine = getattr(module, "create_engine", None)
return PluginLoadResult(
success=True,
manifest=manifest,
model_requirements=model_requirements,
create_engine=create_engine,
module=module,
)
def discover_plugins(plugin_dirs: list[Path]) -> list[PluginLoadResult]:
"""Discover and load plugins from multiple directories.
Args:
plugin_dirs: List of directories to scan for plugins.
Returns:
List of PluginLoadResult, one per plugin directory found.
"""
results: list[PluginLoadResult] = []
for plugin_dir in plugin_dirs:
if not plugin_dir.exists():
continue
# Scan for subdirectories (each is a potential plugin)
for item in sorted(plugin_dir.iterdir()):
if item.is_dir() and not item.name.startswith("."):
result = load_plugin_from_dir(item)
results.append(result)
return results
def load_plugin(
plugin_dir: Path,
) -> PluginLoadResult:
"""Load a single plugin from a directory.
This is the main entry point for loading a plugin.
Args:
plugin_dir: Path to the plugin directory.
Returns:
PluginLoadResult with success status and either plugin data or error info.
"""
return load_plugin_from_dir(plugin_dir)
"""Plugin loader infrastructure for the TTS Plugin Architecture.
This module provides functionality to discover, import, validate, and load
TTS plugins. It handles both valid and invalid plugins, providing diagnostic
messages for errors.
The loader does NOT:
- Create Engine instances (that's the plugin's create_engine() responsibility)
- Manage plugin lifecycle (that's the Plugin Manager's responsibility)
- Implement any TTS engine functionality
"""
from __future__ import annotations
import importlib
import re
import sys
import types
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any, Callable
from abogen.tts_plugin.manifest import ModelManifest, PluginManifest
# Host API version for compatibility checking
HOST_API_VERSION = "1.0"
@dataclass(frozen=True)
class PluginLoadError:
"""Diagnostic information for a failed plugin load.
Attributes:
plugin_id: Plugin identifier if available, otherwise directory name.
path: Path to the plugin directory.
errors: List of error messages describing what went wrong.
"""
plugin_id: str
path: Path
errors: tuple[str, ...] = field(default_factory=tuple)
@dataclass(frozen=True)
class PluginLoadResult:
"""Result of loading a plugin.
Attributes:
success: Whether the plugin loaded successfully.
manifest: The plugin manifest if successful.
model_requirements: Model requirements if successful.
create_engine: The create_engine function if successful.
module: The plugin module if successful.
error: Error information if failed.
"""
success: bool
manifest: PluginManifest | None = None
model_requirements: tuple[ModelManifest, ...] | None = None
create_engine: Callable[..., Any] | None = None
module: types.ModuleType | None = None
error: PluginLoadError | None = None
def _parse_api_version(version: str) -> tuple[int, int] | None:
"""Parse an api_version string into (major, minor) tuple.
Args:
version: Version string in format "MAJOR.MINOR".
Returns:
Tuple of (major, minor) or None if invalid format.
"""
match = re.match(r"^(\d+)\.(\d+)$", version)
if match:
return int(match.group(1)), int(match.group(2))
return None
def _check_api_version_compatibility(plugin_version: str) -> str | None:
"""Check if plugin api_version is compatible with host.
Architecture spec:
- Format: semver (MAJOR.MINOR)
- Compatibility: Host rejects plugin if major version differs
- Minor version: backward compatible, Host accepts higher minor
Args:
plugin_version: Plugin's api_version string.
Returns:
Error message if incompatible, None if compatible.
"""
plugin_ver = _parse_api_version(plugin_version)
if plugin_ver is None:
return f"Invalid api_version format: '{plugin_version}'. Expected format: MAJOR.MINOR"
host_ver = _parse_api_version(HOST_API_VERSION)
if host_ver is None:
return f"Invalid host api_version format: '{HOST_API_VERSION}'"
if plugin_ver[0] != host_ver[0]:
return (
f"api_version major mismatch: plugin={plugin_ver[0]}, host={host_ver[0]}. "
f"Major version must match for compatibility."
)
return None
def _validate_manifest(module: types.ModuleType, plugin_dir: Path) -> list[str]:
"""Validate that a plugin module has required exports.
Args:
module: The imported plugin module.
plugin_dir: Path to the plugin directory.
Returns:
List of error messages (empty if valid).
"""
errors: list[str] = []
# Check PLUGIN_MANIFEST
manifest = getattr(module, "PLUGIN_MANIFEST", None)
if manifest is None:
errors.append("Missing PLUGIN_MANIFEST export")
elif not isinstance(manifest, PluginManifest):
errors.append(
f"PLUGIN_MANIFEST must be a PluginManifest instance, "
f"got {type(manifest).__name__}"
)
# Check MODEL_REQUIREMENTS
model_reqs = getattr(module, "MODEL_REQUIREMENTS", None)
if model_reqs is None:
errors.append("Missing MODEL_REQUIREMENTS export")
elif not isinstance(model_reqs, list):
errors.append(
f"MODEL_REQUIREMENTS must be a list, got {type(model_reqs).__name__}"
)
else:
for i, req in enumerate(model_reqs):
if not isinstance(req, ModelManifest):
errors.append(
f"MODEL_REQUIREMENTS[{i}] must be a ModelManifest instance, "
f"got {type(req).__name__}"
)
# Check create_engine
create_engine = getattr(module, "create_engine", None)
if create_engine is None:
errors.append("Missing create_engine export")
elif not callable(create_engine):
errors.append(
f"create_engine must be callable, got {type(create_engine).__name__}"
)
return errors
def _validate_capabilities(manifest: PluginManifest) -> list[str]:
"""Validate plugin capabilities.
Args:
manifest: The plugin manifest to validate.
Returns:
List of error messages (empty if valid).
"""
errors: list[str] = []
# Known capabilities (can be extended)
known_capabilities = frozenset({
"voice_list",
"preview",
"voice_clone",
"voice_blend",
"streaming",
"cancel",
})
for cap in manifest.capabilities:
if cap not in known_capabilities:
errors.append(f"Unknown capability: '{cap}'")
return errors
def _validate_api_version(manifest: PluginManifest) -> list[str]:
"""Validate api_version compatibility.
Args:
manifest: The plugin manifest to validate.
Returns:
List of error messages (empty if valid).
"""
errors: list[str] = []
error = _check_api_version_compatibility(manifest.api_version)
if error:
errors.append(error)
return errors
def load_plugin_from_dir(plugin_dir: Path) -> PluginLoadResult:
"""Load and validate a plugin from a directory.
The plugin directory must contain an __init__.py that exports:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Callable
Args:
plugin_dir: Path to the plugin directory.
Returns:
PluginLoadResult with success status and either plugin data or error info.
"""
plugin_id = plugin_dir.name
errors: list[str] = []
# Check if directory exists
if not plugin_dir.exists():
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=(f"Plugin directory does not exist: {plugin_dir}",),
),
)
# Check for __init__.py
init_file = plugin_dir / "__init__.py"
if not init_file.exists():
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=("Missing __init__.py in plugin directory",),
),
)
# Import the module
module_name = f"abogen.tts_plugin._loaded.{plugin_id}"
try:
# Remove from cache if already imported (for testing)
if module_name in sys.modules:
del sys.modules[module_name]
spec = importlib.util.spec_from_file_location(
module_name, init_file, submodule_search_locations=[]
)
if spec is None or spec.loader is None:
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=(f"Failed to create module spec for {init_file}",),
),
)
module = importlib.util.module_from_spec(spec)
sys.modules[module_name] = module
spec.loader.exec_module(module)
except Exception as e:
# Clean up module from sys.modules on import failure
if module_name in sys.modules:
del sys.modules[module_name]
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=(f"Failed to import plugin module: {e}",),
),
)
# Validate manifest
manifest_errors = _validate_manifest(module, plugin_dir)
errors.extend(manifest_errors)
# If manifest is valid, perform additional validation
manifest = getattr(module, "PLUGIN_MANIFEST", None)
if isinstance(manifest, PluginManifest):
# Validate api_version
api_errors = _validate_api_version(manifest)
errors.extend(api_errors)
# Validate capabilities
cap_errors = _validate_capabilities(manifest)
errors.extend(cap_errors)
# Use manifest id if available
plugin_id = manifest.id
# Check if any errors occurred
if errors:
# Clean up module from sys.modules
if module_name in sys.modules:
del sys.modules[module_name]
return PluginLoadResult(
success=False,
error=PluginLoadError(
plugin_id=plugin_id,
path=plugin_dir,
errors=tuple(errors),
),
)
# Get MODEL_REQUIREMENTS
model_requirements = tuple(getattr(module, "MODEL_REQUIREMENTS", []))
create_engine = getattr(module, "create_engine", None)
return PluginLoadResult(
success=True,
manifest=manifest,
model_requirements=model_requirements,
create_engine=create_engine,
module=module,
)
def discover_plugins(plugin_dirs: list[Path]) -> list[PluginLoadResult]:
"""Discover and load plugins from multiple directories.
Args:
plugin_dirs: List of directories to scan for plugins.
Returns:
List of PluginLoadResult, one per plugin directory found.
"""
results: list[PluginLoadResult] = []
for plugin_dir in plugin_dirs:
if not plugin_dir.exists():
continue
# Scan for subdirectories (each is a potential plugin)
for item in sorted(plugin_dir.iterdir()):
if item.is_dir() and not item.name.startswith("."):
result = load_plugin_from_dir(item)
results.append(result)
return results
def load_plugin(
plugin_dir: Path,
) -> PluginLoadResult:
"""Load a single plugin from a directory.
This is the main entry point for loading a plugin.
Args:
plugin_dir: Path to the plugin directory.
Returns:
PluginLoadResult with success status and either plugin data or error info.
"""
return load_plugin_from_dir(plugin_dir)
abogen/tts_plugin/manifest.py
+189 -189
View File
@@ -1,189 +1,189 @@
"""Plugin manifest types for the TTS Plugin Architecture.
This module contains static metadata types that describe plugins.
These types have no dependencies and are immutable.
"""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Any
@dataclass(frozen=True)
class AudioFormatManifest:
"""Manifest describing an audio format.
Attributes:
mime: MIME type of the audio.
extension: File extension.
"""
mime: str
extension: str
@dataclass(frozen=True)
class EnumOption:
"""Manifest describing an enum option for a parameter.
Attributes:
value: The enum value.
label: Human-readable label.
"""
value: str
label: str
@dataclass(frozen=True)
class ParameterManifest:
"""Manifest describing a synthesis parameter.
Attributes:
id: Parameter identifier.
name: Human-readable name.
description: Parameter description.
type: Parameter type ("float", "int", "string", "boolean", "enum").
default: Default value.
min: Minimum value (optional, for numeric types).
max: Maximum value (optional, for numeric types).
step: Step size (optional, for numeric types).
options: Available options (optional, for enum type).
unit: Unit of measurement (optional).
group: Parameter group (optional).
"""
id: str
name: str
description: str
type: str
default: Any
min: float | None = None
max: float | None = None
step: float | None = None
options: tuple[EnumOption, ...] = field(default_factory=tuple)
unit: str | None = None
group: str | None = None
@dataclass(frozen=True)
class VoiceManifest:
"""Manifest describing a voice.
Attributes:
id: Voice identifier.
name: Human-readable name.
tags: Voice tags (e.g., language, style).
"""
id: str
name: str
tags: tuple[str, ...] = field(default_factory=tuple)
@dataclass(frozen=True)
class VoiceSourceManifest:
"""Manifest describing a voice source.
Attributes:
id: Voice source identifier.
name: Human-readable name.
type: Source type ("list", "speaker_id", "clone", "blend", "generate", "none").
config: Source-specific configuration.
"""
id: str
name: str
type: str
config: Any = None
@dataclass(frozen=True)
class EngineManifest:
"""Manifest describing engine capabilities.
Attributes:
voiceSources: Available voice sources.
parameters: Available synthesis parameters.
audioFormats: Supported audio formats.
"""
voiceSources: tuple[VoiceSourceManifest, ...] = field(default_factory=tuple)
parameters: tuple[ParameterManifest, ...] = field(default_factory=tuple)
audioFormats: tuple[AudioFormatManifest, ...] = field(default_factory=tuple)
@dataclass(frozen=True)
class GpuRequirement:
"""Manifest describing GPU requirements.
Attributes:
required: Whether GPU is required.
type: GPU type (e.g., "cuda", "rocm").
memory: Required GPU memory in GB.
"""
required: bool = False
type: str | None = None
memory: float | None = None
@dataclass(frozen=True)
class RequirementManifest:
"""Manifest describing plugin requirements.
Attributes:
gpu: GPU requirements (optional).
memory: Required RAM in GB (optional).
internet: Whether internet is required (optional).
"""
gpu: GpuRequirement | None = None
memory: float | None = None
internet: bool | None = None
@dataclass(frozen=True)
class ModelManifest:
"""Manifest describing a model requirement.
Attributes:
id: Model identifier.
name: Human-readable name.
size: Model size as string (e.g., "100MB", "2GB").
"""
id: str
name: str
size: str
@dataclass(frozen=True)
class PluginManifest:
"""Main manifest for a TTS plugin.
Attributes:
id: Plugin identifier (unique).
name: Human-readable name.
version: Plugin version.
api_version: API version (semver format: MAJOR.MINOR).
description: Plugin description.
author: Plugin author.
capabilities: List of capability identifiers.
requires: Plugin requirements.
engine: Engine manifest.
voices: Optional static voice catalog. None = not declared (use VoiceLister),
empty tuple = explicitly no static voices, non-empty = static catalog.
"""
id: str
name: str
version: str
api_version: str
description: str
author: str
capabilities: tuple[str, ...] = field(default_factory=tuple)
requires: RequirementManifest = field(default_factory=RequirementManifest)
engine: EngineManifest = field(default_factory=EngineManifest)
voices: tuple[VoiceManifest, ...] | None = None
"""Plugin manifest types for the TTS Plugin Architecture.
This module contains static metadata types that describe plugins.
These types have no dependencies and are immutable.
"""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Any
@dataclass(frozen=True)
class AudioFormatManifest:
"""Manifest describing an audio format.
Attributes:
mime: MIME type of the audio.
extension: File extension.
"""
mime: str
extension: str
@dataclass(frozen=True)
class EnumOption:
"""Manifest describing an enum option for a parameter.
Attributes:
value: The enum value.
label: Human-readable label.
"""
value: str
label: str
@dataclass(frozen=True)
class ParameterManifest:
"""Manifest describing a synthesis parameter.
Attributes:
id: Parameter identifier.
name: Human-readable name.
description: Parameter description.
type: Parameter type ("float", "int", "string", "boolean", "enum").
default: Default value.
min: Minimum value (optional, for numeric types).
max: Maximum value (optional, for numeric types).
step: Step size (optional, for numeric types).
options: Available options (optional, for enum type).
unit: Unit of measurement (optional).
group: Parameter group (optional).
"""
id: str
name: str
description: str
type: str
default: Any
min: float | None = None
max: float | None = None
step: float | None = None
options: tuple[EnumOption, ...] = field(default_factory=tuple)
unit: str | None = None
group: str | None = None
@dataclass(frozen=True)
class VoiceManifest:
"""Manifest describing a voice.
Attributes:
id: Voice identifier.
name: Human-readable name.
tags: Voice tags (e.g., language, style).
"""
id: str
name: str
tags: tuple[str, ...] = field(default_factory=tuple)
@dataclass(frozen=True)
class VoiceSourceManifest:
"""Manifest describing a voice source.
Attributes:
id: Voice source identifier.
name: Human-readable name.
type: Source type ("list", "speaker_id", "clone", "blend", "generate", "none").
config: Source-specific configuration.
"""
id: str
name: str
type: str
config: Any = None
@dataclass(frozen=True)
class EngineManifest:
"""Manifest describing engine capabilities.
Attributes:
voiceSources: Available voice sources.
parameters: Available synthesis parameters.
audioFormats: Supported audio formats.
"""
voiceSources: tuple[VoiceSourceManifest, ...] = field(default_factory=tuple)
parameters: tuple[ParameterManifest, ...] = field(default_factory=tuple)
audioFormats: tuple[AudioFormatManifest, ...] = field(default_factory=tuple)
@dataclass(frozen=True)
class GpuRequirement:
"""Manifest describing GPU requirements.
Attributes:
required: Whether GPU is required.
type: GPU type (e.g., "cuda", "rocm").
memory: Required GPU memory in GB.
"""
required: bool = False
type: str | None = None
memory: float | None = None
@dataclass(frozen=True)
class RequirementManifest:
"""Manifest describing plugin requirements.
Attributes:
gpu: GPU requirements (optional).
memory: Required RAM in GB (optional).
internet: Whether internet is required (optional).
"""
gpu: GpuRequirement | None = None
memory: float | None = None
internet: bool | None = None
@dataclass(frozen=True)
class ModelManifest:
"""Manifest describing a model requirement.
Attributes:
id: Model identifier.
name: Human-readable name.
size: Model size as string (e.g., "100MB", "2GB").
"""
id: str
name: str
size: str
@dataclass(frozen=True)
class PluginManifest:
"""Main manifest for a TTS plugin.
Attributes:
id: Plugin identifier (unique).
name: Human-readable name.
version: Plugin version.
api_version: API version (semver format: MAJOR.MINOR).
description: Plugin description.
author: Plugin author.
capabilities: List of capability identifiers.
requires: Plugin requirements.
engine: Engine manifest.
voices: Optional static voice catalog. None = not declared (use VoiceLister),
empty tuple = explicitly no static voices, non-empty = static catalog.
"""
id: str
name: str
version: str
api_version: str
description: str
author: str
capabilities: tuple[str, ...] = field(default_factory=tuple)
requires: RequirementManifest = field(default_factory=RequirementManifest)
engine: EngineManifest = field(default_factory=EngineManifest)
voices: tuple[VoiceManifest, ...] | None = None
abogen/tts_plugin/plugin.py
+55 -55
View File
@@ -1,55 +1,55 @@
"""Plugin contract for the TTS Plugin Architecture.
This module defines the plugin contract that all TTS plugins must implement.
Each plugin must export:
- PLUGIN_MANIFEST: PluginManifest instance
- MODEL_REQUIREMENTS: list of ModelManifest instances
- create_engine(): Factory function that creates an Engine
The create_engine() function is the entry point for plugin activation.
It must be atomic: succeed fully or raise and clean up.
"""
from __future__ import annotations
from pathlib import Path
from typing import Protocol, runtime_checkable
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
@runtime_checkable
class Plugin(Protocol):
"""Protocol defining the plugin contract.
Every TTS plugin must implement this protocol by exporting:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Callable[[HostContext, Path | None, EngineConfig], Engine]
"""
def create_engine(
self,
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create an engine instance.
This is the factory function that creates an Engine from a plugin.
It must be atomic: succeed fully or raise EngineError and clean up.
Args:
context: Host services (config dir, logger, http client).
model_path: Resolved model path, or None for cloud/no-model engines.
config: Engine initialization settings.
Returns:
A fully initialized Engine instance.
Raises:
EngineError: On failure. Cleans up partially created resources.
"""
...
"""Plugin contract for the TTS Plugin Architecture.
This module defines the plugin contract that all TTS plugins must implement.
Each plugin must export:
- PLUGIN_MANIFEST: PluginManifest instance
- MODEL_REQUIREMENTS: list of ModelManifest instances
- create_engine(): Factory function that creates an Engine
The create_engine() function is the entry point for plugin activation.
It must be atomic: succeed fully or raise and clean up.
"""
from __future__ import annotations
from pathlib import Path
from typing import Protocol, runtime_checkable
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
@runtime_checkable
class Plugin(Protocol):
"""Protocol defining the plugin contract.
Every TTS plugin must implement this protocol by exporting:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Callable[[HostContext, Path | None, EngineConfig], Engine]
"""
def create_engine(
self,
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create an engine instance.
This is the factory function that creates an Engine from a plugin.
It must be atomic: succeed fully or raise EngineError and clean up.
Args:
context: Host services (config dir, logger, http client).
model_path: Resolved model path, or None for cloud/no-model engines.
config: Engine initialization settings.
Returns:
A fully initialized Engine instance.
Raises:
EngineError: On failure. Cleans up partially created resources.
"""
...
abogen/tts_plugin/plugin_manager.py
+153 -153
View File
@@ -1,153 +1,153 @@
"""Plugin Manager
Provides a simple interface for consumers to access TTS engines via the
new Plugin Architecture. Discovers, loads, and manages plugins from the
plugins directory.
Usage:
from abogen.tts_plugin.plugin_manager import get_plugin_manager
manager = get_plugin_manager()
engine = manager.create_engine("kokoro", lang_code="a", device="cpu")
session = engine.create_session()
try:
result = session.synthesize("Hello world")
finally:
session.dispose()
"""
from typing import Any, Dict, List, Optional, Type
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import AudioFormat
class PluginManager:
"""Manages TTS plugins and provides a simple interface for consumers."""
def __init__(self) -> None:
self._plugins: Dict[str, dict] = {}
self._engines: Dict[str, Engine] = {}
self._loaded = False
def discover(self, plugins_dir: str = "plugins") -> None:
"""Discover and load all plugins from the given directory."""
import os
from pathlib import Path
from abogen.tts_plugin.loader import load_plugin_from_dir
self._plugins.clear()
self._engines.clear()
plugins_path = Path(plugins_dir)
if not plugins_path.exists():
self._loaded = True
return
for entry in plugins_path.iterdir():
if entry.is_dir() and (entry / "__init__.py").exists():
try:
result = load_plugin_from_dir(entry)
if result.success and result.manifest is not None:
self._plugins[result.manifest.id] = {
"manifest": result.manifest,
"create_engine": result.create_engine,
"module": result.module,
}
except Exception as e:
# Log error but continue with other plugins
print(f"Warning: Failed to load plugin from {entry}: {e}")
self._loaded = True
def _ensure_loaded(self) -> None:
"""Ensure plugins have been discovered."""
if not self._loaded:
self.discover()
def list_plugins(self) -> List[PluginManifest]:
"""Return manifests for all loaded plugins."""
self._ensure_loaded()
return [info["manifest"] for info in self._plugins.values()]
def get_plugin(self, plugin_id: str) -> Optional[dict]:
"""Get plugin info by ID."""
self._ensure_loaded()
return self._plugins.get(plugin_id)
def has_plugin(self, plugin_id: str) -> bool:
"""Check if a plugin is loaded."""
self._ensure_loaded()
return plugin_id in self._plugins
def create_engine(self, plugin_id: str, **kwargs: Any) -> Engine:
"""Create an engine instance for the given plugin.
Args:
plugin_id: The plugin identifier (e.g., "kokoro")
**kwargs: Arguments passed to the engine constructor
Returns:
An Engine instance
Raises:
KeyError: If plugin_id is not found
Exception: If engine creation fails
"""
self._ensure_loaded()
if plugin_id not in self._plugins:
raise KeyError(f"Plugin not found: {plugin_id}")
plugin_info = self._plugins[plugin_id]
create_engine_func = plugin_info["create_engine"]
# Create engine using the plugin's factory
engine = create_engine_func(**kwargs)
return engine
def get_or_create_engine(self, plugin_id: str, **kwargs: Any) -> Engine:
"""Get an existing engine or create a new one.
Engines are cached by plugin_id. If you need multiple instances
with different parameters, use create_engine() directly.
"""
self._ensure_loaded()
cache_key = plugin_id
if cache_key in self._engines:
return self._engines[cache_key]
engine = self.create_engine(plugin_id, **kwargs)
self._engines[cache_key] = engine
return engine
def dispose_all(self) -> None:
"""Dispose all cached engines."""
for engine in self._engines.values():
try:
engine.dispose()
except Exception:
pass # dispose() should never raise
self._engines.clear()
# Global singleton
_manager: Optional[PluginManager] = None
def get_plugin_manager() -> PluginManager:
"""Get the global PluginManager instance."""
global _manager
if _manager is None:
_manager = PluginManager()
return _manager
def reset_plugin_manager() -> None:
"""Reset the global PluginManager (for testing)."""
global _manager
if _manager is not None:
_manager.dispose_all()
_manager = None
"""Plugin Manager
Provides a simple interface for consumers to access TTS engines via the
new Plugin Architecture. Discovers, loads, and manages plugins from the
plugins directory.
Usage:
from abogen.tts_plugin.plugin_manager import get_plugin_manager
manager = get_plugin_manager()
engine = manager.create_engine("kokoro", lang_code="a", device="cpu")
session = engine.create_session()
try:
result = session.synthesize("Hello world")
finally:
session.dispose()
"""
from typing import Any, Dict, List, Optional, Type
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import AudioFormat
class PluginManager:
"""Manages TTS plugins and provides a simple interface for consumers."""
def __init__(self) -> None:
self._plugins: Dict[str, dict] = {}
self._engines: Dict[str, Engine] = {}
self._loaded = False
def discover(self, plugins_dir: str = "plugins") -> None:
"""Discover and load all plugins from the given directory."""
import os
from pathlib import Path
from abogen.tts_plugin.loader import load_plugin_from_dir
self._plugins.clear()
self._engines.clear()
plugins_path = Path(plugins_dir)
if not plugins_path.exists():
self._loaded = True
return
for entry in plugins_path.iterdir():
if entry.is_dir() and (entry / "__init__.py").exists():
try:
result = load_plugin_from_dir(entry)
if result.success and result.manifest is not None:
self._plugins[result.manifest.id] = {
"manifest": result.manifest,
"create_engine": result.create_engine,
"module": result.module,
}
except Exception as e:
# Log error but continue with other plugins
print(f"Warning: Failed to load plugin from {entry}: {e}")
self._loaded = True
def _ensure_loaded(self) -> None:
"""Ensure plugins have been discovered."""
if not self._loaded:
self.discover()
def list_plugins(self) -> List[PluginManifest]:
"""Return manifests for all loaded plugins."""
self._ensure_loaded()
return [info["manifest"] for info in self._plugins.values()]
def get_plugin(self, plugin_id: str) -> Optional[dict]:
"""Get plugin info by ID."""
self._ensure_loaded()
return self._plugins.get(plugin_id)
def has_plugin(self, plugin_id: str) -> bool:
"""Check if a plugin is loaded."""
self._ensure_loaded()
return plugin_id in self._plugins
def create_engine(self, plugin_id: str, **kwargs: Any) -> Engine:
"""Create an engine instance for the given plugin.
Args:
plugin_id: The plugin identifier (e.g., "kokoro")
**kwargs: Arguments passed to the engine constructor
Returns:
An Engine instance
Raises:
KeyError: If plugin_id is not found
Exception: If engine creation fails
"""
self._ensure_loaded()
if plugin_id not in self._plugins:
raise KeyError(f"Plugin not found: {plugin_id}")
plugin_info = self._plugins[plugin_id]
create_engine_func = plugin_info["create_engine"]
# Create engine using the plugin's factory
engine = create_engine_func(**kwargs)
return engine
def get_or_create_engine(self, plugin_id: str, **kwargs: Any) -> Engine:
"""Get an existing engine or create a new one.
Engines are cached by plugin_id. If you need multiple instances
with different parameters, use create_engine() directly.
"""
self._ensure_loaded()
cache_key = plugin_id
if cache_key in self._engines:
return self._engines[cache_key]
engine = self.create_engine(plugin_id, **kwargs)
self._engines[cache_key] = engine
return engine
def dispose_all(self) -> None:
"""Dispose all cached engines."""
for engine in self._engines.values():
try:
engine.dispose()
except Exception:
pass # dispose() should never raise
self._engines.clear()
# Global singleton
_manager: Optional[PluginManager] = None
def get_plugin_manager() -> PluginManager:
"""Get the global PluginManager instance."""
global _manager
if _manager is None:
_manager = PluginManager()
return _manager
def reset_plugin_manager() -> None:
"""Reset the global PluginManager (for testing)."""
global _manager
if _manager is not None:
_manager.dispose_all()
_manager = None
abogen/tts_plugin/types.py
+111 -111
View File
@@ -1,111 +1,111 @@
"""Core domain types for the TTS Plugin Architecture.
This module contains immutable value objects that form the core domain.
These types have zero dependencies and are used across the plugin system.
"""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Any, Mapping
@dataclass(frozen=True)
class AudioFormat:
"""Immutable value object representing an audio format.
Attributes:
mime: MIME type of the audio (e.g., "audio/wav", "audio/mpeg").
extension: File extension (e.g., "wav", "mp3").
"""
mime: str
extension: str
@dataclass(frozen=True)
class Duration:
"""Immutable value object representing a time duration.
Attributes:
seconds: Duration in seconds.
"""
seconds: float
@dataclass(frozen=True)
class VoiceSelection:
"""Immutable value object for voice selection. Opaque to engine.
Attributes:
source: Voice source identifier (e.g., "builtin", "clone").
key: Voice key within the source.
payload: Optional payload for clone/blend sources.
"""
source: str
key: str
payload: Any = None
@dataclass(frozen=True)
class ParameterValues:
"""Immutable value object for synthesis parameters. Behaves like Mapping[str, Any].
Attributes:
values: Mapping of parameter names to their values.
"""
values: Mapping[str, Any] = field(default_factory=dict)
@dataclass(frozen=True)
class SynthesisRequest:
"""Immutable value object for a synthesis request.
Attributes:
text: Text to synthesize.
voice: Voice selection.
parameters: Synthesis parameters.
format: Desired audio output format.
"""
text: str
voice: VoiceSelection
parameters: ParameterValues
format: AudioFormat
@dataclass(frozen=True)
class SynthesizedAudio:
"""Immutable value object for synthesized audio result.
Attributes:
data: Raw audio bytes.
format: Audio format of the result.
duration: Duration of the audio.
"""
data: bytes
format: AudioFormat
duration: Duration
@dataclass(frozen=True)
class EngineConfig:
"""Immutable configuration of an Engine instance.
Contains parameters that define how a particular Engine instance is
created and that remain constant throughout the lifetime of that Engine.
Plugin implementations may ignore fields that are not applicable to them.
Attributes:
device: Device to use (e.g., "cpu", "cuda:0").
lang_code: Language code for the engine (e.g., "a" for Kokoro English).
Plugins that do not require a language code ignore this field.
"""
device: str = "cpu"
lang_code: str = "a"
"""Core domain types for the TTS Plugin Architecture.
This module contains immutable value objects that form the core domain.
These types have zero dependencies and are used across the plugin system.
"""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Any, Mapping
@dataclass(frozen=True)
class AudioFormat:
"""Immutable value object representing an audio format.
Attributes:
mime: MIME type of the audio (e.g., "audio/wav", "audio/mpeg").
extension: File extension (e.g., "wav", "mp3").
"""
mime: str
extension: str
@dataclass(frozen=True)
class Duration:
"""Immutable value object representing a time duration.
Attributes:
seconds: Duration in seconds.
"""
seconds: float
@dataclass(frozen=True)
class VoiceSelection:
"""Immutable value object for voice selection. Opaque to engine.
Attributes:
source: Voice source identifier (e.g., "builtin", "clone").
key: Voice key within the source.
payload: Optional payload for clone/blend sources.
"""
source: str
key: str
payload: Any = None
@dataclass(frozen=True)
class ParameterValues:
"""Immutable value object for synthesis parameters. Behaves like Mapping[str, Any].
Attributes:
values: Mapping of parameter names to their values.
"""
values: Mapping[str, Any] = field(default_factory=dict)
@dataclass(frozen=True)
class SynthesisRequest:
"""Immutable value object for a synthesis request.
Attributes:
text: Text to synthesize.
voice: Voice selection.
parameters: Synthesis parameters.
format: Desired audio output format.
"""
text: str
voice: VoiceSelection
parameters: ParameterValues
format: AudioFormat
@dataclass(frozen=True)
class SynthesizedAudio:
"""Immutable value object for synthesized audio result.
Attributes:
data: Raw audio bytes.
format: Audio format of the result.
duration: Duration of the audio.
"""
data: bytes
format: AudioFormat
duration: Duration
@dataclass(frozen=True)
class EngineConfig:
"""Immutable configuration of an Engine instance.
Contains parameters that define how a particular Engine instance is
created and that remain constant throughout the lifetime of that Engine.
Plugin implementations may ignore fields that are not applicable to them.
Attributes:
device: Device to use (e.g., "cpu", "cuda:0").
lang_code: Language code for the engine (e.g., "a" for Kokoro English).
Plugins that do not require a language code ignore this field.
"""
device: str = "cpu"
lang_code: str = "a"
abogen/tts_plugin/utils.py
+235 -235
View File
@@ -1,235 +1,235 @@
"""TTS Plugin Architecture — direct utility functions.
Provides helpers that replace the former compatibility adapter by
calling the Plugin Manager directly.
"""
from __future__ import annotations
from typing import Any, Iterator
import numpy as np
from abogen.tts_plugin.plugin_manager import get_plugin_manager
def get_voices(plugin_id: str) -> tuple[str, ...]:
"""Return the voice-id tuple for *plugin_id*.
Uses the official Plugin Architecture: PluginManager → Engine → VoiceLister.
First checks plugin manifest for static voice catalog.
"""
import logging
import tempfile
from pathlib import Path
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
manager = get_plugin_manager()
if not manager.has_plugin(plugin_id):
return ()
# Check manifest for static voice catalog
plugin_info = manager.get_plugin(plugin_id)
if plugin_info is not None:
manifest = plugin_info.get("manifest")
if manifest is not None and manifest.voices is not None:
return tuple(v.id for v in manifest.voices)
ctx = HostContext(
config_dir=Path(tempfile.gettempdir()),
logger=logging.getLogger(f"abogen.utils.{plugin_id}"),
http_client=type("_StubHttpClient", (), {
"get": staticmethod(lambda url, **kw: None),
"post": staticmethod(lambda url, **kw: None),
})(),
)
try:
engine = manager.create_engine(
plugin_id,
context=ctx,
model_path=None,
config=EngineConfig(device="cpu"),
)
except Exception:
return ()
try:
from abogen.tts_plugin.capabilities import VoiceLister
if isinstance(engine, VoiceLister):
manifests = engine.listVoices("builtin")
return tuple(v.id for v in manifests)
return ()
except Exception:
return ()
finally:
engine.dispose()
def get_default_voice(plugin_id: str, fallback: str = "") -> str:
"""Return the first voice of *plugin_id*, or *fallback*."""
voices = get_voices(plugin_id)
return voices[0] if voices else fallback
def is_plugin_registered(plugin_id: str) -> bool:
"""Check whether *plugin_id* is loaded by the Plugin Manager."""
return get_plugin_manager().has_plugin(plugin_id)
def resolve_voice_to_plugin(spec: str, fallback: str = "kokoro") -> str:
"""Determine which plugin owns the given voice specification.
Resolution rules:
1. Empty spec -> fallback
2. Kokoro formula (contains '*' or '+') -> "kokoro"
3. Exact voice-id match against loaded plugins -> plugin id
4. Unknown voice -> fallback
"""
raw = str(spec or "").strip()
if not raw:
return fallback
if "*" in raw or "+" in raw:
return "kokoro"
upper = raw.upper()
manager = get_plugin_manager()
for manifest in manager.list_plugins():
for voice_source in manifest.engine.voiceSources:
if voice_source.type == "list" and isinstance(voice_source.config, dict):
try:
engine = manager.create_engine(manifest.id)
try:
if hasattr(engine, "listVoices"):
voice_manifests = engine.listVoices(voice_source.id)
voice_ids = [v.id.upper() for v in voice_manifests]
if upper in voice_ids:
return manifest.id
finally:
engine.dispose()
except Exception:
continue
return fallback
class Pipeline:
"""Callable wrapper around Engine / EngineSession.
Presents the same interface that old callers expect::
pipeline = create_pipeline("kokoro", lang_code="a", device="cpu")
for segment in pipeline(text, voice="af_nova", speed=1.0):
audio = segment.audio
"""
def __init__(self, engine: Any, **engine_kwargs: Any) -> None:
self._engine = engine
self._engine_kwargs = engine_kwargs
self._session: Any = None
def _ensure_session(self) -> Any:
if self._session is None:
self._session = self._engine.createSession()
return self._session
def __call__(
self,
text: str,
voice: str = "default",
speed: float = 1.0,
split_pattern: str | None = None,
**kwargs: Any,
) -> Iterator[Any]:
from abogen.tts_plugin.types import (
AudioFormat,
ParameterValues,
SynthesisRequest,
VoiceSelection,
)
session = self._ensure_session()
params: dict[str, Any] = {"speed": speed}
if split_pattern is not None:
params["split_pattern"] = split_pattern
params.update(kwargs)
request = SynthesisRequest(
text=text,
voice=VoiceSelection(source="builtin", key=voice),
parameters=ParameterValues(values=params),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
audio_array = np.frombuffer(result.data, dtype=np.float32)
from dataclasses import dataclass
@dataclass
class Segment:
graphemes: str
audio: np.ndarray
yield Segment(graphemes=text, audio=audio_array)
def dispose(self) -> None:
if self._session is not None:
try:
self._session.dispose()
except Exception:
pass
self._session = None
def __del__(self) -> None:
self.dispose()
def create_pipeline(
plugin_id: str,
*,
lang_code: str = "a",
device: str = "cpu",
) -> Pipeline:
"""Create a callable TTS pipeline via the Plugin Architecture.
Builds a proper HostContext and EngineConfig, then delegates to the
PluginManager to create the engine. Returns a :class:`Pipeline` whose
``__call__`` interface matches the callable protocol used by consumers.
Args:
plugin_id: Plugin identifier (e.g., "kokoro", "supertonic").
lang_code: Language code for the engine.
device: Device to use (e.g., "cpu", "cuda:0").
Returns:
A callable Pipeline instance.
"""
import logging
import tempfile
from pathlib import Path
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
manager = get_plugin_manager()
ctx = HostContext(
config_dir=Path(tempfile.gettempdir()),
logger=logging.getLogger(f"abogen.pipeline.{plugin_id}"),
http_client=type("_StubHttpClient", (), {
"get": staticmethod(lambda url, **kw: None),
"post": staticmethod(lambda url, **kw: None),
})(),
)
config = EngineConfig(device=device, lang_code=lang_code)
engine = manager.create_engine(plugin_id, context=ctx, model_path=None, config=config)
return Pipeline(engine)
"""TTS Plugin Architecture — direct utility functions.
Provides helpers that replace the former compatibility adapter by
calling the Plugin Manager directly.
"""
from __future__ import annotations
from typing import Any, Iterator
import numpy as np
from abogen.tts_plugin.plugin_manager import get_plugin_manager
def get_voices(plugin_id: str) -> tuple[str, ...]:
"""Return the voice-id tuple for *plugin_id*.
Uses the official Plugin Architecture: PluginManager → Engine → VoiceLister.
First checks plugin manifest for static voice catalog.
"""
import logging
import tempfile
from pathlib import Path
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
manager = get_plugin_manager()
if not manager.has_plugin(plugin_id):
return ()
# Check manifest for static voice catalog
plugin_info = manager.get_plugin(plugin_id)
if plugin_info is not None:
manifest = plugin_info.get("manifest")
if manifest is not None and manifest.voices is not None:
return tuple(v.id for v in manifest.voices)
ctx = HostContext(
config_dir=Path(tempfile.gettempdir()),
logger=logging.getLogger(f"abogen.utils.{plugin_id}"),
http_client=type("_StubHttpClient", (), {
"get": staticmethod(lambda url, **kw: None),
"post": staticmethod(lambda url, **kw: None),
})(),
)
try:
engine = manager.create_engine(
plugin_id,
context=ctx,
model_path=None,
config=EngineConfig(device="cpu"),
)
except Exception:
return ()
try:
from abogen.tts_plugin.capabilities import VoiceLister
if isinstance(engine, VoiceLister):
manifests = engine.listVoices("builtin")
return tuple(v.id for v in manifests)
return ()
except Exception:
return ()
finally:
engine.dispose()
def get_default_voice(plugin_id: str, fallback: str = "") -> str:
"""Return the first voice of *plugin_id*, or *fallback*."""
voices = get_voices(plugin_id)
return voices[0] if voices else fallback
def is_plugin_registered(plugin_id: str) -> bool:
"""Check whether *plugin_id* is loaded by the Plugin Manager."""
return get_plugin_manager().has_plugin(plugin_id)
def resolve_voice_to_plugin(spec: str, fallback: str = "kokoro") -> str:
"""Determine which plugin owns the given voice specification.
Resolution rules:
1. Empty spec -> fallback
2. Kokoro formula (contains '*' or '+') -> "kokoro"
3. Exact voice-id match against loaded plugins -> plugin id
4. Unknown voice -> fallback
"""
raw = str(spec or "").strip()
if not raw:
return fallback
if "*" in raw or "+" in raw:
return "kokoro"
upper = raw.upper()
manager = get_plugin_manager()
for manifest in manager.list_plugins():
for voice_source in manifest.engine.voiceSources:
if voice_source.type == "list" and isinstance(voice_source.config, dict):
try:
engine = manager.create_engine(manifest.id)
try:
if hasattr(engine, "listVoices"):
voice_manifests = engine.listVoices(voice_source.id)
voice_ids = [v.id.upper() for v in voice_manifests]
if upper in voice_ids:
return manifest.id
finally:
engine.dispose()
except Exception:
continue
return fallback
class Pipeline:
"""Callable wrapper around Engine / EngineSession.
Presents the same interface that old callers expect::
pipeline = create_pipeline("kokoro", lang_code="a", device="cpu")
for segment in pipeline(text, voice="af_nova", speed=1.0):
audio = segment.audio
"""
def __init__(self, engine: Any, **engine_kwargs: Any) -> None:
self._engine = engine
self._engine_kwargs = engine_kwargs
self._session: Any = None
def _ensure_session(self) -> Any:
if self._session is None:
self._session = self._engine.createSession()
return self._session
def __call__(
self,
text: str,
voice: str = "default",
speed: float = 1.0,
split_pattern: str | None = None,
**kwargs: Any,
) -> Iterator[Any]:
from abogen.tts_plugin.types import (
AudioFormat,
ParameterValues,
SynthesisRequest,
VoiceSelection,
)
session = self._ensure_session()
params: dict[str, Any] = {"speed": speed}
if split_pattern is not None:
params["split_pattern"] = split_pattern
params.update(kwargs)
request = SynthesisRequest(
text=text,
voice=VoiceSelection(source="builtin", key=voice),
parameters=ParameterValues(values=params),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
audio_array = np.frombuffer(result.data, dtype=np.float32)
from dataclasses import dataclass
@dataclass
class Segment:
graphemes: str
audio: np.ndarray
yield Segment(graphemes=text, audio=audio_array)
def dispose(self) -> None:
if self._session is not None:
try:
self._session.dispose()
except Exception:
pass
self._session = None
def __del__(self) -> None:
self.dispose()
def create_pipeline(
plugin_id: str,
*,
lang_code: str = "a",
device: str = "cpu",
) -> Pipeline:
"""Create a callable TTS pipeline via the Plugin Architecture.
Builds a proper HostContext and EngineConfig, then delegates to the
PluginManager to create the engine. Returns a :class:`Pipeline` whose
``__call__`` interface matches the callable protocol used by consumers.
Args:
plugin_id: Plugin identifier (e.g., "kokoro", "supertonic").
lang_code: Language code for the engine.
device: Device to use (e.g., "cpu", "cuda:0").
Returns:
A callable Pipeline instance.
"""
import logging
import tempfile
from pathlib import Path
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
manager = get_plugin_manager()
ctx = HostContext(
config_dir=Path(tempfile.gettempdir()),
logger=logging.getLogger(f"abogen.pipeline.{plugin_id}"),
http_client=type("_StubHttpClient", (), {
"get": staticmethod(lambda url, **kw: None),
"post": staticmethod(lambda url, **kw: None),
})(),
)
config = EngineConfig(device=device, lang_code=lang_code)
engine = manager.create_engine(plugin_id, context=ctx, model_path=None, config=config)
return Pipeline(engine)
abogen/utils.py
+548 -548
View File
File diff suppressed because it is too large Load Diff
abogen/voice_cache.py
+146 -146
View File
@@ -1,146 +1,146 @@
from __future__ import annotations
import os
import threading
from typing import Callable, Dict, Iterable, Optional, Set, Tuple
try: # pragma: no cover - optional dependency guard
from huggingface_hub import hf_hub_download # type: ignore
from huggingface_hub.utils import LocalEntryNotFoundError # type: ignore
except Exception: # pragma: no cover - import fallback
hf_hub_download = None # type: ignore[assignment]
LocalEntryNotFoundError = None # type: ignore[assignment]
if LocalEntryNotFoundError is None: # pragma: no cover - fallback for tests
class LocalEntryNotFoundError(Exception):
pass
from abogen.tts_plugin.utils import get_voices
_CACHE_LOCK = threading.Lock()
_CACHED_VOICES: Set[str] = set()
_BOOTSTRAP_LOCK = threading.Lock()
_BOOTSTRAPPED = False
def _normalize_targets(voices: Optional[Iterable[str]]) -> Set[str]:
kokoro_voices = get_voices("kokoro")
if not voices:
return set(kokoro_voices)
normalized: Set[str] = set()
for voice in voices:
if not voice:
continue
voice_id = str(voice).strip()
if not voice_id:
continue
if voice_id in kokoro_voices:
normalized.add(voice_id)
return normalized
def ensure_voice_assets(
voices: Optional[Iterable[str]] = None,
*,
repo_id: str = "hexgrad/Kokoro-82M",
cache_dir: Optional[str] = None,
on_progress: Optional[Callable[[str], None]] = None,
) -> Tuple[Set[str], Dict[str, str]]:
"""Ensure Kokoro voice weight files are present locally.
Returns a tuple of (downloaded voices, errors) where errors maps the
voice id to the underlying exception message.
"""
if hf_hub_download is None:
raise RuntimeError("huggingface_hub is required to cache voices")
effective_cache_dir = cache_dir
if effective_cache_dir is None:
env_cache_dir = os.environ.get("ABOGEN_VOICE_CACHE_DIR", "").strip()
effective_cache_dir = env_cache_dir or None
targets = _normalize_targets(voices)
if not targets:
return set(), {}
with _CACHE_LOCK:
missing = [voice for voice in targets if voice not in _CACHED_VOICES]
downloaded: Set[str] = set()
errors: Dict[str, str] = {}
for voice_id in missing:
if on_progress:
on_progress(f"Fetching voice asset '{voice_id}'")
try:
downloaded_flag = _ensure_single_voice_asset(
voice_id,
repo_id=repo_id,
cache_dir=effective_cache_dir,
)
except Exception as exc: # pragma: no cover - network variance
errors[voice_id] = str(exc)
continue
if downloaded_flag:
downloaded.add(voice_id)
with _CACHE_LOCK:
_CACHED_VOICES.add(voice_id)
return downloaded, errors
def bootstrap_voice_cache(
voices: Optional[Iterable[str]] = None,
*,
repo_id: str = "hexgrad/Kokoro-82M",
cache_dir: Optional[str] = None,
on_progress: Optional[Callable[[str], None]] = None,
) -> Tuple[Set[str], Dict[str, str]]:
"""Ensure voices are cached once per process.
Subsequent calls are no-ops and return empty structures.
"""
global _BOOTSTRAPPED
with _BOOTSTRAP_LOCK:
if _BOOTSTRAPPED:
return set(), {}
downloaded, errors = ensure_voice_assets(
voices,
repo_id=repo_id,
cache_dir=cache_dir,
on_progress=on_progress,
)
_BOOTSTRAPPED = True
return downloaded, errors
def _ensure_single_voice_asset(
voice_id: str,
*,
repo_id: str,
cache_dir: Optional[str],
) -> bool:
if hf_hub_download is None:
raise RuntimeError("huggingface_hub is required to cache voices")
filename = f"voices/{voice_id}.pt"
common_kwargs = {
"repo_id": repo_id,
"filename": filename,
}
if cache_dir is not None:
common_kwargs["cache_dir"] = cache_dir
try:
hf_hub_download(local_files_only=True, **common_kwargs)
return False
except LocalEntryNotFoundError:
pass
hf_hub_download(resume_download=True, **common_kwargs)
return True
from __future__ import annotations
import os
import threading
from typing import Callable, Dict, Iterable, Optional, Set, Tuple
try: # pragma: no cover - optional dependency guard
from huggingface_hub import hf_hub_download # type: ignore
from huggingface_hub.utils import LocalEntryNotFoundError # type: ignore
except Exception: # pragma: no cover - import fallback
hf_hub_download = None # type: ignore[assignment]
LocalEntryNotFoundError = None # type: ignore[assignment]
if LocalEntryNotFoundError is None: # pragma: no cover - fallback for tests
class LocalEntryNotFoundError(Exception):
pass
from abogen.tts_plugin.utils import get_voices
_CACHE_LOCK = threading.Lock()
_CACHED_VOICES: Set[str] = set()
_BOOTSTRAP_LOCK = threading.Lock()
_BOOTSTRAPPED = False
def _normalize_targets(voices: Optional[Iterable[str]]) -> Set[str]:
kokoro_voices = get_voices("kokoro")
if not voices:
return set(kokoro_voices)
normalized: Set[str] = set()
for voice in voices:
if not voice:
continue
voice_id = str(voice).strip()
if not voice_id:
continue
if voice_id in kokoro_voices:
normalized.add(voice_id)
return normalized
def ensure_voice_assets(
voices: Optional[Iterable[str]] = None,
*,
repo_id: str = "hexgrad/Kokoro-82M",
cache_dir: Optional[str] = None,
on_progress: Optional[Callable[[str], None]] = None,
) -> Tuple[Set[str], Dict[str, str]]:
"""Ensure Kokoro voice weight files are present locally.
Returns a tuple of (downloaded voices, errors) where errors maps the
voice id to the underlying exception message.
"""
if hf_hub_download is None:
raise RuntimeError("huggingface_hub is required to cache voices")
effective_cache_dir = cache_dir
if effective_cache_dir is None:
env_cache_dir = os.environ.get("ABOGEN_VOICE_CACHE_DIR", "").strip()
effective_cache_dir = env_cache_dir or None
targets = _normalize_targets(voices)
if not targets:
return set(), {}
with _CACHE_LOCK:
missing = [voice for voice in targets if voice not in _CACHED_VOICES]
downloaded: Set[str] = set()
errors: Dict[str, str] = {}
for voice_id in missing:
if on_progress:
on_progress(f"Fetching voice asset '{voice_id}'")
try:
downloaded_flag = _ensure_single_voice_asset(
voice_id,
repo_id=repo_id,
cache_dir=effective_cache_dir,
)
except Exception as exc: # pragma: no cover - network variance
errors[voice_id] = str(exc)
continue
if downloaded_flag:
downloaded.add(voice_id)
with _CACHE_LOCK:
_CACHED_VOICES.add(voice_id)
return downloaded, errors
def bootstrap_voice_cache(
voices: Optional[Iterable[str]] = None,
*,
repo_id: str = "hexgrad/Kokoro-82M",
cache_dir: Optional[str] = None,
on_progress: Optional[Callable[[str], None]] = None,
) -> Tuple[Set[str], Dict[str, str]]:
"""Ensure voices are cached once per process.
Subsequent calls are no-ops and return empty structures.
"""
global _BOOTSTRAPPED
with _BOOTSTRAP_LOCK:
if _BOOTSTRAPPED:
return set(), {}
downloaded, errors = ensure_voice_assets(
voices,
repo_id=repo_id,
cache_dir=cache_dir,
on_progress=on_progress,
)
_BOOTSTRAPPED = True
return downloaded, errors
def _ensure_single_voice_asset(
voice_id: str,
*,
repo_id: str,
cache_dir: Optional[str],
) -> bool:
if hf_hub_download is None:
raise RuntimeError("huggingface_hub is required to cache voices")
filename = f"voices/{voice_id}.pt"
common_kwargs = {
"repo_id": repo_id,
"filename": filename,
}
if cache_dir is not None:
common_kwargs["cache_dir"] = cache_dir
try:
hf_hub_download(local_files_only=True, **common_kwargs)
return False
except LocalEntryNotFoundError:
pass
hf_hub_download(resume_download=True, **common_kwargs)
return True
abogen/voice_formulas.py
+82 -82
View File
@@ -1,82 +1,82 @@
import re
from typing import List, Tuple
from abogen.tts_plugin.utils import get_voices
# Calls parsing and loads the voice to gpu or cpu
def get_new_voice(pipeline, formula, use_gpu):
try:
weighted_voice = parse_voice_formula(pipeline, formula)
# device = "cuda" if use_gpu else "cpu"
# Setting the device "cuda" gives "Error occurred: split_with_sizes(): argument 'split_sizes' (position 2)"
# error when the device is gpu. So disabling this for now.
device = "cpu"
return weighted_voice.to(device)
except Exception as e:
raise ValueError(f"Failed to create voice: {str(e)}")
def parse_formula_terms(formula: str) -> List[Tuple[str, float]]:
if not formula or not formula.strip():
raise ValueError("Empty voice formula")
terms: List[Tuple[str, float]] = []
kokoro_voices = get_voices("kokoro")
for segment in formula.split("+"):
part = segment.strip()
if not part:
continue
if "*" not in part:
raise ValueError("Each component must be in the form voice*weight")
voice_name, raw_weight = part.split("*", 1)
voice_name = voice_name.strip()
if voice_name not in kokoro_voices:
raise ValueError(f"Unknown voice: {voice_name}")
try:
weight = float(raw_weight.strip())
except ValueError as exc:
raise ValueError(f"Invalid weight for {voice_name}") from exc
if weight <= 0:
raise ValueError(f"Weight for {voice_name} must be positive")
terms.append((voice_name, weight))
if not terms:
raise ValueError("Voice weights must sum to a positive value")
return terms
def parse_voice_formula(pipeline, formula):
terms = parse_formula_terms(formula)
total_weight = sum(weight for _, weight in terms)
if total_weight <= 0:
raise ValueError("Voice weights must sum to a positive value")
weighted_sum = None
for voice_name, weight in terms:
normalized_weight = weight / total_weight if total_weight > 0 else weight
voice_tensor = pipeline.load_single_voice(voice_name)
if weighted_sum is None:
weighted_sum = normalized_weight * voice_tensor
else:
weighted_sum += normalized_weight * voice_tensor
if weighted_sum is None:
raise ValueError("Voice formula produced no components")
return weighted_sum
def calculate_sum_from_formula(formula):
weights = re.findall(r"\* *([\d.]+)", formula)
total_sum = sum(float(weight) for weight in weights)
return total_sum
def extract_voice_ids(formula: str) -> List[str]:
return [voice for voice, _ in parse_formula_terms(formula)]
import re
from typing import List, Tuple
from abogen.tts_plugin.utils import get_voices
# Calls parsing and loads the voice to gpu or cpu
def get_new_voice(pipeline, formula, use_gpu):
try:
weighted_voice = parse_voice_formula(pipeline, formula)
# device = "cuda" if use_gpu else "cpu"
# Setting the device "cuda" gives "Error occurred: split_with_sizes(): argument 'split_sizes' (position 2)"
# error when the device is gpu. So disabling this for now.
device = "cpu"
return weighted_voice.to(device)
except Exception as e:
raise ValueError(f"Failed to create voice: {str(e)}")
def parse_formula_terms(formula: str) -> List[Tuple[str, float]]:
if not formula or not formula.strip():
raise ValueError("Empty voice formula")
terms: List[Tuple[str, float]] = []
kokoro_voices = get_voices("kokoro")
for segment in formula.split("+"):
part = segment.strip()
if not part:
continue
if "*" not in part:
raise ValueError("Each component must be in the form voice*weight")
voice_name, raw_weight = part.split("*", 1)
voice_name = voice_name.strip()
if voice_name not in kokoro_voices:
raise ValueError(f"Unknown voice: {voice_name}")
try:
weight = float(raw_weight.strip())
except ValueError as exc:
raise ValueError(f"Invalid weight for {voice_name}") from exc
if weight <= 0:
raise ValueError(f"Weight for {voice_name} must be positive")
terms.append((voice_name, weight))
if not terms:
raise ValueError("Voice weights must sum to a positive value")
return terms
def parse_voice_formula(pipeline, formula):
terms = parse_formula_terms(formula)
total_weight = sum(weight for _, weight in terms)
if total_weight <= 0:
raise ValueError("Voice weights must sum to a positive value")
weighted_sum = None
for voice_name, weight in terms:
normalized_weight = weight / total_weight if total_weight > 0 else weight
voice_tensor = pipeline.load_single_voice(voice_name)
if weighted_sum is None:
weighted_sum = normalized_weight * voice_tensor
else:
weighted_sum += normalized_weight * voice_tensor
if weighted_sum is None:
raise ValueError("Voice formula produced no components")
return weighted_sum
def calculate_sum_from_formula(formula):
weights = re.findall(r"\* *([\d.]+)", formula)
total_sum = sum(float(weight) for weight in weights)
return total_sum
def extract_voice_ids(formula: str) -> List[str]:
return [voice for voice, _ in parse_formula_terms(formula)]
abogen/voice_profiles.py
+230 -230
View File
@@ -1,230 +1,230 @@
import json
import os
from typing import Any, Dict, Iterable, List, Tuple
from abogen.tts_plugin.utils import get_voices, is_plugin_registered
from abogen.utils import get_user_config_path
def _get_profiles_path():
config_path = get_user_config_path()
config_dir = os.path.dirname(config_path)
return os.path.join(config_dir, "voice_profiles.json")
def load_profiles():
"""Load all voice profiles from JSON file."""
path = _get_profiles_path()
if os.path.exists(path):
try:
with open(path, "r", encoding="utf-8") as f:
data = json.load(f)
# always expect abogen_voice_profiles wrapper
if isinstance(data, dict) and "abogen_voice_profiles" in data:
return data["abogen_voice_profiles"]
# fallback: treat as profiles dict
if isinstance(data, dict):
return data
except Exception:
return {}
return {}
def save_profiles(profiles):
"""Save all voice profiles to JSON file."""
path = _get_profiles_path()
os.makedirs(os.path.dirname(path), exist_ok=True)
with open(path, "w", encoding="utf-8") as f:
# always save with abogen_voice_profiles wrapper
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
def delete_profile(name):
"""Remove a profile by name."""
profiles = load_profiles()
if name in profiles:
del profiles[name]
save_profiles(profiles)
def duplicate_profile(src, dest):
"""Duplicate an existing profile."""
profiles = load_profiles()
if src in profiles and dest:
profiles[dest] = profiles[src]
save_profiles(profiles)
def export_profiles(export_path):
"""Export all profiles to specified JSON file."""
profiles = load_profiles()
with open(export_path, "w", encoding="utf-8") as f:
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
def serialize_profiles() -> Dict[str, Dict[str, Iterable[Tuple[str, float]]]]:
"""Return profiles in canonical dictionary form."""
return load_profiles()
def _normalize_supertonic_voice(value: Any) -> str:
raw = str(value or "").strip().upper()
supertonic_voices = get_voices("supertonic")
return raw if raw in supertonic_voices else "M1"
def _coerce_supertonic_steps(value: Any) -> int:
try:
steps = int(value)
except (TypeError, ValueError):
return 5
return max(2, min(15, steps))
def _coerce_supertonic_speed(value: Any) -> float:
try:
speed = float(value)
except (TypeError, ValueError):
return 1.0
return max(0.7, min(2.0, speed))
def normalize_profile_entry(entry: Any) -> Dict[str, Any]:
"""Normalize a stored profile entry.
Backwards compatible:
- Legacy Kokoro-only entries: {language, voices}
- New entries: include provider.
"""
if not isinstance(entry, dict):
return {}
provider = str(entry.get("provider") or "kokoro").strip().lower()
if not is_plugin_registered(provider):
provider = "kokoro"
language = str(entry.get("language") or "a").strip().lower() or "a"
if provider == "supertonic":
return {
"provider": "supertonic",
"language": language,
"voice": _normalize_supertonic_voice(
entry.get("voice") or entry.get("voice_name") or entry.get("name")
),
"total_steps": _coerce_supertonic_steps(
entry.get("total_steps")
or entry.get("supertonic_total_steps")
or entry.get("quality")
),
"speed": _coerce_supertonic_speed(
entry.get("speed") or entry.get("supertonic_speed")
),
}
voices = _normalize_voice_entries(entry.get("voices", []))
if not voices:
return {}
return {
"provider": "kokoro",
"language": language,
"voices": voices,
}
def _normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
normalized: List[Tuple[str, float]] = []
kokoro_voices = get_voices("kokoro")
for item in entries or []:
if isinstance(item, dict):
voice = item.get("id") or item.get("voice")
weight = item.get("weight")
elif isinstance(item, (list, tuple)) and len(item) >= 2:
voice, weight = item[0], item[1]
else:
continue
if voice not in kokoro_voices:
continue
if weight is None:
continue
try:
weight_val = float(weight)
except (TypeError, ValueError):
continue
if weight_val <= 0:
continue
normalized.append((voice, weight_val))
return normalized
def normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
"""Public helper to normalize voice-weight pairs from arbitrary payloads."""
return _normalize_voice_entries(entries)
def save_profile(name: str, *, language: str, voices: Iterable) -> None:
"""Persist a single profile after validating its data."""
name = (name or "").strip()
if not name:
raise ValueError("Profile name is required")
normalized = _normalize_voice_entries(voices)
if not normalized:
raise ValueError("At least one voice with a weight above zero is required")
if not language:
language = "a"
profiles = load_profiles()
profiles[name] = {"provider": "kokoro", "language": language, "voices": normalized}
save_profiles(profiles)
def remove_profile(name: str) -> None:
delete_profile(name)
def import_profiles_data(data: Dict, *, replace_existing: bool = False) -> List[str]:
"""Merge profiles from a dictionary structure and persist them.
Returns the list of profile names that were added or updated.
"""
if not isinstance(data, dict):
raise ValueError("Invalid profile payload")
if "abogen_voice_profiles" in data:
data = data["abogen_voice_profiles"]
if not isinstance(data, dict):
raise ValueError("Invalid profile payload")
current = load_profiles()
updated: List[str] = []
for name, entry in data.items():
normalized = normalize_profile_entry(entry)
if not normalized:
continue
if name in current and not replace_existing:
# skip duplicates unless explicit replacement is requested
continue
current[name] = normalized
updated.append(name)
if updated:
save_profiles(current)
return updated
def export_profiles_payload(names: Iterable[str] | None = None) -> Dict[str, Dict]:
"""Return profiles limited to the provided names for download/export."""
profiles = load_profiles()
if names is None:
subset = profiles
else:
subset = {name: profiles[name] for name in names if name in profiles}
return {"abogen_voice_profiles": subset}
import json
import os
from typing import Any, Dict, Iterable, List, Tuple
from abogen.tts_plugin.utils import get_voices, is_plugin_registered
from abogen.utils import get_user_config_path
def _get_profiles_path():
config_path = get_user_config_path()
config_dir = os.path.dirname(config_path)
return os.path.join(config_dir, "voice_profiles.json")
def load_profiles():
"""Load all voice profiles from JSON file."""
path = _get_profiles_path()
if os.path.exists(path):
try:
with open(path, "r", encoding="utf-8") as f:
data = json.load(f)
# always expect abogen_voice_profiles wrapper
if isinstance(data, dict) and "abogen_voice_profiles" in data:
return data["abogen_voice_profiles"]
# fallback: treat as profiles dict
if isinstance(data, dict):
return data
except Exception:
return {}
return {}
def save_profiles(profiles):
"""Save all voice profiles to JSON file."""
path = _get_profiles_path()
os.makedirs(os.path.dirname(path), exist_ok=True)
with open(path, "w", encoding="utf-8") as f:
# always save with abogen_voice_profiles wrapper
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
def delete_profile(name):
"""Remove a profile by name."""
profiles = load_profiles()
if name in profiles:
del profiles[name]
save_profiles(profiles)
def duplicate_profile(src, dest):
"""Duplicate an existing profile."""
profiles = load_profiles()
if src in profiles and dest:
profiles[dest] = profiles[src]
save_profiles(profiles)
def export_profiles(export_path):
"""Export all profiles to specified JSON file."""
profiles = load_profiles()
with open(export_path, "w", encoding="utf-8") as f:
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
def serialize_profiles() -> Dict[str, Dict[str, Iterable[Tuple[str, float]]]]:
"""Return profiles in canonical dictionary form."""
return load_profiles()
def _normalize_supertonic_voice(value: Any) -> str:
raw = str(value or "").strip().upper()
supertonic_voices = get_voices("supertonic")
return raw if raw in supertonic_voices else "M1"
def _coerce_supertonic_steps(value: Any) -> int:
try:
steps = int(value)
except (TypeError, ValueError):
return 5
return max(2, min(15, steps))
def _coerce_supertonic_speed(value: Any) -> float:
try:
speed = float(value)
except (TypeError, ValueError):
return 1.0
return max(0.7, min(2.0, speed))
def normalize_profile_entry(entry: Any) -> Dict[str, Any]:
"""Normalize a stored profile entry.
Backwards compatible:
- Legacy Kokoro-only entries: {language, voices}
- New entries: include provider.
"""
if not isinstance(entry, dict):
return {}
provider = str(entry.get("provider") or "kokoro").strip().lower()
if not is_plugin_registered(provider):
provider = "kokoro"
language = str(entry.get("language") or "a").strip().lower() or "a"
if provider == "supertonic":
return {
"provider": "supertonic",
"language": language,
"voice": _normalize_supertonic_voice(
entry.get("voice") or entry.get("voice_name") or entry.get("name")
),
"total_steps": _coerce_supertonic_steps(
entry.get("total_steps")
or entry.get("supertonic_total_steps")
or entry.get("quality")
),
"speed": _coerce_supertonic_speed(
entry.get("speed") or entry.get("supertonic_speed")
),
}
voices = _normalize_voice_entries(entry.get("voices", []))
if not voices:
return {}
return {
"provider": "kokoro",
"language": language,
"voices": voices,
}
def _normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
normalized: List[Tuple[str, float]] = []
kokoro_voices = get_voices("kokoro")
for item in entries or []:
if isinstance(item, dict):
voice = item.get("id") or item.get("voice")
weight = item.get("weight")
elif isinstance(item, (list, tuple)) and len(item) >= 2:
voice, weight = item[0], item[1]
else:
continue
if voice not in kokoro_voices:
continue
if weight is None:
continue
try:
weight_val = float(weight)
except (TypeError, ValueError):
continue
if weight_val <= 0:
continue
normalized.append((voice, weight_val))
return normalized
def normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
"""Public helper to normalize voice-weight pairs from arbitrary payloads."""
return _normalize_voice_entries(entries)
def save_profile(name: str, *, language: str, voices: Iterable) -> None:
"""Persist a single profile after validating its data."""
name = (name or "").strip()
if not name:
raise ValueError("Profile name is required")
normalized = _normalize_voice_entries(voices)
if not normalized:
raise ValueError("At least one voice with a weight above zero is required")
if not language:
language = "a"
profiles = load_profiles()
profiles[name] = {"provider": "kokoro", "language": language, "voices": normalized}
save_profiles(profiles)
def remove_profile(name: str) -> None:
delete_profile(name)
def import_profiles_data(data: Dict, *, replace_existing: bool = False) -> List[str]:
"""Merge profiles from a dictionary structure and persist them.
Returns the list of profile names that were added or updated.
"""
if not isinstance(data, dict):
raise ValueError("Invalid profile payload")
if "abogen_voice_profiles" in data:
data = data["abogen_voice_profiles"]
if not isinstance(data, dict):
raise ValueError("Invalid profile payload")
current = load_profiles()
updated: List[str] = []
for name, entry in data.items():
normalized = normalize_profile_entry(entry)
if not normalized:
continue
if name in current and not replace_existing:
# skip duplicates unless explicit replacement is requested
continue
current[name] = normalized
updated.append(name)
if updated:
save_profiles(current)
return updated
def export_profiles_payload(names: Iterable[str] | None = None) -> Dict[str, Dict]:
"""Return profiles limited to the provided names for download/export."""
profiles = load_profiles()
if names is None:
subset = profiles
else:
subset = {name: profiles[name] for name in names if name in profiles}
return {"abogen_voice_profiles": subset}
abogen/webui/debug_tts_runner.py
+254 -254
View File
@@ -1,254 +1,254 @@
from __future__ import annotations
import json
import re
import uuid
from dataclasses import dataclass
from pathlib import Path
from typing import Any, Dict, List, Mapping, Optional, Sequence, Tuple
import numpy as np
from abogen.debug_tts_samples import MARKER_PREFIX, MARKER_SUFFIX, build_debug_epub, iter_expected_codes
from abogen.kokoro_text_normalization import normalize_for_pipeline
from abogen.normalization_settings import build_apostrophe_config
from abogen.text_extractor import extract_from_path
from abogen.voice_cache import ensure_voice_assets
from abogen.webui.conversion_runner import SAMPLE_RATE, SPLIT_PATTERN, _select_device, _to_float32, _resolve_voice, _spec_to_voice_ids
from abogen.tts_plugin.utils import create_pipeline
_MARKER_RE = re.compile(re.escape(MARKER_PREFIX) + r"(?P<code>[A-Z0-9_]+)" + re.escape(MARKER_SUFFIX))
@dataclass(frozen=True)
class DebugWavArtifact:
label: str
filename: str
code: Optional[str] = None
text: Optional[str] = None
def _resolve_voice_setting(value: str) -> tuple[str, Optional[str], Optional[str]]:
"""Resolve settings voice strings into a pipeline-ready voice spec.
Supports "profile:<name>" by converting it into a concrete voice formula.
Returns (resolved_voice_spec, profile_name, profile_language).
"""
from abogen.webui.routes.utils.voice import resolve_voice_setting
return resolve_voice_setting(value)
def _load_pipeline(language: str, use_gpu: bool) -> Any:
device = "cpu"
if use_gpu:
device = _select_device()
return create_pipeline("kokoro", lang_code=language, device=device)
def _extract_cases_from_text(text: str) -> List[Tuple[str, str]]:
raw = str(text or "")
matches = list(_MARKER_RE.finditer(raw))
cases: List[Tuple[str, str]] = []
if not matches:
return cases
for idx, match in enumerate(matches):
code = match.group("code")
start = match.end()
end = matches[idx + 1].start() if idx + 1 < len(matches) else len(raw)
snippet = raw[start:end]
# Keep it small and predictable: collapse whitespace.
snippet = " ".join(snippet.strip().split())
cases.append((code, snippet))
return cases
def _spoken_id(code: str) -> str:
# Make IDs pronounceable and stable (avoid reading as a word).
out: List[str] = []
for ch in str(code or ""):
if ch == "_":
out.append(" ")
elif ch.isalnum():
out.append(ch)
else:
out.append(" ")
# Add spaces between alnum to encourage letter-by-letter reading.
spaced = " ".join("".join(out).split())
return spaced
def run_debug_tts_wavs(
*,
output_root: Path,
settings: Mapping[str, Any],
epub_path: Optional[Path] = None,
) -> Dict[str, Any]:
"""Generate WAV artifacts for the debug EPUB samples.
Writes:
- overall.wav: concatenation of all samples
- case_<CODE>.wav: each sample rendered separately
- manifest.json: metadata + file list
"""
output_root = Path(output_root)
output_root.mkdir(parents=True, exist_ok=True)
run_id = uuid.uuid4().hex
run_dir = output_root / "debug" / run_id
run_dir.mkdir(parents=True, exist_ok=True)
if epub_path is None:
epub_path = run_dir / "abogen_debug_samples.epub"
build_debug_epub(epub_path)
else:
epub_path = Path(epub_path)
extraction = extract_from_path(epub_path)
combined_text = extraction.combined_text or "\n\n".join((c.text or "") for c in extraction.chapters)
cases = _extract_cases_from_text(combined_text)
# Prefer the canonical sample catalog for text (EPUB extraction may include headings).
try:
from abogen.debug_tts_samples import DEBUG_TTS_SAMPLES
sample_text_by_code = {sample.code: sample.text for sample in DEBUG_TTS_SAMPLES}
except Exception:
sample_text_by_code = {}
expected = list(iter_expected_codes())
found_codes = {code for code, _ in cases}
missing = [code for code in expected if code not in found_codes]
if missing:
raise RuntimeError(f"Debug EPUB missing expected codes: {', '.join(missing)}")
language = str(settings.get("language") or "a").strip() or "a"
# Kokoro's KPipeline expects short language codes like "a" (American English),
# but older settings may store ISO-like values such as "en".
language_aliases = {
"en": "a",
"en-us": "a",
"en_us": "a",
"en-gb": "b",
"en_gb": "b",
"es": "e",
"es-es": "e",
"fr": "f",
"fr-fr": "f",
"hi": "h",
"it": "i",
"pt": "p",
"pt-br": "p",
"ja": "j",
"jp": "j",
"zh": "z",
"zh-cn": "z",
}
language = language_aliases.get(language.lower(), language)
voice_spec = str(settings.get("default_voice") or "").strip()
use_gpu = bool(settings.get("use_gpu", False))
speed = float(settings.get("default_speed", 1.0) or 1.0)
# Settings may store "profile:<name>" which is not a Kokoro voice ID.
# Resolve it to a concrete voice formula (e.g. "af_heart*0.5+...") so Kokoro
# doesn't attempt to download a non-existent "voices/profile:<name>.pt".
try:
resolved_voice, _profile_name, profile_language = _resolve_voice_setting(voice_spec)
if resolved_voice:
voice_spec = resolved_voice
if profile_language:
language = str(profile_language).strip() or language
except Exception:
# Voice profile resolution is best-effort; fall back to raw voice_spec.
pass
# Best-effort voice caching (only for known Kokoro internal voices).
voice_ids = _spec_to_voice_ids(voice_spec)
if voice_ids:
try:
ensure_voice_assets(voice_ids)
except Exception:
# Network / optional dependency variance; debug runner can still proceed.
pass
pipeline = _load_pipeline(language, use_gpu)
voice_choice = _resolve_voice(pipeline, voice_spec, use_gpu)
apostrophe_config = build_apostrophe_config(settings=settings)
normalization_settings = dict(settings)
artifacts: List[DebugWavArtifact] = []
overall_path = run_dir / "overall.wav"
overall_audio: List[np.ndarray] = []
def synth(text: str, *, apply_normalization: bool = True) -> np.ndarray:
normalized = (
normalize_for_pipeline(
text,
config=apostrophe_config,
settings=normalization_settings,
)
if apply_normalization
else str(text or "")
)
parts: List[np.ndarray] = []
for segment in pipeline(
normalized,
voice=voice_choice,
speed=speed,
split_pattern=SPLIT_PATTERN,
):
audio = _to_float32(getattr(segment, "audio", None))
if audio.size:
parts.append(audio)
if not parts:
return np.zeros(0, dtype="float32")
return np.concatenate(parts).astype("float32", copy=False)
pause_1s = np.zeros(int(1.0 * SAMPLE_RATE), dtype="float32")
between_cases = np.zeros(int(0.35 * SAMPLE_RATE), dtype="float32")
# Per sample
for code, snippet in cases:
snippet = sample_text_by_code.get(code, snippet)
if not snippet:
continue
id_audio = synth(_spoken_id(code), apply_normalization=False)
text_audio = synth(snippet, apply_normalization=True)
audio = np.concatenate([id_audio, pause_1s, text_audio]).astype("float32", copy=False)
filename = f"case_{code}.wav"
path = run_dir / filename
# Write float32 PCM WAV.
import soundfile as sf
sf.write(path, audio, SAMPLE_RATE, subtype="FLOAT")
artifacts.append(DebugWavArtifact(label=f"{code}", filename=filename, code=code, text=snippet))
overall_audio.append(audio)
overall_audio.append(between_cases)
# Overall
if overall_audio:
combined = np.concatenate(overall_audio).astype("float32", copy=False)
else:
combined = np.zeros(0, dtype="float32")
import soundfile as sf
sf.write(overall_path, combined, SAMPLE_RATE, subtype="FLOAT")
artifacts.insert(0, DebugWavArtifact(label="Overall", filename="overall.wav", code=None, text=None))
manifest = {
"run_id": run_id,
"epub": str(epub_path),
"artifacts": [artifact.__dict__ for artifact in artifacts],
"sample_rate": SAMPLE_RATE,
}
(run_dir / "manifest.json").write_text(json.dumps(manifest, indent=2), encoding="utf-8")
try:
pipeline.dispose()
except Exception:
pass
return manifest
from __future__ import annotations
import json
import re
import uuid
from dataclasses import dataclass
from pathlib import Path
from typing import Any, Dict, List, Mapping, Optional, Sequence, Tuple
import numpy as np
from abogen.debug_tts_samples import MARKER_PREFIX, MARKER_SUFFIX, build_debug_epub, iter_expected_codes
from abogen.kokoro_text_normalization import normalize_for_pipeline
from abogen.normalization_settings import build_apostrophe_config
from abogen.text_extractor import extract_from_path
from abogen.voice_cache import ensure_voice_assets
from abogen.webui.conversion_runner import SAMPLE_RATE, SPLIT_PATTERN, _select_device, _to_float32, _resolve_voice, _spec_to_voice_ids
from abogen.tts_plugin.utils import create_pipeline
_MARKER_RE = re.compile(re.escape(MARKER_PREFIX) + r"(?P<code>[A-Z0-9_]+)" + re.escape(MARKER_SUFFIX))
@dataclass(frozen=True)
class DebugWavArtifact:
label: str
filename: str
code: Optional[str] = None
text: Optional[str] = None
def _resolve_voice_setting(value: str) -> tuple[str, Optional[str], Optional[str]]:
"""Resolve settings voice strings into a pipeline-ready voice spec.
Supports "profile:<name>" by converting it into a concrete voice formula.
Returns (resolved_voice_spec, profile_name, profile_language).
"""
from abogen.webui.routes.utils.voice import resolve_voice_setting
return resolve_voice_setting(value)
def _load_pipeline(language: str, use_gpu: bool) -> Any:
device = "cpu"
if use_gpu:
device = _select_device()
return create_pipeline("kokoro", lang_code=language, device=device)
def _extract_cases_from_text(text: str) -> List[Tuple[str, str]]:
raw = str(text or "")
matches = list(_MARKER_RE.finditer(raw))
cases: List[Tuple[str, str]] = []
if not matches:
return cases
for idx, match in enumerate(matches):
code = match.group("code")
start = match.end()
end = matches[idx + 1].start() if idx + 1 < len(matches) else len(raw)
snippet = raw[start:end]
# Keep it small and predictable: collapse whitespace.
snippet = " ".join(snippet.strip().split())
cases.append((code, snippet))
return cases
def _spoken_id(code: str) -> str:
# Make IDs pronounceable and stable (avoid reading as a word).
out: List[str] = []
for ch in str(code or ""):
if ch == "_":
out.append(" ")
elif ch.isalnum():
out.append(ch)
else:
out.append(" ")
# Add spaces between alnum to encourage letter-by-letter reading.
spaced = " ".join("".join(out).split())
return spaced
def run_debug_tts_wavs(
*,
output_root: Path,
settings: Mapping[str, Any],
epub_path: Optional[Path] = None,
) -> Dict[str, Any]:
"""Generate WAV artifacts for the debug EPUB samples.
Writes:
- overall.wav: concatenation of all samples
- case_<CODE>.wav: each sample rendered separately
- manifest.json: metadata + file list
"""
output_root = Path(output_root)
output_root.mkdir(parents=True, exist_ok=True)
run_id = uuid.uuid4().hex
run_dir = output_root / "debug" / run_id
run_dir.mkdir(parents=True, exist_ok=True)
if epub_path is None:
epub_path = run_dir / "abogen_debug_samples.epub"
build_debug_epub(epub_path)
else:
epub_path = Path(epub_path)
extraction = extract_from_path(epub_path)
combined_text = extraction.combined_text or "\n\n".join((c.text or "") for c in extraction.chapters)
cases = _extract_cases_from_text(combined_text)
# Prefer the canonical sample catalog for text (EPUB extraction may include headings).
try:
from abogen.debug_tts_samples import DEBUG_TTS_SAMPLES
sample_text_by_code = {sample.code: sample.text for sample in DEBUG_TTS_SAMPLES}
except Exception:
sample_text_by_code = {}
expected = list(iter_expected_codes())
found_codes = {code for code, _ in cases}
missing = [code for code in expected if code not in found_codes]
if missing:
raise RuntimeError(f"Debug EPUB missing expected codes: {', '.join(missing)}")
language = str(settings.get("language") or "a").strip() or "a"
# Kokoro's KPipeline expects short language codes like "a" (American English),
# but older settings may store ISO-like values such as "en".
language_aliases = {
"en": "a",
"en-us": "a",
"en_us": "a",
"en-gb": "b",
"en_gb": "b",
"es": "e",
"es-es": "e",
"fr": "f",
"fr-fr": "f",
"hi": "h",
"it": "i",
"pt": "p",
"pt-br": "p",
"ja": "j",
"jp": "j",
"zh": "z",
"zh-cn": "z",
}
language = language_aliases.get(language.lower(), language)
voice_spec = str(settings.get("default_voice") or "").strip()
use_gpu = bool(settings.get("use_gpu", False))
speed = float(settings.get("default_speed", 1.0) or 1.0)
# Settings may store "profile:<name>" which is not a Kokoro voice ID.
# Resolve it to a concrete voice formula (e.g. "af_heart*0.5+...") so Kokoro
# doesn't attempt to download a non-existent "voices/profile:<name>.pt".
try:
resolved_voice, _profile_name, profile_language = _resolve_voice_setting(voice_spec)
if resolved_voice:
voice_spec = resolved_voice
if profile_language:
language = str(profile_language).strip() or language
except Exception:
# Voice profile resolution is best-effort; fall back to raw voice_spec.
pass
# Best-effort voice caching (only for known Kokoro internal voices).
voice_ids = _spec_to_voice_ids(voice_spec)
if voice_ids:
try:
ensure_voice_assets(voice_ids)
except Exception:
# Network / optional dependency variance; debug runner can still proceed.
pass
pipeline = _load_pipeline(language, use_gpu)
voice_choice = _resolve_voice(pipeline, voice_spec, use_gpu)
apostrophe_config = build_apostrophe_config(settings=settings)
normalization_settings = dict(settings)
artifacts: List[DebugWavArtifact] = []
overall_path = run_dir / "overall.wav"
overall_audio: List[np.ndarray] = []
def synth(text: str, *, apply_normalization: bool = True) -> np.ndarray:
normalized = (
normalize_for_pipeline(
text,
config=apostrophe_config,
settings=normalization_settings,
)
if apply_normalization
else str(text or "")
)
parts: List[np.ndarray] = []
for segment in pipeline(
normalized,
voice=voice_choice,
speed=speed,
split_pattern=SPLIT_PATTERN,
):
audio = _to_float32(getattr(segment, "audio", None))
if audio.size:
parts.append(audio)
if not parts:
return np.zeros(0, dtype="float32")
return np.concatenate(parts).astype("float32", copy=False)
pause_1s = np.zeros(int(1.0 * SAMPLE_RATE), dtype="float32")
between_cases = np.zeros(int(0.35 * SAMPLE_RATE), dtype="float32")
# Per sample
for code, snippet in cases:
snippet = sample_text_by_code.get(code, snippet)
if not snippet:
continue
id_audio = synth(_spoken_id(code), apply_normalization=False)
text_audio = synth(snippet, apply_normalization=True)
audio = np.concatenate([id_audio, pause_1s, text_audio]).astype("float32", copy=False)
filename = f"case_{code}.wav"
path = run_dir / filename
# Write float32 PCM WAV.
import soundfile as sf
sf.write(path, audio, SAMPLE_RATE, subtype="FLOAT")
artifacts.append(DebugWavArtifact(label=f"{code}", filename=filename, code=code, text=snippet))
overall_audio.append(audio)
overall_audio.append(between_cases)
# Overall
if overall_audio:
combined = np.concatenate(overall_audio).astype("float32", copy=False)
else:
combined = np.zeros(0, dtype="float32")
import soundfile as sf
sf.write(overall_path, combined, SAMPLE_RATE, subtype="FLOAT")
artifacts.insert(0, DebugWavArtifact(label="Overall", filename="overall.wav", code=None, text=None))
manifest = {
"run_id": run_id,
"epub": str(epub_path),
"artifacts": [artifact.__dict__ for artifact in artifacts],
"sample_rate": SAMPLE_RATE,
}
(run_dir / "manifest.json").write_text(json.dumps(manifest, indent=2), encoding="utf-8")
try:
pipeline.dispose()
except Exception:
pass
return manifest
abogen/webui/routes/api.py
+681 -681
View File
File diff suppressed because it is too large Load Diff
abogen/webui/routes/utils/form.py
+1098 -1098
View File
File diff suppressed because it is too large Load Diff
abogen/webui/routes/utils/preview.py
+245 -245
View File
@@ -1,245 +1,245 @@
import io
import threading
from typing import Any, Dict, Iterable, List, Mapping, Optional, Tuple
import numpy as np
import soundfile as sf
from flask import current_app, send_file
from flask.typing import ResponseReturnValue
SPLIT_PATTERN = r"\n+"
SAMPLE_RATE = 24000
_preview_pipelines: Dict[Tuple[str, str], Any] = {}
_preview_pipeline_lock = threading.Lock()
def clear_preview_pipelines() -> None:
"""Dispose all cached preview pipelines and clear the cache."""
with _preview_pipeline_lock:
for pipeline in _preview_pipelines.values():
try:
pipeline.dispose()
except Exception:
pass
_preview_pipelines.clear()
def _select_device() -> str:
import platform
try:
import torch # type: ignore[import-not-found]
except Exception:
return "cpu"
system = platform.system()
if system == "Darwin" and platform.processor() == "arm":
try:
if torch.backends.mps.is_available():
return "mps"
except Exception:
pass
return "cpu"
try:
if torch.cuda.is_available():
return "cuda"
except Exception:
pass
return "cpu"
def _resolve_pipeline(language: str, use_gpu: bool) -> Tuple[Any, bool]:
devices: List[str] = ["cpu"]
if use_gpu:
preferred = _select_device()
if preferred != "cpu":
devices.insert(0, preferred)
last_error: Optional[Exception] = None
for device in devices:
try:
return get_preview_pipeline(language, device), device != "cpu"
except Exception as exc:
last_error = exc
raise RuntimeError("Preview pipeline is unavailable") from last_error
def _to_float32(audio_segment) -> np.ndarray:
if audio_segment is None:
return np.zeros(0, dtype="float32")
tensor = audio_segment
if hasattr(tensor, "detach"):
tensor = tensor.detach()
if hasattr(tensor, "cpu"):
try:
tensor = tensor.cpu()
except Exception:
pass
if hasattr(tensor, "numpy"):
return np.asarray(tensor.numpy(), dtype="float32").reshape(-1)
return np.asarray(tensor, dtype="float32").reshape(-1)
def get_preview_pipeline(language: str, device: str) -> Any:
key = (language, device)
with _preview_pipeline_lock:
pipeline = _preview_pipelines.get(key)
if pipeline is not None:
return pipeline
from abogen.tts_plugin.utils import create_pipeline
pipeline = create_pipeline("kokoro", lang_code=language, device=device)
_preview_pipelines[key] = pipeline
return pipeline
def generate_preview_audio(
text: str,
voice_spec: str,
language: str,
speed: float,
use_gpu: bool,
tts_provider: str = "kokoro",
supertonic_total_steps: int = 5,
max_seconds: float = 8.0,
pronunciation_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
manual_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
speakers: Optional[Mapping[str, Any]] = None,
) -> bytes:
if not text.strip():
raise ValueError("Preview text is required")
provider = (tts_provider or "kokoro").strip().lower()
# Apply pronunciation/manual overrides first so tokens like `Unfu*k` still match
# before any downstream normalization potentially strips punctuation.
source_text = text
if pronunciation_overrides or manual_overrides or speakers:
try:
from abogen.webui import conversion_runner as runner
class _PreviewJob:
def __init__(self):
self.language = language
self.voice = voice_spec
self.speakers = speakers
self.manual_overrides = list(manual_overrides or [])
self.pronunciation_overrides = list(pronunciation_overrides or [])
job = _PreviewJob()
merged = runner._merge_pronunciation_overrides(job)
rules = runner._compile_pronunciation_rules(merged)
source_text = runner._apply_pronunciation_rules(source_text, rules)
except Exception:
current_app.logger.exception("Preview override application failed; using raw text")
source_text = text
normalized_text = source_text
if provider != "supertonic":
try:
from abogen.kokoro_text_normalization import normalize_for_pipeline
normalized_text = normalize_for_pipeline(source_text)
except Exception:
current_app.logger.exception("Preview normalization failed; using raw text")
normalized_text = source_text
if provider == "supertonic":
from abogen.tts_plugin.utils import create_pipeline
pipeline = create_pipeline("supertonic")
segments = pipeline(
normalized_text,
voice=voice_spec,
speed=speed,
split_pattern=SPLIT_PATTERN,
total_steps=supertonic_total_steps,
)
else:
pipeline, pipeline_uses_gpu = _resolve_pipeline(language, use_gpu)
if pipeline is None:
raise RuntimeError("Preview pipeline is unavailable")
voice_choice: Any = voice_spec
if voice_spec and "*" in voice_spec:
from abogen.voice_formulas import get_new_voice
voice_choice = get_new_voice(pipeline, voice_spec, pipeline_uses_gpu)
segments = pipeline(
normalized_text,
voice=voice_choice,
speed=speed,
split_pattern=SPLIT_PATTERN,
)
audio_chunks: List[np.ndarray] = []
accumulated = 0
max_samples = int(max(1.0, max_seconds) * SAMPLE_RATE)
for segment in segments:
graphemes = getattr(segment, "graphemes", "").strip()
if not graphemes:
continue
audio = _to_float32(getattr(segment, "audio", None))
if audio.size == 0:
continue
remaining = max_samples - accumulated
if remaining <= 0:
break
if audio.shape[0] > remaining:
audio = audio[:remaining]
audio_chunks.append(audio)
accumulated += audio.shape[0]
if accumulated >= max_samples:
break
if not audio_chunks:
raise RuntimeError("Preview could not be generated")
audio_data = np.concatenate(audio_chunks)
buffer = io.BytesIO()
sf.write(buffer, audio_data, SAMPLE_RATE, format="WAV")
return buffer.getvalue()
def synthesize_preview(
text: str,
voice_spec: str,
language: str,
speed: float,
use_gpu: bool,
tts_provider: str = "kokoro",
supertonic_total_steps: int = 5,
max_seconds: float = 8.0,
pronunciation_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
manual_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
speakers: Optional[Mapping[str, Any]] = None,
) -> ResponseReturnValue:
try:
audio_bytes = generate_preview_audio(
text=text,
voice_spec=voice_spec,
language=language,
speed=speed,
use_gpu=use_gpu,
tts_provider=tts_provider,
supertonic_total_steps=supertonic_total_steps,
max_seconds=max_seconds,
pronunciation_overrides=pronunciation_overrides,
manual_overrides=manual_overrides,
speakers=speakers,
)
except Exception as e:
raise e
buffer = io.BytesIO(audio_bytes)
response = send_file(
buffer,
mimetype="audio/wav",
as_attachment=False,
download_name="speaker_preview.wav",
)
response.headers["Cache-Control"] = "no-store"
return response
import io
import threading
from typing import Any, Dict, Iterable, List, Mapping, Optional, Tuple
import numpy as np
import soundfile as sf
from flask import current_app, send_file
from flask.typing import ResponseReturnValue
SPLIT_PATTERN = r"\n+"
SAMPLE_RATE = 24000
_preview_pipelines: Dict[Tuple[str, str], Any] = {}
_preview_pipeline_lock = threading.Lock()
def clear_preview_pipelines() -> None:
"""Dispose all cached preview pipelines and clear the cache."""
with _preview_pipeline_lock:
for pipeline in _preview_pipelines.values():
try:
pipeline.dispose()
except Exception:
pass
_preview_pipelines.clear()
def _select_device() -> str:
import platform
try:
import torch # type: ignore[import-not-found]
except Exception:
return "cpu"
system = platform.system()
if system == "Darwin" and platform.processor() == "arm":
try:
if torch.backends.mps.is_available():
return "mps"
except Exception:
pass
return "cpu"
try:
if torch.cuda.is_available():
return "cuda"
except Exception:
pass
return "cpu"
def _resolve_pipeline(language: str, use_gpu: bool) -> Tuple[Any, bool]:
devices: List[str] = ["cpu"]
if use_gpu:
preferred = _select_device()
if preferred != "cpu":
devices.insert(0, preferred)
last_error: Optional[Exception] = None
for device in devices:
try:
return get_preview_pipeline(language, device), device != "cpu"
except Exception as exc:
last_error = exc
raise RuntimeError("Preview pipeline is unavailable") from last_error
def _to_float32(audio_segment) -> np.ndarray:
if audio_segment is None:
return np.zeros(0, dtype="float32")
tensor = audio_segment
if hasattr(tensor, "detach"):
tensor = tensor.detach()
if hasattr(tensor, "cpu"):
try:
tensor = tensor.cpu()
except Exception:
pass
if hasattr(tensor, "numpy"):
return np.asarray(tensor.numpy(), dtype="float32").reshape(-1)
return np.asarray(tensor, dtype="float32").reshape(-1)
def get_preview_pipeline(language: str, device: str) -> Any:
key = (language, device)
with _preview_pipeline_lock:
pipeline = _preview_pipelines.get(key)
if pipeline is not None:
return pipeline
from abogen.tts_plugin.utils import create_pipeline
pipeline = create_pipeline("kokoro", lang_code=language, device=device)
_preview_pipelines[key] = pipeline
return pipeline
def generate_preview_audio(
text: str,
voice_spec: str,
language: str,
speed: float,
use_gpu: bool,
tts_provider: str = "kokoro",
supertonic_total_steps: int = 5,
max_seconds: float = 8.0,
pronunciation_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
manual_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
speakers: Optional[Mapping[str, Any]] = None,
) -> bytes:
if not text.strip():
raise ValueError("Preview text is required")
provider = (tts_provider or "kokoro").strip().lower()
# Apply pronunciation/manual overrides first so tokens like `Unfu*k` still match
# before any downstream normalization potentially strips punctuation.
source_text = text
if pronunciation_overrides or manual_overrides or speakers:
try:
from abogen.webui import conversion_runner as runner
class _PreviewJob:
def __init__(self):
self.language = language
self.voice = voice_spec
self.speakers = speakers
self.manual_overrides = list(manual_overrides or [])
self.pronunciation_overrides = list(pronunciation_overrides or [])
job = _PreviewJob()
merged = runner._merge_pronunciation_overrides(job)
rules = runner._compile_pronunciation_rules(merged)
source_text = runner._apply_pronunciation_rules(source_text, rules)
except Exception:
current_app.logger.exception("Preview override application failed; using raw text")
source_text = text
normalized_text = source_text
if provider != "supertonic":
try:
from abogen.kokoro_text_normalization import normalize_for_pipeline
normalized_text = normalize_for_pipeline(source_text)
except Exception:
current_app.logger.exception("Preview normalization failed; using raw text")
normalized_text = source_text
if provider == "supertonic":
from abogen.tts_plugin.utils import create_pipeline
pipeline = create_pipeline("supertonic")
segments = pipeline(
normalized_text,
voice=voice_spec,
speed=speed,
split_pattern=SPLIT_PATTERN,
total_steps=supertonic_total_steps,
)
else:
pipeline, pipeline_uses_gpu = _resolve_pipeline(language, use_gpu)
if pipeline is None:
raise RuntimeError("Preview pipeline is unavailable")
voice_choice: Any = voice_spec
if voice_spec and "*" in voice_spec:
from abogen.voice_formulas import get_new_voice
voice_choice = get_new_voice(pipeline, voice_spec, pipeline_uses_gpu)
segments = pipeline(
normalized_text,
voice=voice_choice,
speed=speed,
split_pattern=SPLIT_PATTERN,
)
audio_chunks: List[np.ndarray] = []
accumulated = 0
max_samples = int(max(1.0, max_seconds) * SAMPLE_RATE)
for segment in segments:
graphemes = getattr(segment, "graphemes", "").strip()
if not graphemes:
continue
audio = _to_float32(getattr(segment, "audio", None))
if audio.size == 0:
continue
remaining = max_samples - accumulated
if remaining <= 0:
break
if audio.shape[0] > remaining:
audio = audio[:remaining]
audio_chunks.append(audio)
accumulated += audio.shape[0]
if accumulated >= max_samples:
break
if not audio_chunks:
raise RuntimeError("Preview could not be generated")
audio_data = np.concatenate(audio_chunks)
buffer = io.BytesIO()
sf.write(buffer, audio_data, SAMPLE_RATE, format="WAV")
return buffer.getvalue()
def synthesize_preview(
text: str,
voice_spec: str,
language: str,
speed: float,
use_gpu: bool,
tts_provider: str = "kokoro",
supertonic_total_steps: int = 5,
max_seconds: float = 8.0,
pronunciation_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
manual_overrides: Optional[Iterable[Mapping[str, Any]]] = None,
speakers: Optional[Mapping[str, Any]] = None,
) -> ResponseReturnValue:
try:
audio_bytes = generate_preview_audio(
text=text,
voice_spec=voice_spec,
language=language,
speed=speed,
use_gpu=use_gpu,
tts_provider=tts_provider,
supertonic_total_steps=supertonic_total_steps,
max_seconds=max_seconds,
pronunciation_overrides=pronunciation_overrides,
manual_overrides=manual_overrides,
speakers=speakers,
)
except Exception as e:
raise e
buffer = io.BytesIO(audio_bytes)
response = send_file(
buffer,
mimetype="audio/wav",
as_attachment=False,
download_name="speaker_preview.wav",
)
response.headers["Cache-Control"] = "no-store"
return response
abogen/webui/routes/utils/settings.py
+752 -752
View File
File diff suppressed because it is too large Load Diff
abogen/webui/routes/utils/voice.py
+808 -808
View File
File diff suppressed because it is too large Load Diff
plugins/kokoro/__init__.py
+177 -177
View File
@@ -1,178 +1,178 @@
"""Kokoro TTS Plugin for the TTS Plugin Architecture.
This plugin provides a Kokoro-based TTS engine that implements the
Plugin API contract. It wraps the existing Kokoro backend in the
new Engine/EngineSession architecture.
Exports:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Factory function
"""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
"""Kokoro TTS Plugin for the TTS Plugin Architecture.
This plugin provides a Kokoro-based TTS engine that implements the
Plugin API contract. It wraps the existing Kokoro backend in the
new Engine/EngineSession architecture.
Exports:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Factory function
"""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import EngineConfig
from .engine import KokoroEngine
def _load_kpipeline() -> Any:
"""Lazy-load Kokoro dependencies."""
from kokoro import KPipeline # type: ignore[import-not-found]
return KPipeline
PLUGIN_MANIFEST = PluginManifest(
id="kokoro",
name="Kokoro",
version="0.9.4",
api_version="1.0",
description="Kokoro TTS engine - high quality multilingual text-to-speech",
author="Kokoro Team",
capabilities=("voice_list",),
requires=RequirementManifest(
internet=False,
),
engine=EngineManifest(
voiceSources=(
VoiceSourceManifest(
id="builtin",
name="Built-in Voices",
type="list",
config={"voices": "See listVoices()"},
),
),
parameters=(
ParameterManifest(
id="speed",
name="Speed",
description="Speech speed multiplier",
type="float",
default=1.0,
min=0.5,
max=2.0,
step=0.1,
),
),
audioFormats=(
AudioFormatManifest(mime="audio/wav", extension="wav"),
),
),
voices=(
VoiceManifest(id="af_alloy", name="Alloy", tags=("en", "female")),
VoiceManifest(id="af_aoede", name="Aoede", tags=("en", "female")),
VoiceManifest(id="af_bella", name="Bella", tags=("en", "female")),
VoiceManifest(id="af_heart", name="Heart", tags=("en", "female")),
VoiceManifest(id="af_jessica", name="Jessica", tags=("en", "female")),
VoiceManifest(id="af_kore", name="Kore", tags=("en", "female")),
VoiceManifest(id="af_nicole", name="Nicole", tags=("en", "female")),
VoiceManifest(id="af_nova", name="Nova", tags=("en", "female")),
VoiceManifest(id="af_river", name="River", tags=("en", "female")),
VoiceManifest(id="af_sarah", name="Sarah", tags=("en", "female")),
VoiceManifest(id="af_sky", name="Sky", tags=("en", "female")),
VoiceManifest(id="am_adam", name="Adam", tags=("en", "male")),
VoiceManifest(id="am_echo", name="Echo", tags=("en", "male")),
VoiceManifest(id="am_eric", name="Eric", tags=("en", "male")),
VoiceManifest(id="am_fenrir", name="Fenrir", tags=("en", "male")),
VoiceManifest(id="am_liam", name="Liam", tags=("en", "male")),
VoiceManifest(id="am_michael", name="Michael", tags=("en", "male")),
VoiceManifest(id="am_onyx", name="Onyx", tags=("en", "male")),
VoiceManifest(id="am_puck", name="Puck", tags=("en", "male")),
VoiceManifest(id="am_santa", name="Santa", tags=("en", "male")),
VoiceManifest(id="bf_alice", name="Alice", tags=("en", "female")),
VoiceManifest(id="bf_emma", name="Emma", tags=("en", "female")),
VoiceManifest(id="bf_isabella", name="Isabella", tags=("en", "female")),
VoiceManifest(id="bf_lily", name="Lily", tags=("en", "female")),
VoiceManifest(id="bm_daniel", name="Daniel", tags=("en", "male")),
VoiceManifest(id="bm_fable", name="Fable", tags=("en", "male")),
VoiceManifest(id="bm_george", name="George", tags=("en", "male")),
VoiceManifest(id="bm_lewis", name="Lewis", tags=("en", "male")),
VoiceManifest(id="ef_dora", name="Dora", tags=("es", "female")),
VoiceManifest(id="em_alex", name="Alex", tags=("es", "male")),
VoiceManifest(id="em_santa", name="Santa", tags=("es", "male")),
VoiceManifest(id="ff_siwis", name="Siwis", tags=("fr", "female")),
VoiceManifest(id="hf_alpha", name="Alpha", tags=("hi", "female")),
VoiceManifest(id="hf_beta", name="Beta", tags=("hi", "female")),
VoiceManifest(id="hm_omega", name="Omega", tags=("hi", "male")),
VoiceManifest(id="hm_psi", name="Psi", tags=("hi", "male")),
VoiceManifest(id="if_sara", name="Sara", tags=("it", "female")),
VoiceManifest(id="im_nicola", name="Nicola", tags=("it", "male")),
VoiceManifest(id="jf_alpha", name="Alpha", tags=("ja", "female")),
VoiceManifest(id="jf_gongitsune", name="Gongitsune", tags=("ja", "female")),
VoiceManifest(id="jf_nezumi", name="Nezumi", tags=("ja", "female")),
VoiceManifest(id="jf_tebukuro", name="Tebukuro", tags=("ja", "female")),
VoiceManifest(id="jm_kumo", name="Kumo", tags=("ja", "male")),
VoiceManifest(id="pf_dora", name="Dora", tags=("pt", "female")),
VoiceManifest(id="pm_alex", name="Alex", tags=("pt", "male")),
VoiceManifest(id="pm_santa", name="Santa", tags=("pt", "male")),
VoiceManifest(id="zf_xiaobei", name="Xiaobei", tags=("zh", "female")),
VoiceManifest(id="zf_xiaoni", name="Xiaoni", tags=("zh", "female")),
VoiceManifest(id="zf_xiaoxiao", name="Xiaoxiao", tags=("zh", "female")),
VoiceManifest(id="zf_xiaoyi", name="Xiaoyi", tags=("zh", "female")),
VoiceManifest(id="zm_yunjian", name="Yunjian", tags=("zh", "male")),
VoiceManifest(id="zm_yunxi", name="Yunxi", tags=("zh", "male")),
VoiceManifest(id="zm_yunxia", name="Yunxia", tags=("zh", "female")),
VoiceManifest(id="zm_yunyang", name="Yunyang", tags=("zh", "male")),
),
)
MODEL_REQUIREMENTS: list[ModelManifest] = []
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create a Kokoro engine instance.
This function is the plugin entry point. It must be atomic:
succeed fully or raise EngineError and clean up.
Args:
context: Host services (config dir, logger, http client).
model_path: Resolved model path, or None for default.
config: Engine initialization settings (device, etc.).
Returns:
A fully initialized KokoroEngine instance.
Raises:
EngineError: On failure. Cleans up partially created resources.
"""
try:
KPipeline = _load_kpipeline()
# Determine repo_id from model_path or use default
repo_id = "hexgrad/Kokoro-82M"
if model_path is not None:
# If a specific model path is provided, use it as repo_id
repo_id = str(model_path)
pipeline = KPipeline(
lang_code=config.lang_code,
repo_id=repo_id,
device=config.device,
)
engine = KokoroEngine(pipeline)
return engine
except Exception as e:
from abogen.tts_plugin.errors import EngineError as EngineErrorClass
raise EngineErrorClass(f"Failed to create Kokoro engine: {e}") from e
from abogen.tts_plugin.types import EngineConfig
from .engine import KokoroEngine
def _load_kpipeline() -> Any:
"""Lazy-load Kokoro dependencies."""
from kokoro import KPipeline # type: ignore[import-not-found]
return KPipeline
PLUGIN_MANIFEST = PluginManifest(
id="kokoro",
name="Kokoro",
version="0.9.4",
api_version="1.0",
description="Kokoro TTS engine - high quality multilingual text-to-speech",
author="Kokoro Team",
capabilities=("voice_list",),
requires=RequirementManifest(
internet=False,
),
engine=EngineManifest(
voiceSources=(
VoiceSourceManifest(
id="builtin",
name="Built-in Voices",
type="list",
config={"voices": "See listVoices()"},
),
),
parameters=(
ParameterManifest(
id="speed",
name="Speed",
description="Speech speed multiplier",
type="float",
default=1.0,
min=0.5,
max=2.0,
step=0.1,
),
),
audioFormats=(
AudioFormatManifest(mime="audio/wav", extension="wav"),
),
),
voices=(
VoiceManifest(id="af_alloy", name="Alloy", tags=("en", "female")),
VoiceManifest(id="af_aoede", name="Aoede", tags=("en", "female")),
VoiceManifest(id="af_bella", name="Bella", tags=("en", "female")),
VoiceManifest(id="af_heart", name="Heart", tags=("en", "female")),
VoiceManifest(id="af_jessica", name="Jessica", tags=("en", "female")),
VoiceManifest(id="af_kore", name="Kore", tags=("en", "female")),
VoiceManifest(id="af_nicole", name="Nicole", tags=("en", "female")),
VoiceManifest(id="af_nova", name="Nova", tags=("en", "female")),
VoiceManifest(id="af_river", name="River", tags=("en", "female")),
VoiceManifest(id="af_sarah", name="Sarah", tags=("en", "female")),
VoiceManifest(id="af_sky", name="Sky", tags=("en", "female")),
VoiceManifest(id="am_adam", name="Adam", tags=("en", "male")),
VoiceManifest(id="am_echo", name="Echo", tags=("en", "male")),
VoiceManifest(id="am_eric", name="Eric", tags=("en", "male")),
VoiceManifest(id="am_fenrir", name="Fenrir", tags=("en", "male")),
VoiceManifest(id="am_liam", name="Liam", tags=("en", "male")),
VoiceManifest(id="am_michael", name="Michael", tags=("en", "male")),
VoiceManifest(id="am_onyx", name="Onyx", tags=("en", "male")),
VoiceManifest(id="am_puck", name="Puck", tags=("en", "male")),
VoiceManifest(id="am_santa", name="Santa", tags=("en", "male")),
VoiceManifest(id="bf_alice", name="Alice", tags=("en", "female")),
VoiceManifest(id="bf_emma", name="Emma", tags=("en", "female")),
VoiceManifest(id="bf_isabella", name="Isabella", tags=("en", "female")),
VoiceManifest(id="bf_lily", name="Lily", tags=("en", "female")),
VoiceManifest(id="bm_daniel", name="Daniel", tags=("en", "male")),
VoiceManifest(id="bm_fable", name="Fable", tags=("en", "male")),
VoiceManifest(id="bm_george", name="George", tags=("en", "male")),
VoiceManifest(id="bm_lewis", name="Lewis", tags=("en", "male")),
VoiceManifest(id="ef_dora", name="Dora", tags=("es", "female")),
VoiceManifest(id="em_alex", name="Alex", tags=("es", "male")),
VoiceManifest(id="em_santa", name="Santa", tags=("es", "male")),
VoiceManifest(id="ff_siwis", name="Siwis", tags=("fr", "female")),
VoiceManifest(id="hf_alpha", name="Alpha", tags=("hi", "female")),
VoiceManifest(id="hf_beta", name="Beta", tags=("hi", "female")),
VoiceManifest(id="hm_omega", name="Omega", tags=("hi", "male")),
VoiceManifest(id="hm_psi", name="Psi", tags=("hi", "male")),
VoiceManifest(id="if_sara", name="Sara", tags=("it", "female")),
VoiceManifest(id="im_nicola", name="Nicola", tags=("it", "male")),
VoiceManifest(id="jf_alpha", name="Alpha", tags=("ja", "female")),
VoiceManifest(id="jf_gongitsune", name="Gongitsune", tags=("ja", "female")),
VoiceManifest(id="jf_nezumi", name="Nezumi", tags=("ja", "female")),
VoiceManifest(id="jf_tebukuro", name="Tebukuro", tags=("ja", "female")),
VoiceManifest(id="jm_kumo", name="Kumo", tags=("ja", "male")),
VoiceManifest(id="pf_dora", name="Dora", tags=("pt", "female")),
VoiceManifest(id="pm_alex", name="Alex", tags=("pt", "male")),
VoiceManifest(id="pm_santa", name="Santa", tags=("pt", "male")),
VoiceManifest(id="zf_xiaobei", name="Xiaobei", tags=("zh", "female")),
VoiceManifest(id="zf_xiaoni", name="Xiaoni", tags=("zh", "female")),
VoiceManifest(id="zf_xiaoxiao", name="Xiaoxiao", tags=("zh", "female")),
VoiceManifest(id="zf_xiaoyi", name="Xiaoyi", tags=("zh", "female")),
VoiceManifest(id="zm_yunjian", name="Yunjian", tags=("zh", "male")),
VoiceManifest(id="zm_yunxi", name="Yunxi", tags=("zh", "male")),
VoiceManifest(id="zm_yunxia", name="Yunxia", tags=("zh", "female")),
VoiceManifest(id="zm_yunyang", name="Yunyang", tags=("zh", "male")),
),
)
MODEL_REQUIREMENTS: list[ModelManifest] = []
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create a Kokoro engine instance.
This function is the plugin entry point. It must be atomic:
succeed fully or raise EngineError and clean up.
Args:
context: Host services (config dir, logger, http client).
model_path: Resolved model path, or None for default.
config: Engine initialization settings (device, etc.).
Returns:
A fully initialized KokoroEngine instance.
Raises:
EngineError: On failure. Cleans up partially created resources.
"""
try:
KPipeline = _load_kpipeline()
# Determine repo_id from model_path or use default
repo_id = "hexgrad/Kokoro-82M"
if model_path is not None:
# If a specific model path is provided, use it as repo_id
repo_id = str(model_path)
pipeline = KPipeline(
lang_code=config.lang_code,
repo_id=repo_id,
device=config.device,
)
engine = KokoroEngine(pipeline)
return engine
except Exception as e:
from abogen.tts_plugin.errors import EngineError as EngineErrorClass
raise EngineErrorClass(f"Failed to create Kokoro engine: {e}") from e
plugins/kokoro/engine.py
+118 -118
View File
@@ -1,118 +1,118 @@
"""Kokoro Engine adapter for the TTS Plugin Architecture.
This module adapts the existing Kokoro backend to the new Engine/EngineSession
protocol. It wraps the KokoroBackend without modifying it.
"""
from __future__ import annotations
import logging
from typing import Any
import numpy as np
from abogen.tts_plugin.capabilities import VoiceLister
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
SynthesisRequest,
SynthesizedAudio,
)
logger = logging.getLogger(__name__)
# Sample rate for Kokoro audio
_KOKORO_SAMPLE_RATE = 24000
class KokoroSession:
"""EngineSession implementation for Kokoro.
Owns mutable execution state for synthesis.
NOT thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
"""Synthesize audio from text using Kokoro."""
if self._disposed:
raise EngineError("Session disposed")
try:
voice = request.voice.key
speed = request.parameters.values.get("speed", 1.0)
split_pattern = request.parameters.values.get("split_pattern", None)
audio_parts: list[np.ndarray] = []
for segment in self._pipeline(
request.text,
voice=voice,
speed=speed,
split_pattern=split_pattern,
):
audio = segment.audio
if hasattr(audio, "numpy"):
audio = audio.numpy()
audio_parts.append(np.asarray(audio, dtype="float32"))
if not audio_parts:
return SynthesizedAudio(
data=b"",
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.0),
)
combined = np.concatenate(audio_parts).astype("float32", copy=False)
audio_bytes = combined.tobytes()
duration_seconds = len(combined) / _KOKORO_SAMPLE_RATE
return SynthesizedAudio(
data=audio_bytes,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=duration_seconds),
)
except EngineError:
raise
except Exception as e:
raise EngineError(f"Synthesis failed: {e}") from e
def dispose(self) -> None:
"""Release session resources. Idempotent."""
self._disposed = True
class KokoroEngine:
"""Engine implementation for Kokoro.
Factory for KokoroSession instances. Stateless and thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def createSession(self) -> KokoroSession:
"""Create a new KokoroSession."""
if self._disposed:
raise EngineError("Engine disposed")
return KokoroSession(self._pipeline)
def dispose(self) -> None:
"""Release engine resources. Idempotent."""
self._disposed = True
def listVoices(self, sourceId: str) -> list[VoiceManifest]:
"""List available Kokoro voices. Implements VoiceLister capability.
Note: Static voices are declared in the plugin manifest.
This method is a fallback for dynamic plugins.
"""
if self._disposed:
raise EngineError("Engine disposed")
return []
"""Kokoro Engine adapter for the TTS Plugin Architecture.
This module adapts the existing Kokoro backend to the new Engine/EngineSession
protocol. It wraps the KokoroBackend without modifying it.
"""
from __future__ import annotations
import logging
from typing import Any
import numpy as np
from abogen.tts_plugin.capabilities import VoiceLister
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
SynthesisRequest,
SynthesizedAudio,
)
logger = logging.getLogger(__name__)
# Sample rate for Kokoro audio
_KOKORO_SAMPLE_RATE = 24000
class KokoroSession:
"""EngineSession implementation for Kokoro.
Owns mutable execution state for synthesis.
NOT thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
"""Synthesize audio from text using Kokoro."""
if self._disposed:
raise EngineError("Session disposed")
try:
voice = request.voice.key
speed = request.parameters.values.get("speed", 1.0)
split_pattern = request.parameters.values.get("split_pattern", None)
audio_parts: list[np.ndarray] = []
for segment in self._pipeline(
request.text,
voice=voice,
speed=speed,
split_pattern=split_pattern,
):
audio = segment.audio
if hasattr(audio, "numpy"):
audio = audio.numpy()
audio_parts.append(np.asarray(audio, dtype="float32"))
if not audio_parts:
return SynthesizedAudio(
data=b"",
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.0),
)
combined = np.concatenate(audio_parts).astype("float32", copy=False)
audio_bytes = combined.tobytes()
duration_seconds = len(combined) / _KOKORO_SAMPLE_RATE
return SynthesizedAudio(
data=audio_bytes,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=duration_seconds),
)
except EngineError:
raise
except Exception as e:
raise EngineError(f"Synthesis failed: {e}") from e
def dispose(self) -> None:
"""Release session resources. Idempotent."""
self._disposed = True
class KokoroEngine:
"""Engine implementation for Kokoro.
Factory for KokoroSession instances. Stateless and thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def createSession(self) -> KokoroSession:
"""Create a new KokoroSession."""
if self._disposed:
raise EngineError("Engine disposed")
return KokoroSession(self._pipeline)
def dispose(self) -> None:
"""Release engine resources. Idempotent."""
self._disposed = True
def listVoices(self, sourceId: str) -> list[VoiceManifest]:
"""List available Kokoro voices. Implements VoiceLister capability.
Note: Static voices are declared in the plugin manifest.
This method is a fallback for dynamic plugins.
"""
if self._disposed:
raise EngineError("Engine disposed")
return []
plugins/supertonic/__init__.py
+136 -136
View File
@@ -1,136 +1,136 @@
"""SuperTonic TTS Plugin for the TTS Plugin Architecture.
This plugin provides a SuperTonic-based TTS engine that implements the
Plugin API contract. It wraps the existing SuperTonic backend in the
new Engine/EngineSession architecture.
Exports:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Factory function
"""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import EngineConfig
from .engine import SuperTonicEngine
def _load_supertonic_pipeline() -> Any:
"""Lazy-load SuperTonic dependencies and create pipeline."""
from plugins.supertonic.pipeline import SupertonicPipeline
return SupertonicPipeline(
sample_rate=24000,
auto_download=True,
total_steps=5,
)
PLUGIN_MANIFEST = PluginManifest(
id="supertonic",
name="SuperTonic",
version="0.1.0",
api_version="1.0",
description="SuperTonic TTS engine - fast high-quality text-to-speech",
author="SuperTonic Team",
capabilities=("voice_list",),
requires=RequirementManifest(
internet=False,
),
engine=EngineManifest(
voiceSources=(
VoiceSourceManifest(
id="builtin",
name="Built-in Voices",
type="list",
config={"voices": "See listVoices()"},
),
),
parameters=(
ParameterManifest(
id="speed",
name="Speed",
description="Speech speed multiplier",
type="float",
default=1.0,
min=0.7,
max=2.0,
step=0.1,
),
ParameterManifest(
id="total_steps",
name="Quality Steps",
description="Inference steps (higher = better quality, slower)",
type="int",
default=5,
min=2,
max=15,
step=1,
),
),
audioFormats=(
AudioFormatManifest(mime="audio/wav", extension="wav"),
),
),
voices=(
VoiceManifest(id="M1", name="Male 1", tags=("male",)),
VoiceManifest(id="M2", name="Male 2", tags=("male",)),
VoiceManifest(id="M3", name="Male 3", tags=("male",)),
VoiceManifest(id="M4", name="Male 4", tags=("male",)),
VoiceManifest(id="M5", name="Male 5", tags=("male",)),
VoiceManifest(id="F1", name="Female 1", tags=("female",)),
VoiceManifest(id="F2", name="Female 2", tags=("female",)),
VoiceManifest(id="F3", name="Female 3", tags=("female",)),
VoiceManifest(id="F4", name="Female 4", tags=("female",)),
VoiceManifest(id="F5", name="Female 5", tags=("female",)),
),
)
MODEL_REQUIREMENTS: list[ModelManifest] = []
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create a SuperTonic engine instance.
This function is the plugin entry point. It must be atomic:
succeed fully or raise EngineError and clean up.
Args:
context: Host services (config dir, logger, http client).
model_path: Resolved model path, or None for default.
config: Engine initialization settings (device, etc.).
Returns:
A fully initialized SuperTonicEngine instance.
Raises:
EngineError: On failure. Cleans up partially created resources.
"""
try:
pipeline = _load_supertonic_pipeline()
engine = SuperTonicEngine(pipeline)
return engine
except Exception as e:
from abogen.tts_plugin.errors import EngineError as EngineErrorClass
raise EngineErrorClass(f"Failed to create SuperTonic engine: {e}") from e
"""SuperTonic TTS Plugin for the TTS Plugin Architecture.
This plugin provides a SuperTonic-based TTS engine that implements the
Plugin API contract. It wraps the existing SuperTonic backend in the
new Engine/EngineSession architecture.
Exports:
- PLUGIN_MANIFEST: PluginManifest
- MODEL_REQUIREMENTS: list[ModelManifest]
- create_engine: Factory function
"""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import EngineConfig
from .engine import SuperTonicEngine
def _load_supertonic_pipeline() -> Any:
"""Lazy-load SuperTonic dependencies and create pipeline."""
from plugins.supertonic.pipeline import SupertonicPipeline
return SupertonicPipeline(
sample_rate=24000,
auto_download=True,
total_steps=5,
)
PLUGIN_MANIFEST = PluginManifest(
id="supertonic",
name="SuperTonic",
version="0.1.0",
api_version="1.0",
description="SuperTonic TTS engine - fast high-quality text-to-speech",
author="SuperTonic Team",
capabilities=("voice_list",),
requires=RequirementManifest(
internet=False,
),
engine=EngineManifest(
voiceSources=(
VoiceSourceManifest(
id="builtin",
name="Built-in Voices",
type="list",
config={"voices": "See listVoices()"},
),
),
parameters=(
ParameterManifest(
id="speed",
name="Speed",
description="Speech speed multiplier",
type="float",
default=1.0,
min=0.7,
max=2.0,
step=0.1,
),
ParameterManifest(
id="total_steps",
name="Quality Steps",
description="Inference steps (higher = better quality, slower)",
type="int",
default=5,
min=2,
max=15,
step=1,
),
),
audioFormats=(
AudioFormatManifest(mime="audio/wav", extension="wav"),
),
),
voices=(
VoiceManifest(id="M1", name="Male 1", tags=("male",)),
VoiceManifest(id="M2", name="Male 2", tags=("male",)),
VoiceManifest(id="M3", name="Male 3", tags=("male",)),
VoiceManifest(id="M4", name="Male 4", tags=("male",)),
VoiceManifest(id="M5", name="Male 5", tags=("male",)),
VoiceManifest(id="F1", name="Female 1", tags=("female",)),
VoiceManifest(id="F2", name="Female 2", tags=("female",)),
VoiceManifest(id="F3", name="Female 3", tags=("female",)),
VoiceManifest(id="F4", name="Female 4", tags=("female",)),
VoiceManifest(id="F5", name="Female 5", tags=("female",)),
),
)
MODEL_REQUIREMENTS: list[ModelManifest] = []
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create a SuperTonic engine instance.
This function is the plugin entry point. It must be atomic:
succeed fully or raise EngineError and clean up.
Args:
context: Host services (config dir, logger, http client).
model_path: Resolved model path, or None for default.
config: Engine initialization settings (device, etc.).
Returns:
A fully initialized SuperTonicEngine instance.
Raises:
EngineError: On failure. Cleans up partially created resources.
"""
try:
pipeline = _load_supertonic_pipeline()
engine = SuperTonicEngine(pipeline)
return engine
except Exception as e:
from abogen.tts_plugin.errors import EngineError as EngineErrorClass
raise EngineErrorClass(f"Failed to create SuperTonic engine: {e}") from e
plugins/supertonic/engine.py
+125 -125
View File
@@ -1,125 +1,125 @@
"""SuperTonic Engine adapter for the TTS Plugin Architecture.
This module adapts the existing SuperTonic backend to the new Engine/EngineSession
protocol. It wraps the SupertonicPipeline without modifying it.
"""
from __future__ import annotations
import io
import logging
from typing import Any
import numpy as np
from abogen.tts_plugin.capabilities import VoiceLister
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
SynthesisRequest,
SynthesizedAudio,
)
logger = logging.getLogger(__name__)
# Sample rate for SuperTonic audio
_SUPERTONIC_SAMPLE_RATE = 24000
class SuperTonicSession:
"""EngineSession implementation for SuperTonic.
Owns mutable execution state for synthesis.
NOT thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
"""Synthesize audio from text using SuperTonic."""
if self._disposed:
raise EngineError("Session disposed")
try:
import soundfile as sf
voice = request.voice.key
speed = float(request.parameters.values.get("speed", 1.0))
total_steps = request.parameters.values.get("total_steps", None)
split_pattern = request.parameters.values.get("split_pattern", None)
if total_steps is not None:
total_steps = int(total_steps)
audio_parts: list[np.ndarray] = []
for segment in self._pipeline(
request.text,
voice=voice,
speed=speed,
split_pattern=split_pattern,
total_steps=total_steps,
):
audio_parts.append(segment.audio)
if not audio_parts:
return SynthesizedAudio(
data=b"",
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.0),
)
combined = np.concatenate(audio_parts).astype("float32", copy=False)
buf = io.BytesIO()
sf.write(buf, combined, self._pipeline.sample_rate, format="WAV")
audio_bytes = buf.getvalue()
duration_seconds = len(combined) / self._pipeline.sample_rate
return SynthesizedAudio(
data=audio_bytes,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=duration_seconds),
)
except EngineError:
raise
except Exception as e:
raise EngineError(f"Synthesis failed: {e}") from e
def dispose(self) -> None:
"""Release session resources. Idempotent."""
self._disposed = True
class SuperTonicEngine:
"""Engine implementation for SuperTonic.
Factory for SuperTonicSession instances. Stateless and thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def createSession(self) -> SuperTonicSession:
"""Create a new SuperTonicSession."""
if self._disposed:
raise EngineError("Engine disposed")
return SuperTonicSession(self._pipeline)
def dispose(self) -> None:
"""Release engine resources. Idempotent."""
self._disposed = True
def listVoices(self, sourceId: str) -> list[VoiceManifest]:
"""List available SuperTonic voices. Implements VoiceLister capability.
Note: Static voice catalog is declared in plugin manifest.
This method is retained for VoiceLister interface compliance.
"""
if self._disposed:
raise EngineError("Engine disposed")
return []
"""SuperTonic Engine adapter for the TTS Plugin Architecture.
This module adapts the existing SuperTonic backend to the new Engine/EngineSession
protocol. It wraps the SupertonicPipeline without modifying it.
"""
from __future__ import annotations
import io
import logging
from typing import Any
import numpy as np
from abogen.tts_plugin.capabilities import VoiceLister
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
SynthesisRequest,
SynthesizedAudio,
)
logger = logging.getLogger(__name__)
# Sample rate for SuperTonic audio
_SUPERTONIC_SAMPLE_RATE = 24000
class SuperTonicSession:
"""EngineSession implementation for SuperTonic.
Owns mutable execution state for synthesis.
NOT thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
"""Synthesize audio from text using SuperTonic."""
if self._disposed:
raise EngineError("Session disposed")
try:
import soundfile as sf
voice = request.voice.key
speed = float(request.parameters.values.get("speed", 1.0))
total_steps = request.parameters.values.get("total_steps", None)
split_pattern = request.parameters.values.get("split_pattern", None)
if total_steps is not None:
total_steps = int(total_steps)
audio_parts: list[np.ndarray] = []
for segment in self._pipeline(
request.text,
voice=voice,
speed=speed,
split_pattern=split_pattern,
total_steps=total_steps,
):
audio_parts.append(segment.audio)
if not audio_parts:
return SynthesizedAudio(
data=b"",
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.0),
)
combined = np.concatenate(audio_parts).astype("float32", copy=False)
buf = io.BytesIO()
sf.write(buf, combined, self._pipeline.sample_rate, format="WAV")
audio_bytes = buf.getvalue()
duration_seconds = len(combined) / self._pipeline.sample_rate
return SynthesizedAudio(
data=audio_bytes,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=duration_seconds),
)
except EngineError:
raise
except Exception as e:
raise EngineError(f"Synthesis failed: {e}") from e
def dispose(self) -> None:
"""Release session resources. Idempotent."""
self._disposed = True
class SuperTonicEngine:
"""Engine implementation for SuperTonic.
Factory for SuperTonicSession instances. Stateless and thread-safe.
"""
def __init__(self, pipeline: Any) -> None:
self._pipeline = pipeline
self._disposed = False
def createSession(self) -> SuperTonicSession:
"""Create a new SuperTonicSession."""
if self._disposed:
raise EngineError("Engine disposed")
return SuperTonicSession(self._pipeline)
def dispose(self) -> None:
"""Release engine resources. Idempotent."""
self._disposed = True
def listVoices(self, sourceId: str) -> list[VoiceManifest]:
"""List available SuperTonic voices. Implements VoiceLister capability.
Note: Static voice catalog is declared in plugin manifest.
This method is retained for VoiceLister interface compliance.
"""
if self._disposed:
raise EngineError("Engine disposed")
return []
plugins/supertonic/pipeline.py
+266 -266
View File
@@ -1,266 +1,266 @@
"""SuperTonic Pipeline — self-contained TTS pipeline for the plugin.
This module provides the SuperTonicPipeline class and supporting utilities
used by the SuperTonic plugin. It is independent of the legacy
abogen.tts_backends module.
"""
from __future__ import annotations
import ast
import logging
import re
from typing import Any, Iterable, Iterator, Optional
import numpy as np
logger = logging.getLogger(__name__)
def _ensure_float32_mono(wav: Any) -> np.ndarray:
arr = np.asarray(wav, dtype="float32")
if arr.ndim == 2:
if arr.shape[0] == 1 and arr.shape[1] > 1:
arr = arr.reshape(-1)
else:
arr = arr[:, 0]
return arr.reshape(-1)
def _resample_linear(audio: np.ndarray, src_rate: int, dst_rate: int) -> np.ndarray:
if src_rate == dst_rate:
return audio
if audio.size == 0:
return audio
ratio = dst_rate / float(src_rate)
new_len = int(round(audio.size * ratio))
if new_len <= 1:
return np.zeros(0, dtype="float32")
x_old = np.linspace(0.0, 1.0, num=audio.size, endpoint=False)
x_new = np.linspace(0.0, 1.0, num=new_len, endpoint=False)
return np.interp(x_new, x_old, audio).astype("float32", copy=False)
def _split_text(
text: str, *, split_pattern: Optional[str], max_chunk_length: int
) -> list[str]:
stripped = (text or "").strip()
if not stripped:
return []
parts: list[str]
if split_pattern:
try:
parts = [p.strip() for p in re.split(split_pattern, stripped) if p.strip()]
except re.error:
parts = [stripped]
else:
parts = [stripped]
result: list[str] = []
for part in parts:
if len(part) <= max_chunk_length:
result.append(part)
continue
start = 0
while start < len(part):
end = min(len(part), start + max_chunk_length)
if end < len(part):
ws = part.rfind(" ", start, end)
if ws > start + 40:
end = ws
chunk = part[start:end].strip()
if chunk:
result.append(chunk)
start = end
return result
_UNSUPPORTED_CHARS_RE = re.compile(
r"unsupported character\(s\):\s*(\[[^\]]*\])", re.IGNORECASE
)
def _parse_unsupported_characters(error: BaseException) -> list[str]:
"""Best-effort extraction of unsupported characters from SuperTonic errors."""
message = " ".join(
str(part) for part in getattr(error, "args", ()) if part is not None
) or str(error)
match = _UNSUPPORTED_CHARS_RE.search(message)
if not match:
return []
raw = match.group(1)
try:
value = ast.literal_eval(raw)
except Exception:
return []
if isinstance(value, (list, tuple)):
out: list[str] = []
for item in value:
if item is None:
continue
s = str(item)
if s:
out.append(s)
return out
if isinstance(value, str) and value:
return [value]
return []
def _remove_unsupported_characters(text: str, unsupported: Iterable[str]) -> str:
result = text
for item in unsupported:
if not item:
continue
result = result.replace(item, "")
return result
def _configure_supertonic_gpu() -> None:
"""Patch supertonic's config to enable GPU acceleration if available."""
try:
import onnxruntime as ort
available = ort.get_available_providers()
providers = []
if "CUDAExecutionProvider" in available:
providers.append("CUDAExecutionProvider")
providers.append("CPUExecutionProvider")
import supertonic.config as supertonic_config
import supertonic.loader as supertonic_loader
supertonic_config.DEFAULT_ONNX_PROVIDERS = providers
supertonic_loader.DEFAULT_ONNX_PROVIDERS = providers
logger.info("Supertonic ONNX providers configured: %s", providers)
except Exception as exc:
logger.warning("Could not configure supertonic GPU providers: %s", exc)
class SupertonicSegment:
"""A single synthesized audio segment."""
__slots__ = ("graphemes", "audio")
def __init__(self, graphemes: str, audio: np.ndarray) -> None:
self.graphemes = graphemes
self.audio = audio
class SupertonicPipeline:
"""Minimal adapter that mimics Kokoro's pipeline iteration interface."""
def __init__(
self,
*,
sample_rate: int,
auto_download: bool = True,
total_steps: int = 5,
max_chunk_length: int = 300,
) -> None:
self.sample_rate = int(sample_rate)
self.total_steps = int(total_steps)
self.max_chunk_length = int(max_chunk_length)
_configure_supertonic_gpu()
try:
from supertonic import TTS # type: ignore[import-not-found]
except Exception as exc: # pragma: no cover
raise RuntimeError(
"Supertonic is not installed. Install it with `pip install supertonic`."
) from exc
self._tts = TTS(auto_download=auto_download)
def __call__(
self,
text: str,
*,
voice: str,
speed: float,
split_pattern: Optional[str] = None,
total_steps: Optional[int] = None,
) -> Iterator[SupertonicSegment]:
voice_name = (voice or "").strip() or "M1"
steps = int(total_steps) if total_steps is not None else self.total_steps
steps = max(2, min(15, steps))
speed_value = float(speed) if speed is not None else 1.0
speed_value = max(0.7, min(2.0, speed_value))
style = self._tts.get_voice_style(voice_name=voice_name)
chunks = _split_text(
text, split_pattern=split_pattern, max_chunk_length=self.max_chunk_length
)
for chunk in chunks:
chunk_to_speak = chunk
removed: set[str] = set()
last_exc: Exception | None = None
for attempt in range(3):
try:
wav, duration = self._tts.synthesize(
text=chunk_to_speak,
voice_style=style,
total_steps=steps,
speed=speed_value,
max_chunk_length=self.max_chunk_length,
silence_duration=0.0,
verbose=False,
)
break
except ValueError as exc:
last_exc = exc
unsupported = _parse_unsupported_characters(exc)
if not unsupported:
raise
removed.update(unsupported)
sanitized = _remove_unsupported_characters(
chunk_to_speak, unsupported
).strip()
if sanitized == chunk_to_speak.strip():
raise
chunk_to_speak = sanitized
if not chunk_to_speak:
logger.warning(
"SuperTonic: dropped a chunk after removing unsupported characters: %s",
sorted(removed),
)
break
if attempt == 0:
logger.warning(
"SuperTonic: removed unsupported characters %s and retried.",
sorted(removed),
)
else:
assert last_exc is not None
raise last_exc
if not chunk_to_speak:
continue
audio = _ensure_float32_mono(wav)
src_rate = self.sample_rate
try:
dur = float(duration)
if dur > 0 and audio.size > 0:
inferred = int(round(audio.size / dur))
if 8000 <= inferred <= 96000:
src_rate = inferred
except Exception:
pass
if src_rate != self.sample_rate:
audio = _resample_linear(audio, src_rate, self.sample_rate)
yield SupertonicSegment(graphemes=chunk_to_speak, audio=audio)
"""SuperTonic Pipeline — self-contained TTS pipeline for the plugin.
This module provides the SuperTonicPipeline class and supporting utilities
used by the SuperTonic plugin. It is independent of the legacy
abogen.tts_backends module.
"""
from __future__ import annotations
import ast
import logging
import re
from typing import Any, Iterable, Iterator, Optional
import numpy as np
logger = logging.getLogger(__name__)
def _ensure_float32_mono(wav: Any) -> np.ndarray:
arr = np.asarray(wav, dtype="float32")
if arr.ndim == 2:
if arr.shape[0] == 1 and arr.shape[1] > 1:
arr = arr.reshape(-1)
else:
arr = arr[:, 0]
return arr.reshape(-1)
def _resample_linear(audio: np.ndarray, src_rate: int, dst_rate: int) -> np.ndarray:
if src_rate == dst_rate:
return audio
if audio.size == 0:
return audio
ratio = dst_rate / float(src_rate)
new_len = int(round(audio.size * ratio))
if new_len <= 1:
return np.zeros(0, dtype="float32")
x_old = np.linspace(0.0, 1.0, num=audio.size, endpoint=False)
x_new = np.linspace(0.0, 1.0, num=new_len, endpoint=False)
return np.interp(x_new, x_old, audio).astype("float32", copy=False)
def _split_text(
text: str, *, split_pattern: Optional[str], max_chunk_length: int
) -> list[str]:
stripped = (text or "").strip()
if not stripped:
return []
parts: list[str]
if split_pattern:
try:
parts = [p.strip() for p in re.split(split_pattern, stripped) if p.strip()]
except re.error:
parts = [stripped]
else:
parts = [stripped]
result: list[str] = []
for part in parts:
if len(part) <= max_chunk_length:
result.append(part)
continue
start = 0
while start < len(part):
end = min(len(part), start + max_chunk_length)
if end < len(part):
ws = part.rfind(" ", start, end)
if ws > start + 40:
end = ws
chunk = part[start:end].strip()
if chunk:
result.append(chunk)
start = end
return result
_UNSUPPORTED_CHARS_RE = re.compile(
r"unsupported character\(s\):\s*(\[[^\]]*\])", re.IGNORECASE
)
def _parse_unsupported_characters(error: BaseException) -> list[str]:
"""Best-effort extraction of unsupported characters from SuperTonic errors."""
message = " ".join(
str(part) for part in getattr(error, "args", ()) if part is not None
) or str(error)
match = _UNSUPPORTED_CHARS_RE.search(message)
if not match:
return []
raw = match.group(1)
try:
value = ast.literal_eval(raw)
except Exception:
return []
if isinstance(value, (list, tuple)):
out: list[str] = []
for item in value:
if item is None:
continue
s = str(item)
if s:
out.append(s)
return out
if isinstance(value, str) and value:
return [value]
return []
def _remove_unsupported_characters(text: str, unsupported: Iterable[str]) -> str:
result = text
for item in unsupported:
if not item:
continue
result = result.replace(item, "")
return result
def _configure_supertonic_gpu() -> None:
"""Patch supertonic's config to enable GPU acceleration if available."""
try:
import onnxruntime as ort
available = ort.get_available_providers()
providers = []
if "CUDAExecutionProvider" in available:
providers.append("CUDAExecutionProvider")
providers.append("CPUExecutionProvider")
import supertonic.config as supertonic_config
import supertonic.loader as supertonic_loader
supertonic_config.DEFAULT_ONNX_PROVIDERS = providers
supertonic_loader.DEFAULT_ONNX_PROVIDERS = providers
logger.info("Supertonic ONNX providers configured: %s", providers)
except Exception as exc:
logger.warning("Could not configure supertonic GPU providers: %s", exc)
class SupertonicSegment:
"""A single synthesized audio segment."""
__slots__ = ("graphemes", "audio")
def __init__(self, graphemes: str, audio: np.ndarray) -> None:
self.graphemes = graphemes
self.audio = audio
class SupertonicPipeline:
"""Minimal adapter that mimics Kokoro's pipeline iteration interface."""
def __init__(
self,
*,
sample_rate: int,
auto_download: bool = True,
total_steps: int = 5,
max_chunk_length: int = 300,
) -> None:
self.sample_rate = int(sample_rate)
self.total_steps = int(total_steps)
self.max_chunk_length = int(max_chunk_length)
_configure_supertonic_gpu()
try:
from supertonic import TTS # type: ignore[import-not-found]
except Exception as exc: # pragma: no cover
raise RuntimeError(
"Supertonic is not installed. Install it with `pip install supertonic`."
) from exc
self._tts = TTS(auto_download=auto_download)
def __call__(
self,
text: str,
*,
voice: str,
speed: float,
split_pattern: Optional[str] = None,
total_steps: Optional[int] = None,
) -> Iterator[SupertonicSegment]:
voice_name = (voice or "").strip() or "M1"
steps = int(total_steps) if total_steps is not None else self.total_steps
steps = max(2, min(15, steps))
speed_value = float(speed) if speed is not None else 1.0
speed_value = max(0.7, min(2.0, speed_value))
style = self._tts.get_voice_style(voice_name=voice_name)
chunks = _split_text(
text, split_pattern=split_pattern, max_chunk_length=self.max_chunk_length
)
for chunk in chunks:
chunk_to_speak = chunk
removed: set[str] = set()
last_exc: Exception | None = None
for attempt in range(3):
try:
wav, duration = self._tts.synthesize(
text=chunk_to_speak,
voice_style=style,
total_steps=steps,
speed=speed_value,
max_chunk_length=self.max_chunk_length,
silence_duration=0.0,
verbose=False,
)
break
except ValueError as exc:
last_exc = exc
unsupported = _parse_unsupported_characters(exc)
if not unsupported:
raise
removed.update(unsupported)
sanitized = _remove_unsupported_characters(
chunk_to_speak, unsupported
).strip()
if sanitized == chunk_to_speak.strip():
raise
chunk_to_speak = sanitized
if not chunk_to_speak:
logger.warning(
"SuperTonic: dropped a chunk after removing unsupported characters: %s",
sorted(removed),
)
break
if attempt == 0:
logger.warning(
"SuperTonic: removed unsupported characters %s and retried.",
sorted(removed),
)
else:
assert last_exc is not None
raise last_exc
if not chunk_to_speak:
continue
audio = _ensure_float32_mono(wav)
src_rate = self.sample_rate
try:
dur = float(duration)
if dur > 0 and audio.size > 0:
inferred = int(round(audio.size / dur))
if 8000 <= inferred <= 96000:
src_rate = inferred
except Exception:
pass
if src_rate != self.sample_rate:
audio = _resample_linear(audio, src_rate, self.sample_rate)
yield SupertonicSegment(graphemes=chunk_to_speak, audio=audio)
tests/contracts/__init__.py
+5 -5
View File
@@ -1,5 +1,5 @@
"""Contract tests for the TTS Plugin API.
This package contains reusable contract tests that any TTS plugin implementation
must satisfy. Tests use only the public API and are engine-agnostic.
"""
"""Contract tests for the TTS Plugin API.
This package contains reusable contract tests that any TTS plugin implementation
must satisfy. Tests use only the public API and are engine-agnostic.
"""
tests/contracts/conftest.py
+231 -231
View File
@@ -1,231 +1,231 @@
"""Shared fixtures and stubs for contract tests.
This module provides minimal stub implementations that satisfy the public API
for testing purposes. These stubs do NOT contain real business logic.
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Iterator
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class FakeHttpClient:
"""Stub HTTP client that satisfies the HttpClient protocol."""
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
class FakeEngineSession:
"""Stub EngineSession for testing protocol compliance."""
def __init__(self) -> None:
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def dispose(self) -> None:
self._disposed = True
class FakeStreamingSession:
"""Stub EngineSession with StreamingSynthesizer capability."""
def __init__(self) -> None:
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def synthesizeStream(self, request: SynthesisRequest) -> Iterator[bytes]:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
for i in range(3):
yield b"\x00" * 50
def dispose(self) -> None:
self._disposed = True
class FakeCancelableSession:
"""Stub EngineSession with CancelableSession capability."""
def __init__(self) -> None:
self._disposed = False
self._cancelled = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
if self._cancelled:
from abogen.tts_plugin.errors import CancelledError
raise CancelledError("Cancelled")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def cancel(self) -> None:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
self._cancelled = True
def dispose(self) -> None:
self._disposed = True
class FakeEngine:
"""Stub Engine for testing protocol compliance."""
def __init__(self, session_class: type = FakeEngineSession) -> None:
self._disposed = False
self._session_class = session_class
def createSession(self) -> EngineSession:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return self._session_class()
def dispose(self) -> None:
self._disposed = True
class FakeVoiceListerEngine:
"""Stub Engine that also implements VoiceLister."""
def __init__(self) -> None:
self._disposed = False
def createSession(self) -> EngineSession:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return FakeEngineSession()
def listVoices(self, sourceId: str) -> list:
from abogen.tts_plugin.manifest import VoiceManifest
return [
VoiceManifest(id="voice1", name="Voice 1", tags=("en",)),
VoiceManifest(id="voice2", name="Voice 2", tags=("es",)),
]
def dispose(self) -> None:
self._disposed = True
class FakePreviewEngine:
"""Stub Engine that also implements PreviewGenerator."""
def __init__(self) -> None:
self._disposed = False
def createSession(self) -> EngineSession:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return FakeEngineSession()
def generatePreview(self, voice: VoiceSelection, text: str) -> SynthesizedAudio:
return SynthesizedAudio(
data=b"\x00" * 50,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.5),
)
def dispose(self) -> None:
self._disposed = True
@pytest.fixture
def fake_http_client() -> FakeHttpClient:
return FakeHttpClient()
@pytest.fixture
def host_context(tmp_path: Path, fake_http_client: FakeHttpClient) -> HostContext:
return HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=fake_http_client,
)
@pytest.fixture
def fake_engine() -> FakeEngine:
return FakeEngine()
@pytest.fixture
def fake_session() -> FakeEngineSession:
return FakeEngineSession()
@pytest.fixture
def default_voice() -> VoiceSelection:
return VoiceSelection(source="builtin", key="af_nova")
@pytest.fixture
def default_format() -> AudioFormat:
return AudioFormat(mime="audio/wav", extension="wav")
@pytest.fixture
def default_request(
default_voice: VoiceSelection, default_format: AudioFormat
) -> SynthesisRequest:
return SynthesisRequest(
text="Hello, world!",
voice=default_voice,
parameters=ParameterValues(values={}),
format=default_format,
)
"""Shared fixtures and stubs for contract tests.
This module provides minimal stub implementations that satisfy the public API
for testing purposes. These stubs do NOT contain real business logic.
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Iterator
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class FakeHttpClient:
"""Stub HTTP client that satisfies the HttpClient protocol."""
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
class FakeEngineSession:
"""Stub EngineSession for testing protocol compliance."""
def __init__(self) -> None:
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def dispose(self) -> None:
self._disposed = True
class FakeStreamingSession:
"""Stub EngineSession with StreamingSynthesizer capability."""
def __init__(self) -> None:
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def synthesizeStream(self, request: SynthesisRequest) -> Iterator[bytes]:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
for i in range(3):
yield b"\x00" * 50
def dispose(self) -> None:
self._disposed = True
class FakeCancelableSession:
"""Stub EngineSession with CancelableSession capability."""
def __init__(self) -> None:
self._disposed = False
self._cancelled = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
if self._cancelled:
from abogen.tts_plugin.errors import CancelledError
raise CancelledError("Cancelled")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def cancel(self) -> None:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Session disposed")
self._cancelled = True
def dispose(self) -> None:
self._disposed = True
class FakeEngine:
"""Stub Engine for testing protocol compliance."""
def __init__(self, session_class: type = FakeEngineSession) -> None:
self._disposed = False
self._session_class = session_class
def createSession(self) -> EngineSession:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return self._session_class()
def dispose(self) -> None:
self._disposed = True
class FakeVoiceListerEngine:
"""Stub Engine that also implements VoiceLister."""
def __init__(self) -> None:
self._disposed = False
def createSession(self) -> EngineSession:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return FakeEngineSession()
def listVoices(self, sourceId: str) -> list:
from abogen.tts_plugin.manifest import VoiceManifest
return [
VoiceManifest(id="voice1", name="Voice 1", tags=("en",)),
VoiceManifest(id="voice2", name="Voice 2", tags=("es",)),
]
def dispose(self) -> None:
self._disposed = True
class FakePreviewEngine:
"""Stub Engine that also implements PreviewGenerator."""
def __init__(self) -> None:
self._disposed = False
def createSession(self) -> EngineSession:
if self._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return FakeEngineSession()
def generatePreview(self, voice: VoiceSelection, text: str) -> SynthesizedAudio:
return SynthesizedAudio(
data=b"\x00" * 50,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.5),
)
def dispose(self) -> None:
self._disposed = True
@pytest.fixture
def fake_http_client() -> FakeHttpClient:
return FakeHttpClient()
@pytest.fixture
def host_context(tmp_path: Path, fake_http_client: FakeHttpClient) -> HostContext:
return HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=fake_http_client,
)
@pytest.fixture
def fake_engine() -> FakeEngine:
return FakeEngine()
@pytest.fixture
def fake_session() -> FakeEngineSession:
return FakeEngineSession()
@pytest.fixture
def default_voice() -> VoiceSelection:
return VoiceSelection(source="builtin", key="af_nova")
@pytest.fixture
def default_format() -> AudioFormat:
return AudioFormat(mime="audio/wav", extension="wav")
@pytest.fixture
def default_request(
default_voice: VoiceSelection, default_format: AudioFormat
) -> SynthesisRequest:
return SynthesisRequest(
text="Hello, world!",
voice=default_voice,
parameters=ParameterValues(values={}),
format=default_format,
)
tests/contracts/engine_contract.py
+120 -120
View File
@@ -1,120 +1,120 @@
"""Base contract tests for Engine implementations.
Any new TTS plugin must inherit from these classes to verify
it satisfies the Engine/EngineSession protocol.
Usage:
from tests.contracts.engine_contract import EngineContractMixin
class TestMyEngine(EngineContractMixin):
@pytest.fixture
def engine(self):
return create_my_engine()
"""
from __future__ import annotations
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.types import (
AudioFormat,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class EngineContractMixin:
"""Base contract tests for Engine implementations.
Subclasses must define a module-level ``engine`` fixture returning
a fully initialized Engine instance. The tests below will use it
via pytest's standard fixture resolution.
"""
def _req(self, text: str = "Hello", voice: str | None = None) -> SynthesisRequest:
return SynthesisRequest(
text=text,
voice=VoiceSelection(source="builtin", key=voice or "default"),
parameters=ParameterValues(values={}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# ── Engine protocol ──────────────────────────────────────
def test_engine_satisfies_protocol(self, engine: Engine) -> None:
assert isinstance(engine, Engine)
def test_create_session_returns_session(self, engine: Engine) -> None:
session = engine.createSession()
assert isinstance(session, EngineSession)
session.dispose()
def test_create_session_returns_new_instances(self, engine: Engine) -> None:
s1 = engine.createSession()
s2 = engine.createSession()
assert s1 is not s2
s1.dispose()
s2.dispose()
def test_dispose_is_idempotent(self, engine: Engine) -> None:
engine.dispose()
engine.dispose()
def test_create_session_after_dispose_raises(self, engine: Engine) -> None:
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
# ── Session protocol ─────────────────────────────────────
def test_session_satisfies_protocol(self, engine: Engine) -> None:
session = engine.createSession()
assert isinstance(session, EngineSession)
session.dispose()
engine.dispose()
def test_session_synthesize_returns_audio(self, engine: Engine) -> None:
session = engine.createSession()
result = session.synthesize(self._req())
assert isinstance(result, SynthesizedAudio)
assert isinstance(result.data, bytes)
assert len(result.data) > 0
session.dispose()
engine.dispose()
def test_session_dispose_is_idempotent(self, engine: Engine) -> None:
session = engine.createSession()
session.dispose()
session.dispose()
engine.dispose()
def test_session_synthesize_after_dispose_raises(self, engine: Engine) -> None:
session = engine.createSession()
session.dispose()
with pytest.raises(EngineError):
session.synthesize(self._req())
engine.dispose()
def test_session_multiple_synthesize(self, engine: Engine) -> None:
session = engine.createSession()
r1 = session.synthesize(self._req())
r2 = session.synthesize(self._req())
assert isinstance(r1.data, bytes)
assert isinstance(r2.data, bytes)
session.dispose()
engine.dispose()
# ── Lifecycle ────────────────────────────────────────────
def test_full_lifecycle(self, engine: Engine) -> None:
s1 = engine.createSession()
s2 = engine.createSession()
s1.synthesize(self._req())
s2.synthesize(self._req())
s1.dispose()
s2.dispose()
engine.dispose()
"""Base contract tests for Engine implementations.
Any new TTS plugin must inherit from these classes to verify
it satisfies the Engine/EngineSession protocol.
Usage:
from tests.contracts.engine_contract import EngineContractMixin
class TestMyEngine(EngineContractMixin):
@pytest.fixture
def engine(self):
return create_my_engine()
"""
from __future__ import annotations
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.types import (
AudioFormat,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class EngineContractMixin:
"""Base contract tests for Engine implementations.
Subclasses must define a module-level ``engine`` fixture returning
a fully initialized Engine instance. The tests below will use it
via pytest's standard fixture resolution.
"""
def _req(self, text: str = "Hello", voice: str | None = None) -> SynthesisRequest:
return SynthesisRequest(
text=text,
voice=VoiceSelection(source="builtin", key=voice or "default"),
parameters=ParameterValues(values={}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# ── Engine protocol ──────────────────────────────────────
def test_engine_satisfies_protocol(self, engine: Engine) -> None:
assert isinstance(engine, Engine)
def test_create_session_returns_session(self, engine: Engine) -> None:
session = engine.createSession()
assert isinstance(session, EngineSession)
session.dispose()
def test_create_session_returns_new_instances(self, engine: Engine) -> None:
s1 = engine.createSession()
s2 = engine.createSession()
assert s1 is not s2
s1.dispose()
s2.dispose()
def test_dispose_is_idempotent(self, engine: Engine) -> None:
engine.dispose()
engine.dispose()
def test_create_session_after_dispose_raises(self, engine: Engine) -> None:
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
# ── Session protocol ─────────────────────────────────────
def test_session_satisfies_protocol(self, engine: Engine) -> None:
session = engine.createSession()
assert isinstance(session, EngineSession)
session.dispose()
engine.dispose()
def test_session_synthesize_returns_audio(self, engine: Engine) -> None:
session = engine.createSession()
result = session.synthesize(self._req())
assert isinstance(result, SynthesizedAudio)
assert isinstance(result.data, bytes)
assert len(result.data) > 0
session.dispose()
engine.dispose()
def test_session_dispose_is_idempotent(self, engine: Engine) -> None:
session = engine.createSession()
session.dispose()
session.dispose()
engine.dispose()
def test_session_synthesize_after_dispose_raises(self, engine: Engine) -> None:
session = engine.createSession()
session.dispose()
with pytest.raises(EngineError):
session.synthesize(self._req())
engine.dispose()
def test_session_multiple_synthesize(self, engine: Engine) -> None:
session = engine.createSession()
r1 = session.synthesize(self._req())
r2 = session.synthesize(self._req())
assert isinstance(r1.data, bytes)
assert isinstance(r2.data, bytes)
session.dispose()
engine.dispose()
# ── Lifecycle ────────────────────────────────────────────
def test_full_lifecycle(self, engine: Engine) -> None:
s1 = engine.createSession()
s2 = engine.createSession()
s1.synthesize(self._req())
s2.synthesize(self._req())
s1.dispose()
s2.dispose()
engine.dispose()
tests/contracts/test_capabilities_contract.py
+183 -183
View File
@@ -1,183 +1,183 @@
"""Contract tests for capability interfaces.
These tests verify that capability interfaces satisfy the architectural requirements:
- VoiceLister: lists voices for a source
- PreviewGenerator: generates preview audio
- StreamingSynthesizer: yields audio chunks
- CancelableSession: cancels in-progress synthesis
"""
import pytest
from abogen.tts_plugin.capabilities import (
CancelableSession,
PreviewGenerator,
StreamingSynthesizer,
VoiceLister,
)
from abogen.tts_plugin.errors import CancelledError, EngineError
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
from .conftest import FakeCancelableSession, FakeStreamingSession, FakeVoiceListerEngine
class TestVoiceListerProtocolContract:
"""Contract tests for VoiceLister protocol."""
def test_voice_lister_is_protocol(self) -> None:
assert hasattr(VoiceLister, "__protocol_attrs__")
def test_voice_lister_satisfied_by_engine(self) -> None:
engine = FakeVoiceListerEngine()
assert isinstance(engine, VoiceLister)
def test_list_voices_returns_list(self) -> None:
engine = FakeVoiceListerEngine()
voices = engine.listVoices("builtin")
assert isinstance(voices, list)
def test_list_voices_returns_voice_manifests(self) -> None:
engine = FakeVoiceListerEngine()
voices = engine.listVoices("builtin")
for voice in voices:
assert isinstance(voice, VoiceManifest)
def test_list_voices_has_required_fields(self) -> None:
engine = FakeVoiceListerEngine()
voices = engine.listVoices("builtin")
for voice in voices:
assert hasattr(voice, "id")
assert hasattr(voice, "name")
assert hasattr(voice, "tags")
class TestPreviewGeneratorProtocolContract:
"""Contract tests for PreviewGenerator protocol."""
def test_preview_generator_is_protocol(self) -> None:
assert hasattr(PreviewGenerator, "__protocol_attrs__")
def test_preview_generator_satisfied_by_engine(self) -> None:
from .conftest import FakePreviewEngine
engine = FakePreviewEngine()
assert isinstance(engine, PreviewGenerator)
def test_generate_preview_returns_synthesized_audio(self) -> None:
from .conftest import FakePreviewEngine
engine = FakePreviewEngine()
voice = VoiceSelection(source="builtin", key="af_nova")
result = engine.generatePreview(voice, "Hello")
assert isinstance(result, SynthesizedAudio)
def test_generate_preview_has_valid_data(self) -> None:
from .conftest import FakePreviewEngine
engine = FakePreviewEngine()
voice = VoiceSelection(source="builtin", key="af_nova")
result = engine.generatePreview(voice, "Hello")
assert isinstance(result.data, bytes)
assert len(result.data) > 0
class TestStreamingSynthesizerProtocolContract:
"""Contract tests for StreamingSynthesizer protocol."""
def test_streaming_synthesizer_is_protocol(self) -> None:
assert hasattr(StreamingSynthesizer, "__protocol_attrs__")
def test_streaming_session_satisfies_protocol(self) -> None:
session = FakeStreamingSession()
assert isinstance(session, StreamingSynthesizer)
def test_synthesize_stream_yields_bytes(self) -> None:
session = FakeStreamingSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
chunks = list(session.synthesizeStream(request))
assert len(chunks) > 0
for chunk in chunks:
assert isinstance(chunk, bytes)
def test_streaming_iterator_exhaustion(self) -> None:
"""Architecture spec: Iterator exhaustion = synthesis complete."""
session = FakeStreamingSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
chunks = list(session.synthesizeStream(request))
assert len(chunks) == 3
def test_streaming_after_dispose_raises(self) -> None:
"""Architecture spec: After dispose(), methods raise EngineError."""
session = FakeStreamingSession()
session.dispose()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError):
list(session.synthesizeStream(request))
class TestCancelableSessionProtocolContract:
"""Contract tests for CancelableSession protocol."""
def test_cancelable_session_is_protocol(self) -> None:
assert hasattr(CancelableSession, "__protocol_attrs__")
def test_cancelable_session_satisfies_protocol(self) -> None:
session = FakeCancelableSession()
assert isinstance(session, CancelableSession)
def test_cancel_causes_synthesize_to_raise_cancelled(self) -> None:
"""Architecture spec: cancel() causes synthesize() to raise CancelledError."""
session = FakeCancelableSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# Cancel
session.cancel()
# synthesize should raise CancelledError
with pytest.raises(CancelledError):
session.synthesize(request)
def test_cancel_after_dispose_raises(self) -> None:
"""Architecture spec: cancel() raises EngineError if called after dispose()."""
session = FakeCancelableSession()
session.dispose()
with pytest.raises(EngineError):
session.cancel()
def test_session_usable_after_cancel(self) -> None:
"""Architecture spec: EngineSession remains usable after cancellation."""
session = FakeCancelableSession()
# Cancel
session.cancel()
# Dispose and create new session for synthesis
session.dispose()
"""Contract tests for capability interfaces.
These tests verify that capability interfaces satisfy the architectural requirements:
- VoiceLister: lists voices for a source
- PreviewGenerator: generates preview audio
- StreamingSynthesizer: yields audio chunks
- CancelableSession: cancels in-progress synthesis
"""
import pytest
from abogen.tts_plugin.capabilities import (
CancelableSession,
PreviewGenerator,
StreamingSynthesizer,
VoiceLister,
)
from abogen.tts_plugin.errors import CancelledError, EngineError
from abogen.tts_plugin.manifest import VoiceManifest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
from .conftest import FakeCancelableSession, FakeStreamingSession, FakeVoiceListerEngine
class TestVoiceListerProtocolContract:
"""Contract tests for VoiceLister protocol."""
def test_voice_lister_is_protocol(self) -> None:
assert hasattr(VoiceLister, "__protocol_attrs__")
def test_voice_lister_satisfied_by_engine(self) -> None:
engine = FakeVoiceListerEngine()
assert isinstance(engine, VoiceLister)
def test_list_voices_returns_list(self) -> None:
engine = FakeVoiceListerEngine()
voices = engine.listVoices("builtin")
assert isinstance(voices, list)
def test_list_voices_returns_voice_manifests(self) -> None:
engine = FakeVoiceListerEngine()
voices = engine.listVoices("builtin")
for voice in voices:
assert isinstance(voice, VoiceManifest)
def test_list_voices_has_required_fields(self) -> None:
engine = FakeVoiceListerEngine()
voices = engine.listVoices("builtin")
for voice in voices:
assert hasattr(voice, "id")
assert hasattr(voice, "name")
assert hasattr(voice, "tags")
class TestPreviewGeneratorProtocolContract:
"""Contract tests for PreviewGenerator protocol."""
def test_preview_generator_is_protocol(self) -> None:
assert hasattr(PreviewGenerator, "__protocol_attrs__")
def test_preview_generator_satisfied_by_engine(self) -> None:
from .conftest import FakePreviewEngine
engine = FakePreviewEngine()
assert isinstance(engine, PreviewGenerator)
def test_generate_preview_returns_synthesized_audio(self) -> None:
from .conftest import FakePreviewEngine
engine = FakePreviewEngine()
voice = VoiceSelection(source="builtin", key="af_nova")
result = engine.generatePreview(voice, "Hello")
assert isinstance(result, SynthesizedAudio)
def test_generate_preview_has_valid_data(self) -> None:
from .conftest import FakePreviewEngine
engine = FakePreviewEngine()
voice = VoiceSelection(source="builtin", key="af_nova")
result = engine.generatePreview(voice, "Hello")
assert isinstance(result.data, bytes)
assert len(result.data) > 0
class TestStreamingSynthesizerProtocolContract:
"""Contract tests for StreamingSynthesizer protocol."""
def test_streaming_synthesizer_is_protocol(self) -> None:
assert hasattr(StreamingSynthesizer, "__protocol_attrs__")
def test_streaming_session_satisfies_protocol(self) -> None:
session = FakeStreamingSession()
assert isinstance(session, StreamingSynthesizer)
def test_synthesize_stream_yields_bytes(self) -> None:
session = FakeStreamingSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
chunks = list(session.synthesizeStream(request))
assert len(chunks) > 0
for chunk in chunks:
assert isinstance(chunk, bytes)
def test_streaming_iterator_exhaustion(self) -> None:
"""Architecture spec: Iterator exhaustion = synthesis complete."""
session = FakeStreamingSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
chunks = list(session.synthesizeStream(request))
assert len(chunks) == 3
def test_streaming_after_dispose_raises(self) -> None:
"""Architecture spec: After dispose(), methods raise EngineError."""
session = FakeStreamingSession()
session.dispose()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError):
list(session.synthesizeStream(request))
class TestCancelableSessionProtocolContract:
"""Contract tests for CancelableSession protocol."""
def test_cancelable_session_is_protocol(self) -> None:
assert hasattr(CancelableSession, "__protocol_attrs__")
def test_cancelable_session_satisfies_protocol(self) -> None:
session = FakeCancelableSession()
assert isinstance(session, CancelableSession)
def test_cancel_causes_synthesize_to_raise_cancelled(self) -> None:
"""Architecture spec: cancel() causes synthesize() to raise CancelledError."""
session = FakeCancelableSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# Cancel
session.cancel()
# synthesize should raise CancelledError
with pytest.raises(CancelledError):
session.synthesize(request)
def test_cancel_after_dispose_raises(self) -> None:
"""Architecture spec: cancel() raises EngineError if called after dispose()."""
session = FakeCancelableSession()
session.dispose()
with pytest.raises(EngineError):
session.cancel()
def test_session_usable_after_cancel(self) -> None:
"""Architecture spec: EngineSession remains usable after cancellation."""
session = FakeCancelableSession()
# Cancel
session.cancel()
# Dispose and create new session for synthesis
session.dispose()
tests/contracts/test_engine_contract.py
+106 -106
View File
@@ -1,106 +1,106 @@
"""Contract tests for Engine protocol.
These tests verify that Engine implementations satisfy the architectural requirements:
- createSession() returns EngineSession
- dispose() is idempotent
- After dispose(), createSession() raises EngineError
- Engine is thread-safe for createSession()
"""
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from .conftest import FakeEngine, FakeEngineSession
class TestEngineProtocolContract:
"""Contract tests for the Engine protocol itself."""
def test_engine_is_protocol(self) -> None:
assert hasattr(Engine, "__protocol_attrs__")
def test_engine_session_is_protocol(self) -> None:
assert hasattr(EngineSession, "__protocol_attrs__")
def test_fake_engine_satisfies_protocol(self) -> None:
engine = FakeEngine()
assert isinstance(engine, Engine)
def test_fake_session_satisfies_protocol(self) -> None:
session = FakeEngineSession()
assert isinstance(session, EngineSession)
class TestEngineCreateSessionContract:
"""Contract tests for Engine.createSession()."""
def test_create_session_returns_engine_session(self) -> None:
engine = FakeEngine()
session = engine.createSession()
assert isinstance(session, EngineSession)
def test_create_session_returns_new_instance(self) -> None:
engine = FakeEngine()
session1 = engine.createSession()
session2 = engine.createSession()
assert session1 is not session2
def test_create_session_ownership_transfers(self) -> None:
"""Architecture spec: Ownership transfers to caller."""
engine = FakeEngine()
session = engine.createSession()
assert isinstance(session, EngineSession)
class TestEngineDisposeContract:
"""Contract tests for Engine.dispose()."""
def test_dispose_is_idempotent(self) -> None:
"""Architecture spec: dispose() is idempotent."""
engine = FakeEngine()
engine.dispose()
engine.dispose() # Should not raise
def test_dispose_never_raises(self) -> None:
"""Architecture spec: dispose() never raises exceptions."""
engine = FakeEngine()
engine.dispose() # Should not raise
def test_create_session_after_dispose_raises(self) -> None:
"""Architecture spec: After dispose(), all methods except dispose() raise EngineError."""
engine = FakeEngine()
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
class TestEngineLifecycleContract:
"""Contract tests for Engine lifecycle."""
def test_full_lifecycle(self) -> None:
"""Test complete engine lifecycle: create -> sessions -> dispose."""
engine = FakeEngine()
# Create sessions
session1 = engine.createSession()
session2 = engine.createSession()
# Use sessions
assert isinstance(session1, EngineSession)
assert isinstance(session2, EngineSession)
# Dispose sessions
session1.dispose()
session2.dispose()
# Dispose engine
engine.dispose()
def test_engine_disposed_session_raises(self) -> None:
"""Architecture spec: After dispose(), all methods except dispose() raise EngineError."""
engine = FakeEngine()
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
"""Contract tests for Engine protocol.
These tests verify that Engine implementations satisfy the architectural requirements:
- createSession() returns EngineSession
- dispose() is idempotent
- After dispose(), createSession() raises EngineError
- Engine is thread-safe for createSession()
"""
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from .conftest import FakeEngine, FakeEngineSession
class TestEngineProtocolContract:
"""Contract tests for the Engine protocol itself."""
def test_engine_is_protocol(self) -> None:
assert hasattr(Engine, "__protocol_attrs__")
def test_engine_session_is_protocol(self) -> None:
assert hasattr(EngineSession, "__protocol_attrs__")
def test_fake_engine_satisfies_protocol(self) -> None:
engine = FakeEngine()
assert isinstance(engine, Engine)
def test_fake_session_satisfies_protocol(self) -> None:
session = FakeEngineSession()
assert isinstance(session, EngineSession)
class TestEngineCreateSessionContract:
"""Contract tests for Engine.createSession()."""
def test_create_session_returns_engine_session(self) -> None:
engine = FakeEngine()
session = engine.createSession()
assert isinstance(session, EngineSession)
def test_create_session_returns_new_instance(self) -> None:
engine = FakeEngine()
session1 = engine.createSession()
session2 = engine.createSession()
assert session1 is not session2
def test_create_session_ownership_transfers(self) -> None:
"""Architecture spec: Ownership transfers to caller."""
engine = FakeEngine()
session = engine.createSession()
assert isinstance(session, EngineSession)
class TestEngineDisposeContract:
"""Contract tests for Engine.dispose()."""
def test_dispose_is_idempotent(self) -> None:
"""Architecture spec: dispose() is idempotent."""
engine = FakeEngine()
engine.dispose()
engine.dispose() # Should not raise
def test_dispose_never_raises(self) -> None:
"""Architecture spec: dispose() never raises exceptions."""
engine = FakeEngine()
engine.dispose() # Should not raise
def test_create_session_after_dispose_raises(self) -> None:
"""Architecture spec: After dispose(), all methods except dispose() raise EngineError."""
engine = FakeEngine()
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
class TestEngineLifecycleContract:
"""Contract tests for Engine lifecycle."""
def test_full_lifecycle(self) -> None:
"""Test complete engine lifecycle: create -> sessions -> dispose."""
engine = FakeEngine()
# Create sessions
session1 = engine.createSession()
session2 = engine.createSession()
# Use sessions
assert isinstance(session1, EngineSession)
assert isinstance(session2, EngineSession)
# Dispose sessions
session1.dispose()
session2.dispose()
# Dispose engine
engine.dispose()
def test_engine_disposed_session_raises(self) -> None:
"""Architecture spec: After dispose(), all methods except dispose() raise EngineError."""
engine = FakeEngine()
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
tests/contracts/test_errors_contract.py
+85 -85
View File
@@ -1,85 +1,85 @@
"""Contract tests for error hierarchy.
These tests verify that the error hierarchy satisfies the architectural requirements:
- All errors inherit from EngineError
- EngineError inherits from Exception
- Each error type is properly classified
"""
import pytest
from abogen.tts_plugin.errors import (
CancelledError,
ConfigurationError,
EngineError,
InternalError,
InvalidInputError,
ModelLoadError,
ModelNotFoundError,
NetworkError,
)
class TestErrorHierarchyContract:
"""Contract tests for the error hierarchy."""
def test_engine_error_is_exception(self) -> None:
assert issubclass(EngineError, Exception)
def test_all_errors_inherit_from_engine_error(self) -> None:
error_classes = [
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
]
for error_class in error_classes:
assert issubclass(error_class, EngineError), (
f"{error_class.__name__} must inherit from EngineError"
)
def test_all_errors_are_catchable(self) -> None:
error_classes = [
EngineError,
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
]
for error_class in error_classes:
with pytest.raises(EngineError):
raise error_class("test message")
def test_error_message_preserved(self) -> None:
msg = "Model not found: bert-base"
with pytest.raises(ModelNotFoundError, match=msg):
raise ModelNotFoundError(msg)
def test_error_can_be_caught_as_engine_error(self) -> None:
with pytest.raises(EngineError):
raise ModelNotFoundError("test")
def test_cancelled_error_is_engine_error(self) -> None:
"""CancelledError is a subtype of EngineError per architecture spec."""
assert issubclass(CancelledError, EngineError)
def test_error_hierarchy_no_cycles(self) -> None:
"""Verify no circular inheritance."""
error_classes = [
EngineError,
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
]
for cls in error_classes:
assert cls not in cls.__bases__
"""Contract tests for error hierarchy.
These tests verify that the error hierarchy satisfies the architectural requirements:
- All errors inherit from EngineError
- EngineError inherits from Exception
- Each error type is properly classified
"""
import pytest
from abogen.tts_plugin.errors import (
CancelledError,
ConfigurationError,
EngineError,
InternalError,
InvalidInputError,
ModelLoadError,
ModelNotFoundError,
NetworkError,
)
class TestErrorHierarchyContract:
"""Contract tests for the error hierarchy."""
def test_engine_error_is_exception(self) -> None:
assert issubclass(EngineError, Exception)
def test_all_errors_inherit_from_engine_error(self) -> None:
error_classes = [
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
]
for error_class in error_classes:
assert issubclass(error_class, EngineError), (
f"{error_class.__name__} must inherit from EngineError"
)
def test_all_errors_are_catchable(self) -> None:
error_classes = [
EngineError,
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
]
for error_class in error_classes:
with pytest.raises(EngineError):
raise error_class("test message")
def test_error_message_preserved(self) -> None:
msg = "Model not found: bert-base"
with pytest.raises(ModelNotFoundError, match=msg):
raise ModelNotFoundError(msg)
def test_error_can_be_caught_as_engine_error(self) -> None:
with pytest.raises(EngineError):
raise ModelNotFoundError("test")
def test_cancelled_error_is_engine_error(self) -> None:
"""CancelledError is a subtype of EngineError per architecture spec."""
assert issubclass(CancelledError, EngineError)
def test_error_hierarchy_no_cycles(self) -> None:
"""Verify no circular inheritance."""
error_classes = [
EngineError,
ModelNotFoundError,
ModelLoadError,
NetworkError,
InvalidInputError,
ConfigurationError,
CancelledError,
InternalError,
]
for cls in error_classes:
assert cls not in cls.__bases__
tests/contracts/test_host_context_contract.py
+89 -89
View File
@@ -1,89 +1,89 @@
"""Contract tests for HostContext.
These tests verify that HostContext satisfies the architectural requirements:
- Minimal (3 fields maximum)
- Frozen dataclass
- config_dir: Path
- logger: Logger
- http_client: HttpClient protocol
"""
import logging
from pathlib import Path
import pytest
from abogen.tts_plugin.host_context import HttpClient, HostContext
class TestHostContextContract:
"""Contract tests for HostContext dataclass."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(HostContext, "__dataclass_params__")
assert HostContext.__dataclass_params__.frozen is True
def test_required_fields(self, tmp_path: Path) -> None:
logger = logging.getLogger("test")
class FakeClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
ctx = HostContext(
config_dir=tmp_path,
logger=logger,
http_client=FakeClient(),
)
assert ctx.config_dir == tmp_path
assert ctx.logger is logger
def test_immutability(self, tmp_path: Path) -> None:
class FakeClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
ctx = HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=FakeClient(),
)
with pytest.raises(AttributeError):
ctx.config_dir = Path("/other") # type: ignore[misc]
def test_max_three_fields(self) -> None:
"""Architecture spec: HostContext is minimal (3 fields max)."""
import dataclasses
fields = dataclasses.fields(HostContext)
assert len(fields) <= 3
class TestHttpClientProtocolContract:
"""Contract tests for HttpClient protocol."""
def test_http_client_is_protocol(self) -> None:
assert hasattr(HttpClient, "__protocol_attrs__")
def test_http_client_has_get(self) -> None:
assert hasattr(HttpClient, "get")
def test_http_client_has_post(self) -> None:
assert hasattr(HttpClient, "post")
def test_http_client_satisfied(self) -> None:
class FakeClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
client = FakeClient()
assert isinstance(client, HttpClient)
"""Contract tests for HostContext.
These tests verify that HostContext satisfies the architectural requirements:
- Minimal (3 fields maximum)
- Frozen dataclass
- config_dir: Path
- logger: Logger
- http_client: HttpClient protocol
"""
import logging
from pathlib import Path
import pytest
from abogen.tts_plugin.host_context import HttpClient, HostContext
class TestHostContextContract:
"""Contract tests for HostContext dataclass."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(HostContext, "__dataclass_params__")
assert HostContext.__dataclass_params__.frozen is True
def test_required_fields(self, tmp_path: Path) -> None:
logger = logging.getLogger("test")
class FakeClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
ctx = HostContext(
config_dir=tmp_path,
logger=logger,
http_client=FakeClient(),
)
assert ctx.config_dir == tmp_path
assert ctx.logger is logger
def test_immutability(self, tmp_path: Path) -> None:
class FakeClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
ctx = HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=FakeClient(),
)
with pytest.raises(AttributeError):
ctx.config_dir = Path("/other") # type: ignore[misc]
def test_max_three_fields(self) -> None:
"""Architecture spec: HostContext is minimal (3 fields max)."""
import dataclasses
fields = dataclasses.fields(HostContext)
assert len(fields) <= 3
class TestHttpClientProtocolContract:
"""Contract tests for HttpClient protocol."""
def test_http_client_is_protocol(self) -> None:
assert hasattr(HttpClient, "__protocol_attrs__")
def test_http_client_has_get(self) -> None:
assert hasattr(HttpClient, "get")
def test_http_client_has_post(self) -> None:
assert hasattr(HttpClient, "post")
def test_http_client_satisfied(self) -> None:
class FakeClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
client = FakeClient()
assert isinstance(client, HttpClient)
tests/contracts/test_integration.py
+420 -420
View File
@@ -1,420 +1,420 @@
"""Integration tests for the TTS Plugin Architecture.
These tests verify:
1. Consumer Flow: consumer → plugin → engine → session → synthesis → result
2. Lifecycle: dispose, no leaks, error handling
3. Regression: old path vs new path equivalence
Tests use mock plugins to avoid requiring real TTS dependencies.
"""
import pytest
from typing import Any, Iterator
from unittest.mock import MagicMock, patch
import numpy as np
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.plugin_manager import PluginManager, get_plugin_manager, reset_plugin_manager
from abogen.tts_plugin.utils import Pipeline, create_pipeline
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class MockEngineSession:
"""Mock EngineSession that records calls for verification."""
def __init__(self):
self._disposed = False
self.synthesize_calls = []
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
raise EngineError("Session disposed")
self.synthesize_calls.append(request)
# Return fake audio
audio = np.ones(1000, dtype=np.float32) * 0.5
return SynthesizedAudio(
data=audio.tobytes(),
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1000 / 24000),
)
def dispose(self) -> None:
self._disposed = True
class MockEngine:
"""Mock Engine that creates MockEngineSessions."""
def __init__(self, **kwargs):
self.kwargs = kwargs
self._disposed = False
self.sessions_created = []
def createSession(self) -> MockEngineSession:
if self._disposed:
raise EngineError("Engine disposed")
session = MockEngineSession()
self.sessions_created.append(session)
return session
def dispose(self) -> None:
self._disposed = True
def create_mock_plugin(create_engine_func=None):
"""Helper to create a mock plugin module."""
if create_engine_func is None:
create_engine_func = lambda **kwargs: MockEngine(**kwargs)
from abogen.tts_plugin.manifest import PluginManifest, EngineManifest
manifest = PluginManifest(
id="mock_tts",
name="Mock TTS",
version="1.0.0",
api_version="1.0",
description="Mock TTS for testing",
author="Test",
capabilities=(),
requires=None,
engine=EngineManifest(
voiceSources=(),
parameters=(),
audioFormats=(),
),
)
return {
"PLUGIN_MANIFEST": manifest,
"MODEL_REQUIREMENTS": [],
"create_engine": create_mock_plugin_engine if create_engine_func is None else create_engine_func,
}
def create_mock_plugin_engine(**kwargs):
"""Default mock plugin engine factory."""
return MockEngine(**kwargs)
class TestConsumerFlow:
"""Consumer Flow Test: consumer → plugin → engine → session → synthesis → result"""
def test_full_consumer_flow(self):
"""Verify complete flow from consumer to audio output."""
manager = PluginManager()
# Register mock plugin
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# Step 1: Consumer gets plugin
assert manager.has_plugin("mock_tts") is True
# Step 2: Plugin creates engine
engine = manager.create_engine("mock_tts")
assert engine is not None
assert isinstance(engine, MockEngine)
# Step 3: Engine creates session
session = engine.createSession()
assert session is not None
assert isinstance(session, MockEngineSession)
# Step 4: Session synthesizes
request = SynthesisRequest(
text="Hello world",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(values={"speed": 1.0}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
# Step 5: Result returned
assert result is not None
assert isinstance(result, SynthesizedAudio)
assert len(result.data) > 0
assert result.format.mime == "audio/wav"
assert result.duration.seconds > 0
def test_consumer_flow_via_pipeline(self):
"""Verify flow through Pipeline utility matches direct flow."""
manager = PluginManager()
# Register mock plugin
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# Use Pipeline utility
with patch("abogen.tts_plugin.utils.get_plugin_manager", return_value=manager):
backend = create_pipeline("mock_tts")
# Call like old TTSBackend
segments = list(backend("Hello world", voice="default", speed=1.0))
# Verify result
assert len(segments) >= 1
segment = segments[0]
assert hasattr(segment, "graphemes")
assert hasattr(segment, "audio")
assert segment.graphemes == "Hello world"
class TestLifecycle:
"""Lifecycle Test: dispose, no leaks, error handling"""
def test_session_dispose_is_idempotent(self):
"""dispose() can be called multiple times safely."""
session = MockEngineSession()
session.dispose()
session.dispose() # Should not raise
assert session._disposed is True
def test_session_synthesize_after_dispose_raises(self):
"""synthesize() after dispose() raises EngineError."""
session = MockEngineSession()
session.dispose()
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError):
session.synthesize(request)
def test_engine_dispose_is_idempotent(self):
"""Engine dispose() can be called multiple times safely."""
engine = MockEngine()
engine.dispose()
engine.dispose() # Should not raise
assert engine._disposed is True
def test_engine_create_session_after_dispose_raises(self):
"""createSession() after dispose() raises EngineError."""
engine = MockEngine()
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
def test_full_lifecycle(self):
"""Test complete lifecycle: create → use → dispose."""
engine = MockEngine()
# Create and use session
session = engine.createSession()
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert len(result.data) > 0
# Dispose session
session.dispose()
assert session._disposed is True
# Dispose engine
engine.dispose()
assert engine._disposed is True
def test_no_session_leak_on_engine_dispose(self):
"""Engine can be disposed even if sessions were created."""
engine = MockEngine()
# Create multiple sessions
session1 = engine.createSession()
session2 = engine.createSession()
# Use sessions
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
session1.synthesize(request)
session2.synthesize(request)
# Dispose engine (sessions still exist but engine is disposed)
engine.dispose()
assert engine._disposed is True
# Sessions can still be used (they hold reference to pipeline)
result = session1.synthesize(request)
assert len(result.data) > 0
def test_error_handling_in_synthesis(self):
"""Error during synthesis is handled correctly."""
class FailingSession:
def synthesize(self, request):
raise EngineError("Synthesis failed")
def dispose(self):
pass
session = FailingSession()
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError, match="Synthesis failed"):
session.synthesize(request)
class TestRegression:
"""Regression Test: old path vs new path equivalence"""
def test_old_path_vs_new_path_same_result(self):
"""Both paths should produce equivalent results."""
# Setup mock plugin
manager = PluginManager()
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# New path: Plugin Manager → Engine → Session → Synthesis
with patch("abogen.tts_plugin.utils.get_plugin_manager", return_value=manager):
new_backend = create_pipeline("mock_tts")
new_segments = list(new_backend("Hello world", voice="default", speed=1.0))
# Old path: Direct MockEngine (simulating old registry)
old_engine = MockEngine()
old_session = old_engine.createSession()
request = SynthesisRequest(
text="Hello world",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(values={"speed": 1.0}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
old_result = old_session.synthesize(request)
# Compare results
# New path returns segments, old path returns SynthesizedAudio
# But both should have valid audio data
assert len(new_segments) >= 1
assert len(old_result.data) > 0
# Both should have same format
assert new_segments[0].audio.dtype == np.float32
def test_pipeline_matches_old_interface(self):
"""Pipeline utility should match old TTSBackend interface."""
manager = PluginManager()
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
with patch("abogen.tts_plugin.utils.get_plugin_manager", return_value=manager):
backend = create_pipeline("mock_tts", lang_code="a", device="cpu")
# Old interface: pipeline(text, voice=..., speed=..., split_pattern=...)
segments = list(backend(
"Hello world",
voice="af_heart",
speed=1.0,
split_pattern=r"\n+"
))
# Should return segments with graphemes and audio
assert len(segments) >= 1
segment = segments[0]
assert segment.graphemes == "Hello world"
assert isinstance(segment.audio, np.ndarray)
assert segment.audio.dtype == np.float32
assert len(segment.audio) > 0
class TestPluginManagerIntegration:
"""Integration tests for PluginManager."""
def test_plugin_manager_singleton_pattern(self):
"""Global plugin manager follows singleton pattern."""
reset_plugin_manager()
manager1 = get_plugin_manager()
manager2 = get_plugin_manager()
assert manager1 is manager2
reset_plugin_manager()
manager3 = get_plugin_manager()
assert manager1 is not manager3
def test_plugin_manager_discover_plugins(self):
"""Plugin manager can discover plugins from directory."""
manager = PluginManager()
# Discover from test plugins directory
manager.discover("tests/plugins")
# Should find valid_plugin
# (This depends on test plugins existing)
plugins = manager.list_plugins()
assert isinstance(plugins, list)
def test_plugin_manager_dispose_all(self):
"""Plugin manager can dispose all cached engines."""
manager = PluginManager()
# Register mock plugin
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# Create engines
engine1 = manager.get_or_create_engine("mock_tts")
engine2 = manager.get_or_create_engine("mock_tts")
# Dispose all
manager.dispose_all()
# Engines should be disposed
assert engine1._disposed is True
assert engine2._disposed is True
# Cache should be empty
assert len(manager._engines) == 0
class TestNoCompatLayer:
"""Regression: confirm the compatibility layer has been removed."""
def test_compat_module_does_not_exist(self):
"""abogen.tts_plugin.compat must not be importable."""
import importlib
with pytest.raises((ImportError, ModuleNotFoundError)):
importlib.import_module("abogen.tts_plugin.compat")
def test_consumers_use_plugin_architecture_directly(self):
"""Key consumers import from abogen.tts_plugin.utils, not compat."""
import inspect, abogen.voice_profiles, abogen.voice_formulas, abogen.voice_cache
for mod in (abogen.voice_profiles, abogen.voice_formulas, abogen.voice_cache):
source = inspect.getsource(mod)
assert "tts_plugin.compat" not in source, (
f"{mod.__name__} still references tts_plugin.compat"
)
"""Integration tests for the TTS Plugin Architecture.
These tests verify:
1. Consumer Flow: consumer → plugin → engine → session → synthesis → result
2. Lifecycle: dispose, no leaks, error handling
3. Regression: old path vs new path equivalence
Tests use mock plugins to avoid requiring real TTS dependencies.
"""
import pytest
from typing import Any, Iterator
from unittest.mock import MagicMock, patch
import numpy as np
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.plugin_manager import PluginManager, get_plugin_manager, reset_plugin_manager
from abogen.tts_plugin.utils import Pipeline, create_pipeline
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class MockEngineSession:
"""Mock EngineSession that records calls for verification."""
def __init__(self):
self._disposed = False
self.synthesize_calls = []
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
raise EngineError("Session disposed")
self.synthesize_calls.append(request)
# Return fake audio
audio = np.ones(1000, dtype=np.float32) * 0.5
return SynthesizedAudio(
data=audio.tobytes(),
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1000 / 24000),
)
def dispose(self) -> None:
self._disposed = True
class MockEngine:
"""Mock Engine that creates MockEngineSessions."""
def __init__(self, **kwargs):
self.kwargs = kwargs
self._disposed = False
self.sessions_created = []
def createSession(self) -> MockEngineSession:
if self._disposed:
raise EngineError("Engine disposed")
session = MockEngineSession()
self.sessions_created.append(session)
return session
def dispose(self) -> None:
self._disposed = True
def create_mock_plugin(create_engine_func=None):
"""Helper to create a mock plugin module."""
if create_engine_func is None:
create_engine_func = lambda **kwargs: MockEngine(**kwargs)
from abogen.tts_plugin.manifest import PluginManifest, EngineManifest
manifest = PluginManifest(
id="mock_tts",
name="Mock TTS",
version="1.0.0",
api_version="1.0",
description="Mock TTS for testing",
author="Test",
capabilities=(),
requires=None,
engine=EngineManifest(
voiceSources=(),
parameters=(),
audioFormats=(),
),
)
return {
"PLUGIN_MANIFEST": manifest,
"MODEL_REQUIREMENTS": [],
"create_engine": create_mock_plugin_engine if create_engine_func is None else create_engine_func,
}
def create_mock_plugin_engine(**kwargs):
"""Default mock plugin engine factory."""
return MockEngine(**kwargs)
class TestConsumerFlow:
"""Consumer Flow Test: consumer → plugin → engine → session → synthesis → result"""
def test_full_consumer_flow(self):
"""Verify complete flow from consumer to audio output."""
manager = PluginManager()
# Register mock plugin
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# Step 1: Consumer gets plugin
assert manager.has_plugin("mock_tts") is True
# Step 2: Plugin creates engine
engine = manager.create_engine("mock_tts")
assert engine is not None
assert isinstance(engine, MockEngine)
# Step 3: Engine creates session
session = engine.createSession()
assert session is not None
assert isinstance(session, MockEngineSession)
# Step 4: Session synthesizes
request = SynthesisRequest(
text="Hello world",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(values={"speed": 1.0}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
# Step 5: Result returned
assert result is not None
assert isinstance(result, SynthesizedAudio)
assert len(result.data) > 0
assert result.format.mime == "audio/wav"
assert result.duration.seconds > 0
def test_consumer_flow_via_pipeline(self):
"""Verify flow through Pipeline utility matches direct flow."""
manager = PluginManager()
# Register mock plugin
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# Use Pipeline utility
with patch("abogen.tts_plugin.utils.get_plugin_manager", return_value=manager):
backend = create_pipeline("mock_tts")
# Call like old TTSBackend
segments = list(backend("Hello world", voice="default", speed=1.0))
# Verify result
assert len(segments) >= 1
segment = segments[0]
assert hasattr(segment, "graphemes")
assert hasattr(segment, "audio")
assert segment.graphemes == "Hello world"
class TestLifecycle:
"""Lifecycle Test: dispose, no leaks, error handling"""
def test_session_dispose_is_idempotent(self):
"""dispose() can be called multiple times safely."""
session = MockEngineSession()
session.dispose()
session.dispose() # Should not raise
assert session._disposed is True
def test_session_synthesize_after_dispose_raises(self):
"""synthesize() after dispose() raises EngineError."""
session = MockEngineSession()
session.dispose()
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError):
session.synthesize(request)
def test_engine_dispose_is_idempotent(self):
"""Engine dispose() can be called multiple times safely."""
engine = MockEngine()
engine.dispose()
engine.dispose() # Should not raise
assert engine._disposed is True
def test_engine_create_session_after_dispose_raises(self):
"""createSession() after dispose() raises EngineError."""
engine = MockEngine()
engine.dispose()
with pytest.raises(EngineError):
engine.createSession()
def test_full_lifecycle(self):
"""Test complete lifecycle: create → use → dispose."""
engine = MockEngine()
# Create and use session
session = engine.createSession()
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert len(result.data) > 0
# Dispose session
session.dispose()
assert session._disposed is True
# Dispose engine
engine.dispose()
assert engine._disposed is True
def test_no_session_leak_on_engine_dispose(self):
"""Engine can be disposed even if sessions were created."""
engine = MockEngine()
# Create multiple sessions
session1 = engine.createSession()
session2 = engine.createSession()
# Use sessions
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
session1.synthesize(request)
session2.synthesize(request)
# Dispose engine (sessions still exist but engine is disposed)
engine.dispose()
assert engine._disposed is True
# Sessions can still be used (they hold reference to pipeline)
result = session1.synthesize(request)
assert len(result.data) > 0
def test_error_handling_in_synthesis(self):
"""Error during synthesis is handled correctly."""
class FailingSession:
def synthesize(self, request):
raise EngineError("Synthesis failed")
def dispose(self):
pass
session = FailingSession()
request = SynthesisRequest(
text="test",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError, match="Synthesis failed"):
session.synthesize(request)
class TestRegression:
"""Regression Test: old path vs new path equivalence"""
def test_old_path_vs_new_path_same_result(self):
"""Both paths should produce equivalent results."""
# Setup mock plugin
manager = PluginManager()
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# New path: Plugin Manager → Engine → Session → Synthesis
with patch("abogen.tts_plugin.utils.get_plugin_manager", return_value=manager):
new_backend = create_pipeline("mock_tts")
new_segments = list(new_backend("Hello world", voice="default", speed=1.0))
# Old path: Direct MockEngine (simulating old registry)
old_engine = MockEngine()
old_session = old_engine.createSession()
request = SynthesisRequest(
text="Hello world",
voice=VoiceSelection(source="builtin", key="default"),
parameters=ParameterValues(values={"speed": 1.0}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
old_result = old_session.synthesize(request)
# Compare results
# New path returns segments, old path returns SynthesizedAudio
# But both should have valid audio data
assert len(new_segments) >= 1
assert len(old_result.data) > 0
# Both should have same format
assert new_segments[0].audio.dtype == np.float32
def test_pipeline_matches_old_interface(self):
"""Pipeline utility should match old TTSBackend interface."""
manager = PluginManager()
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
with patch("abogen.tts_plugin.utils.get_plugin_manager", return_value=manager):
backend = create_pipeline("mock_tts", lang_code="a", device="cpu")
# Old interface: pipeline(text, voice=..., speed=..., split_pattern=...)
segments = list(backend(
"Hello world",
voice="af_heart",
speed=1.0,
split_pattern=r"\n+"
))
# Should return segments with graphemes and audio
assert len(segments) >= 1
segment = segments[0]
assert segment.graphemes == "Hello world"
assert isinstance(segment.audio, np.ndarray)
assert segment.audio.dtype == np.float32
assert len(segment.audio) > 0
class TestPluginManagerIntegration:
"""Integration tests for PluginManager."""
def test_plugin_manager_singleton_pattern(self):
"""Global plugin manager follows singleton pattern."""
reset_plugin_manager()
manager1 = get_plugin_manager()
manager2 = get_plugin_manager()
assert manager1 is manager2
reset_plugin_manager()
manager3 = get_plugin_manager()
assert manager1 is not manager3
def test_plugin_manager_discover_plugins(self):
"""Plugin manager can discover plugins from directory."""
manager = PluginManager()
# Discover from test plugins directory
manager.discover("tests/plugins")
# Should find valid_plugin
# (This depends on test plugins existing)
plugins = manager.list_plugins()
assert isinstance(plugins, list)
def test_plugin_manager_dispose_all(self):
"""Plugin manager can dispose all cached engines."""
manager = PluginManager()
# Register mock plugin
mock_plugin = create_mock_plugin()
manager._plugins["mock_tts"] = mock_plugin
manager._loaded = True
# Create engines
engine1 = manager.get_or_create_engine("mock_tts")
engine2 = manager.get_or_create_engine("mock_tts")
# Dispose all
manager.dispose_all()
# Engines should be disposed
assert engine1._disposed is True
assert engine2._disposed is True
# Cache should be empty
assert len(manager._engines) == 0
class TestNoCompatLayer:
"""Regression: confirm the compatibility layer has been removed."""
def test_compat_module_does_not_exist(self):
"""abogen.tts_plugin.compat must not be importable."""
import importlib
with pytest.raises((ImportError, ModuleNotFoundError)):
importlib.import_module("abogen.tts_plugin.compat")
def test_consumers_use_plugin_architecture_directly(self):
"""Key consumers import from abogen.tts_plugin.utils, not compat."""
import inspect, abogen.voice_profiles, abogen.voice_formulas, abogen.voice_cache
for mod in (abogen.voice_profiles, abogen.voice_formulas, abogen.voice_cache):
source = inspect.getsource(mod)
assert "tts_plugin.compat" not in source, (
f"{mod.__name__} still references tts_plugin.compat"
)
tests/contracts/test_loader_contract.py
+436 -436
View File
@@ -1,436 +1,436 @@
"""Comprehensive tests for the plugin loader infrastructure.
These tests verify that the loader correctly:
- Discovers plugins in directories
- Imports plugin modules
- Validates PLUGIN_MANIFEST, MODEL_REQUIREMENTS, create_engine
- Validates api_version compatibility
- Validates capabilities
- Provides diagnostic messages for errors
- Rejects invalid plugins
"""
from __future__ import annotations
import sys
from pathlib import Path
import pytest
from abogen.tts_plugin.loader import (
HOST_API_VERSION,
PluginLoadError,
PluginLoadResult,
_check_api_version_compatibility,
_parse_api_version,
_validate_api_version,
_validate_capabilities,
_validate_manifest,
discover_plugins,
load_plugin,
load_plugin_from_dir,
)
from abogen.tts_plugin.manifest import (
EngineManifest,
ModelManifest,
PluginManifest,
)
# ──────────────────────────────────────────────────────────────
# Path fixtures
# ──────────────────────────────────────────────────────────────
@pytest.fixture
def plugins_dir() -> Path:
return Path(__file__).parent.parent / "plugins"
@pytest.fixture
def fake_plugin_dir(plugins_dir: Path) -> Path:
return plugins_dir / "fake_plugin"
@pytest.fixture
def missing_manifest_dir(plugins_dir: Path) -> Path:
return plugins_dir / "missing_manifest"
@pytest.fixture
def invalid_api_version_dir(plugins_dir: Path) -> Path:
return plugins_dir / "invalid_api_version"
@pytest.fixture
def invalid_capabilities_dir(plugins_dir: Path) -> Path:
return plugins_dir / "invalid_capabilities"
@pytest.fixture
def missing_create_engine_dir(plugins_dir: Path) -> Path:
return plugins_dir / "missing_create_engine"
@pytest.fixture
def import_error_dir(plugins_dir: Path) -> Path:
return plugins_dir / "import_error"
@pytest.fixture
def missing_model_requirements_dir(plugins_dir: Path) -> Path:
return plugins_dir / "missing_model_requirements"
# ──────────────────────────────────────────────────────────────
# Unit tests: _parse_api_version
# ──────────────────────────────────────────────────────────────
class TestParseApiVersion:
def test_valid_version(self) -> None:
assert _parse_api_version("1.0") == (1, 0)
assert _parse_api_version("2.5") == (2, 5)
assert _parse_api_version("10.20") == (10, 20)
def test_invalid_format(self) -> None:
assert _parse_api_version("1") is None
assert _parse_api_version("1.0.0") is None
assert _parse_api_version("abc") is None
assert _parse_api_version("") is None
assert _parse_api_version("1.x") is None
# ──────────────────────────────────────────────────────────────
# Unit tests: _check_api_version_compatibility
# ──────────────────────────────────────────────────────────────
class TestCheckApiVersionCompatibility:
def test_compatible_version(self) -> None:
assert _check_api_version_compatibility("1.0") is None
assert _check_api_version_compatibility("1.5") is None
def test_major_mismatch(self) -> None:
error = _check_api_version_compatibility("2.0")
assert error is not None
assert "major mismatch" in error
def test_invalid_format(self) -> None:
error = _check_api_version_compatibility("invalid")
assert error is not None
assert "Invalid api_version format" in error
# ──────────────────────────────────────────────────────────────
# Unit tests: _validate_manifest
# ──────────────────────────────────────────────────────────────
class TestValidateManifest:
def test_valid_manifest(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS: list = []
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert errors == []
def test_missing_manifest(self) -> None:
class FakeModule:
MODEL_REQUIREMENTS: list = []
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("Missing PLUGIN_MANIFEST" in e for e in errors)
def test_wrong_manifest_type(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = "not a manifest"
MODEL_REQUIREMENTS: list = []
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("PluginManifest instance" in e for e in errors)
def test_missing_model_requirements(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("Missing MODEL_REQUIREMENTS" in e for e in errors)
def test_wrong_model_requirements_type(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS = "not a list"
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("must be a list" in e for e in errors)
def test_invalid_model_requirements_item(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS = ["not a model manifest"]
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("ModelManifest instance" in e for e in errors)
def test_missing_create_engine(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS: list = []
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("Missing create_engine" in e for e in errors)
def test_create_engine_not_callable(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS: list = []
create_engine = "not callable"
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("must be callable" in e for e in errors)
# ──────────────────────────────────────────────────────────────
# Unit tests: _validate_capabilities
# ──────────────────────────────────────────────────────────────
class TestValidateCapabilities:
def test_valid_capabilities(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
capabilities=("voice_list", "preview"),
)
errors = _validate_capabilities(manifest)
assert errors == []
def test_unknown_capability(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
capabilities=("voice_list", "unknown_cap"),
)
errors = _validate_capabilities(manifest)
assert any("unknown_cap" in e for e in errors)
def test_empty_capabilities(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
capabilities=(),
)
errors = _validate_capabilities(manifest)
assert errors == []
# ──────────────────────────────────────────────────────────────
# Unit tests: _validate_api_version
# ──────────────────────────────────────────────────────────────
class TestValidateApiVersion:
def test_compatible(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
errors = _validate_api_version(manifest)
assert errors == []
def test_incompatible(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="2.0", description="Test", author="Test",
)
errors = _validate_api_version(manifest)
assert len(errors) > 0
# ──────────────────────────────────────────────────────────────
# Integration tests: load_plugin_from_dir
# ──────────────────────────────────────────────────────────────
class TestLoadPluginFromDir:
def test_load_valid_plugin(self, fake_plugin_dir: Path) -> None:
result = load_plugin_from_dir(fake_plugin_dir)
assert result.success is True
assert result.manifest is not None
assert result.manifest.id == "fake_plugin"
assert result.model_requirements is not None
assert result.create_engine is not None
assert result.module is not None
assert result.error is None
def test_plugin_satisfies_protocol(self, fake_plugin_dir: Path) -> None:
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
import logging
result = load_plugin_from_dir(fake_plugin_dir)
assert result.success is True
# Create engine using the loaded create_engine function
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda s, **kw: None, "post": lambda s, **kw: None})(),
)
engine = result.create_engine(ctx, None, __import__("abogen.tts_plugin.types", fromlist=["EngineConfig"]).EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
def test_nonexistent_directory(self, tmp_path: Path) -> None:
result = load_plugin_from_dir(tmp_path / "nonexistent")
assert result.success is False
assert result.error is not None
assert "does not exist" in result.error.errors[0]
def test_missing_init_file(self, tmp_path: Path) -> None:
plugin_dir = tmp_path / "no_init"
plugin_dir.mkdir()
result = load_plugin_from_dir(plugin_dir)
assert result.success is False
assert result.error is not None
assert "__init__.py" in result.error.errors[0]
def test_import_error(self, import_error_dir: Path) -> None:
result = load_plugin_from_dir(import_error_dir)
assert result.success is False
assert result.error is not None
assert "Failed to import" in result.error.errors[0]
# ──────────────────────────────────────────────────────────────
# Integration tests: invalid plugins
# ──────────────────────────────────────────────────────────────
class TestInvalidPlugins:
def test_missing_manifest(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.success is False
assert result.error is not None
assert any("Missing PLUGIN_MANIFEST" in e for e in result.error.errors)
def test_invalid_api_version(self, invalid_api_version_dir: Path) -> None:
result = load_plugin_from_dir(invalid_api_version_dir)
assert result.success is False
assert result.error is not None
assert any("major mismatch" in e for e in result.error.errors)
def test_invalid_capabilities(self, invalid_capabilities_dir: Path) -> None:
result = load_plugin_from_dir(invalid_capabilities_dir)
assert result.success is False
assert result.error is not None
assert any("Unknown capability" in e for e in result.error.errors)
def test_missing_create_engine(self, missing_create_engine_dir: Path) -> None:
result = load_plugin_from_dir(missing_create_engine_dir)
assert result.success is False
assert result.error is not None
assert any("Missing create_engine" in e for e in result.error.errors)
def test_missing_model_requirements(self, missing_model_requirements_dir: Path) -> None:
result = load_plugin_from_dir(missing_model_requirements_dir)
assert result.success is False
assert result.error is not None
assert any("Missing MODEL_REQUIREMENTS" in e for e in result.error.errors)
# ──────────────────────────────────────────────────────────────
# Integration tests: discover_plugins
# ──────────────────────────────────────────────────────────────
class TestDiscoverPlugins:
def test_discover_from_valid_dir(self, plugins_dir: Path) -> None:
results = discover_plugins([plugins_dir])
# Should find multiple plugins (valid and invalid)
assert len(results) > 0
def test_discover_includes_valid_plugin(self, plugins_dir: Path) -> None:
results = discover_plugins([plugins_dir])
valid = [r for r in results if r.success]
assert len(valid) >= 1
assert any(r.manifest and r.manifest.id == "fake_plugin" for r in valid)
def test_discover_includes_invalid_plugins(self, plugins_dir: Path) -> None:
results = discover_plugins([plugins_dir])
invalid = [r for r in results if not r.success]
assert len(invalid) >= 1
def test_discover_nonexistent_dir(self, tmp_path: Path) -> None:
results = discover_plugins([tmp_path / "nonexistent"])
assert results == []
def test_discover_multiple_dirs(self, plugins_dir: Path, tmp_path: Path) -> None:
results = discover_plugins([plugins_dir, tmp_path / "nonexistent"])
assert len(results) > 0
# ──────────────────────────────────────────────────────────────
# Diagnostic messages tests
# ──────────────────────────────────────────────────────────────
class TestDiagnosticMessages:
def test_error_contains_plugin_id(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.error is not None
assert result.error.plugin_id == "missing_manifest"
def test_error_contains_path(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.error is not None
assert result.error.path == missing_manifest_dir
def test_error_contains_messages(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.error is not None
assert len(result.error.errors) > 0
def test_multiple_errors(self, invalid_api_version_dir: Path) -> None:
# This plugin has multiple issues
result = load_plugin_from_dir(invalid_api_version_dir)
assert result.error is not None
# Should have at least the api_version error
assert len(result.error.errors) >= 1
# ──────────────────────────────────────────────────────────────
# No partial registration tests
# ──────────────────────────────────────────────────────────────
class TestNoPartialRegistration:
def test_invalid_plugin_no_manifest_attr(self, missing_manifest_dir: Path) -> None:
"""After failed load, module should not remain in sys.modules."""
result = load_plugin_from_dir(missing_manifest_dir)
assert result.success is False
# Module should not be registered
module_name = f"abogen.tts_plugin._loaded.missing_manifest"
assert module_name not in sys.modules
def test_import_error_no_registration(self, import_error_dir: Path) -> None:
"""After import error, module should not remain in sys.modules."""
result = load_plugin_from_dir(import_error_dir)
assert result.success is False
module_name = f"abogen.tts_plugin._loaded.import_error"
assert module_name not in sys.modules
"""Comprehensive tests for the plugin loader infrastructure.
These tests verify that the loader correctly:
- Discovers plugins in directories
- Imports plugin modules
- Validates PLUGIN_MANIFEST, MODEL_REQUIREMENTS, create_engine
- Validates api_version compatibility
- Validates capabilities
- Provides diagnostic messages for errors
- Rejects invalid plugins
"""
from __future__ import annotations
import sys
from pathlib import Path
import pytest
from abogen.tts_plugin.loader import (
HOST_API_VERSION,
PluginLoadError,
PluginLoadResult,
_check_api_version_compatibility,
_parse_api_version,
_validate_api_version,
_validate_capabilities,
_validate_manifest,
discover_plugins,
load_plugin,
load_plugin_from_dir,
)
from abogen.tts_plugin.manifest import (
EngineManifest,
ModelManifest,
PluginManifest,
)
# ──────────────────────────────────────────────────────────────
# Path fixtures
# ──────────────────────────────────────────────────────────────
@pytest.fixture
def plugins_dir() -> Path:
return Path(__file__).parent.parent / "plugins"
@pytest.fixture
def fake_plugin_dir(plugins_dir: Path) -> Path:
return plugins_dir / "fake_plugin"
@pytest.fixture
def missing_manifest_dir(plugins_dir: Path) -> Path:
return plugins_dir / "missing_manifest"
@pytest.fixture
def invalid_api_version_dir(plugins_dir: Path) -> Path:
return plugins_dir / "invalid_api_version"
@pytest.fixture
def invalid_capabilities_dir(plugins_dir: Path) -> Path:
return plugins_dir / "invalid_capabilities"
@pytest.fixture
def missing_create_engine_dir(plugins_dir: Path) -> Path:
return plugins_dir / "missing_create_engine"
@pytest.fixture
def import_error_dir(plugins_dir: Path) -> Path:
return plugins_dir / "import_error"
@pytest.fixture
def missing_model_requirements_dir(plugins_dir: Path) -> Path:
return plugins_dir / "missing_model_requirements"
# ──────────────────────────────────────────────────────────────
# Unit tests: _parse_api_version
# ──────────────────────────────────────────────────────────────
class TestParseApiVersion:
def test_valid_version(self) -> None:
assert _parse_api_version("1.0") == (1, 0)
assert _parse_api_version("2.5") == (2, 5)
assert _parse_api_version("10.20") == (10, 20)
def test_invalid_format(self) -> None:
assert _parse_api_version("1") is None
assert _parse_api_version("1.0.0") is None
assert _parse_api_version("abc") is None
assert _parse_api_version("") is None
assert _parse_api_version("1.x") is None
# ──────────────────────────────────────────────────────────────
# Unit tests: _check_api_version_compatibility
# ──────────────────────────────────────────────────────────────
class TestCheckApiVersionCompatibility:
def test_compatible_version(self) -> None:
assert _check_api_version_compatibility("1.0") is None
assert _check_api_version_compatibility("1.5") is None
def test_major_mismatch(self) -> None:
error = _check_api_version_compatibility("2.0")
assert error is not None
assert "major mismatch" in error
def test_invalid_format(self) -> None:
error = _check_api_version_compatibility("invalid")
assert error is not None
assert "Invalid api_version format" in error
# ──────────────────────────────────────────────────────────────
# Unit tests: _validate_manifest
# ──────────────────────────────────────────────────────────────
class TestValidateManifest:
def test_valid_manifest(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS: list = []
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert errors == []
def test_missing_manifest(self) -> None:
class FakeModule:
MODEL_REQUIREMENTS: list = []
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("Missing PLUGIN_MANIFEST" in e for e in errors)
def test_wrong_manifest_type(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = "not a manifest"
MODEL_REQUIREMENTS: list = []
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("PluginManifest instance" in e for e in errors)
def test_missing_model_requirements(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("Missing MODEL_REQUIREMENTS" in e for e in errors)
def test_wrong_model_requirements_type(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS = "not a list"
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("must be a list" in e for e in errors)
def test_invalid_model_requirements_item(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS = ["not a model manifest"]
create_engine = lambda *a, **kw: None
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("ModelManifest instance" in e for e in errors)
def test_missing_create_engine(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS: list = []
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("Missing create_engine" in e for e in errors)
def test_create_engine_not_callable(self) -> None:
class FakeModule:
PLUGIN_MANIFEST = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
MODEL_REQUIREMENTS: list = []
create_engine = "not callable"
errors = _validate_manifest(FakeModule(), Path("/tmp"))
assert any("must be callable" in e for e in errors)
# ──────────────────────────────────────────────────────────────
# Unit tests: _validate_capabilities
# ──────────────────────────────────────────────────────────────
class TestValidateCapabilities:
def test_valid_capabilities(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
capabilities=("voice_list", "preview"),
)
errors = _validate_capabilities(manifest)
assert errors == []
def test_unknown_capability(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
capabilities=("voice_list", "unknown_cap"),
)
errors = _validate_capabilities(manifest)
assert any("unknown_cap" in e for e in errors)
def test_empty_capabilities(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
capabilities=(),
)
errors = _validate_capabilities(manifest)
assert errors == []
# ──────────────────────────────────────────────────────────────
# Unit tests: _validate_api_version
# ──────────────────────────────────────────────────────────────
class TestValidateApiVersion:
def test_compatible(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="1.0", description="Test", author="Test",
)
errors = _validate_api_version(manifest)
assert errors == []
def test_incompatible(self) -> None:
manifest = PluginManifest(
id="test", name="Test", version="1.0.0",
api_version="2.0", description="Test", author="Test",
)
errors = _validate_api_version(manifest)
assert len(errors) > 0
# ──────────────────────────────────────────────────────────────
# Integration tests: load_plugin_from_dir
# ──────────────────────────────────────────────────────────────
class TestLoadPluginFromDir:
def test_load_valid_plugin(self, fake_plugin_dir: Path) -> None:
result = load_plugin_from_dir(fake_plugin_dir)
assert result.success is True
assert result.manifest is not None
assert result.manifest.id == "fake_plugin"
assert result.model_requirements is not None
assert result.create_engine is not None
assert result.module is not None
assert result.error is None
def test_plugin_satisfies_protocol(self, fake_plugin_dir: Path) -> None:
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
import logging
result = load_plugin_from_dir(fake_plugin_dir)
assert result.success is True
# Create engine using the loaded create_engine function
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda s, **kw: None, "post": lambda s, **kw: None})(),
)
engine = result.create_engine(ctx, None, __import__("abogen.tts_plugin.types", fromlist=["EngineConfig"]).EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
def test_nonexistent_directory(self, tmp_path: Path) -> None:
result = load_plugin_from_dir(tmp_path / "nonexistent")
assert result.success is False
assert result.error is not None
assert "does not exist" in result.error.errors[0]
def test_missing_init_file(self, tmp_path: Path) -> None:
plugin_dir = tmp_path / "no_init"
plugin_dir.mkdir()
result = load_plugin_from_dir(plugin_dir)
assert result.success is False
assert result.error is not None
assert "__init__.py" in result.error.errors[0]
def test_import_error(self, import_error_dir: Path) -> None:
result = load_plugin_from_dir(import_error_dir)
assert result.success is False
assert result.error is not None
assert "Failed to import" in result.error.errors[0]
# ──────────────────────────────────────────────────────────────
# Integration tests: invalid plugins
# ──────────────────────────────────────────────────────────────
class TestInvalidPlugins:
def test_missing_manifest(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.success is False
assert result.error is not None
assert any("Missing PLUGIN_MANIFEST" in e for e in result.error.errors)
def test_invalid_api_version(self, invalid_api_version_dir: Path) -> None:
result = load_plugin_from_dir(invalid_api_version_dir)
assert result.success is False
assert result.error is not None
assert any("major mismatch" in e for e in result.error.errors)
def test_invalid_capabilities(self, invalid_capabilities_dir: Path) -> None:
result = load_plugin_from_dir(invalid_capabilities_dir)
assert result.success is False
assert result.error is not None
assert any("Unknown capability" in e for e in result.error.errors)
def test_missing_create_engine(self, missing_create_engine_dir: Path) -> None:
result = load_plugin_from_dir(missing_create_engine_dir)
assert result.success is False
assert result.error is not None
assert any("Missing create_engine" in e for e in result.error.errors)
def test_missing_model_requirements(self, missing_model_requirements_dir: Path) -> None:
result = load_plugin_from_dir(missing_model_requirements_dir)
assert result.success is False
assert result.error is not None
assert any("Missing MODEL_REQUIREMENTS" in e for e in result.error.errors)
# ──────────────────────────────────────────────────────────────
# Integration tests: discover_plugins
# ──────────────────────────────────────────────────────────────
class TestDiscoverPlugins:
def test_discover_from_valid_dir(self, plugins_dir: Path) -> None:
results = discover_plugins([plugins_dir])
# Should find multiple plugins (valid and invalid)
assert len(results) > 0
def test_discover_includes_valid_plugin(self, plugins_dir: Path) -> None:
results = discover_plugins([plugins_dir])
valid = [r for r in results if r.success]
assert len(valid) >= 1
assert any(r.manifest and r.manifest.id == "fake_plugin" for r in valid)
def test_discover_includes_invalid_plugins(self, plugins_dir: Path) -> None:
results = discover_plugins([plugins_dir])
invalid = [r for r in results if not r.success]
assert len(invalid) >= 1
def test_discover_nonexistent_dir(self, tmp_path: Path) -> None:
results = discover_plugins([tmp_path / "nonexistent"])
assert results == []
def test_discover_multiple_dirs(self, plugins_dir: Path, tmp_path: Path) -> None:
results = discover_plugins([plugins_dir, tmp_path / "nonexistent"])
assert len(results) > 0
# ──────────────────────────────────────────────────────────────
# Diagnostic messages tests
# ──────────────────────────────────────────────────────────────
class TestDiagnosticMessages:
def test_error_contains_plugin_id(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.error is not None
assert result.error.plugin_id == "missing_manifest"
def test_error_contains_path(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.error is not None
assert result.error.path == missing_manifest_dir
def test_error_contains_messages(self, missing_manifest_dir: Path) -> None:
result = load_plugin_from_dir(missing_manifest_dir)
assert result.error is not None
assert len(result.error.errors) > 0
def test_multiple_errors(self, invalid_api_version_dir: Path) -> None:
# This plugin has multiple issues
result = load_plugin_from_dir(invalid_api_version_dir)
assert result.error is not None
# Should have at least the api_version error
assert len(result.error.errors) >= 1
# ──────────────────────────────────────────────────────────────
# No partial registration tests
# ──────────────────────────────────────────────────────────────
class TestNoPartialRegistration:
def test_invalid_plugin_no_manifest_attr(self, missing_manifest_dir: Path) -> None:
"""After failed load, module should not remain in sys.modules."""
result = load_plugin_from_dir(missing_manifest_dir)
assert result.success is False
# Module should not be registered
module_name = f"abogen.tts_plugin._loaded.missing_manifest"
assert module_name not in sys.modules
def test_import_error_no_registration(self, import_error_dir: Path) -> None:
"""After import error, module should not remain in sys.modules."""
result = load_plugin_from_dir(import_error_dir)
assert result.success is False
module_name = f"abogen.tts_plugin._loaded.import_error"
assert module_name not in sys.modules
tests/contracts/test_manifest_contract.py
+290 -290
View File
@@ -1,290 +1,290 @@
"""Contract tests for plugin manifest types.
These tests verify that manifest types satisfy the architectural requirements:
- All required fields are present
- api_version follows semver format
- capabilities are properly defined
- engine manifest describes the engine correctly
"""
import re
import pytest
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
EnumOption,
GpuRequirement,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
class TestPluginManifestContract:
"""Contract tests for PluginManifest."""
def test_required_fields(self) -> None:
manifest = PluginManifest(
id="test-plugin",
name="Test Plugin",
version="1.0.0",
api_version="1.0",
description="A test plugin",
author="Test Author",
)
assert manifest.id == "test-plugin"
assert manifest.name == "Test Plugin"
assert manifest.version == "1.0.0"
assert manifest.api_version == "1.0"
assert manifest.description == "A test plugin"
assert manifest.author == "Test Author"
def test_api_version_semver_format(self) -> None:
"""Architecture spec: api_version format is semver (MAJOR.MINOR)."""
valid_versions = ["1.0", "2.1", "10.5"]
for version in valid_versions:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version=version,
description="Test",
author="Test",
)
assert re.match(r"^\d+\.\d+$", manifest.api_version)
def test_capabilities_default_empty(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
)
assert manifest.capabilities == ()
def test_capabilities_tuple(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
capabilities=("voice_list", "preview"),
)
assert "voice_list" in manifest.capabilities
assert "preview" in manifest.capabilities
def test_requires_default(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
)
assert isinstance(manifest.requires, RequirementManifest)
def test_engine_default(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
)
assert isinstance(manifest.engine, EngineManifest)
class TestEngineManifestContract:
"""Contract tests for EngineManifest."""
def test_required_fields(self) -> None:
manifest = EngineManifest(
voiceSources=(
VoiceSourceManifest(id="builtin", name="Builtin", type="list"),
),
parameters=(
ParameterManifest(
id="speed", name="Speed", description="Speed", type="float", default=1.0
),
),
audioFormats=(AudioFormatManifest(mime="audio/wav", extension="wav"),),
)
assert len(manifest.voiceSources) == 1
assert len(manifest.parameters) == 1
assert len(manifest.audioFormats) == 1
def test_defaults_empty(self) -> None:
manifest = EngineManifest()
assert manifest.voiceSources == ()
assert manifest.parameters == ()
assert manifest.audioFormats == ()
class TestVoiceSourceManifestContract:
"""Contract tests for VoiceSourceManifest."""
def test_required_fields(self) -> None:
vs = VoiceSourceManifest(id="builtin", name="Builtin", type="list")
assert vs.id == "builtin"
assert vs.name == "Builtin"
assert vs.type == "list"
def test_valid_types(self) -> None:
valid_types = ["list", "speaker_id", "clone", "blend", "generate", "none"]
for vtype in valid_types:
vs = VoiceSourceManifest(id="test", name="Test", type=vtype)
assert vs.type == vtype
def test_config_optional(self) -> None:
vs = VoiceSourceManifest(id="test", name="Test", type="list")
assert vs.config is None
def test_config_any(self) -> None:
config = {"voices": ["af_nova", "af_sky"]}
vs = VoiceSourceManifest(id="test", name="Test", type="list", config=config)
assert vs.config == config
class TestVoiceManifestContract:
"""Contract tests for VoiceManifest."""
def test_required_fields(self) -> None:
v = VoiceManifest(id="af_nova", name="Nova")
assert v.id == "af_nova"
assert v.name == "Nova"
def test_tags_default_empty(self) -> None:
v = VoiceManifest(id="af_nova", name="Nova")
assert v.tags == ()
def test_tags_tuple(self) -> None:
v = VoiceManifest(id="af_nova", name="Nova", tags=("en", "female"))
assert "en" in v.tags
assert "female" in v.tags
class TestParameterManifestContract:
"""Contract tests for ParameterManifest."""
def test_required_fields(self) -> None:
p = ParameterManifest(
id="speed", name="Speed", description="Speech speed", type="float", default=1.0
)
assert p.id == "speed"
assert p.name == "Speed"
assert p.description == "Speech speed"
assert p.type == "float"
assert p.default == 1.0
def test_valid_types(self) -> None:
valid_types = ["float", "int", "string", "boolean", "enum"]
for ptype in valid_types:
p = ParameterManifest(
id="test", name="Test", description="Test", type=ptype, default=None
)
assert p.type == ptype
def test_optional_numeric_bounds(self) -> None:
p = ParameterManifest(
id="speed",
name="Speed",
description="Speed",
type="float",
default=1.0,
min=0.5,
max=2.0,
step=0.1,
)
assert p.min == 0.5
assert p.max == 2.0
assert p.step == 0.1
def test_enum_options(self) -> None:
options = (
EnumOption(value="low", label="Low"),
EnumOption(value="high", label="High"),
)
p = ParameterManifest(
id="quality",
name="Quality",
description="Quality",
type="enum",
default="low",
options=options,
)
assert len(p.options) == 2
assert p.options[0].value == "low"
class TestAudioFormatManifestContract:
"""Contract tests for AudioFormatManifest."""
def test_required_fields(self) -> None:
af = AudioFormatManifest(mime="audio/wav", extension="wav")
assert af.mime == "audio/wav"
assert af.extension == "wav"
class TestEnumOptionContract:
"""Contract tests for EnumOption."""
def test_required_fields(self) -> None:
opt = EnumOption(value="low", label="Low Quality")
assert opt.value == "low"
assert opt.label == "Low Quality"
class TestRequirementManifestContract:
"""Contract tests for RequirementManifest."""
def test_defaults(self) -> None:
req = RequirementManifest()
assert req.gpu is None
assert req.memory is None
assert req.internet is None
def test_with_gpu(self) -> None:
gpu = GpuRequirement(required=True, type="cuda", memory=8.0)
req = RequirementManifest(gpu=gpu)
assert req.gpu.required is True
assert req.gpu.type == "cuda"
assert req.gpu.memory == 8.0
def test_with_internet(self) -> None:
req = RequirementManifest(internet=True)
assert req.internet is True
class TestGpuRequirementContract:
"""Contract tests for GpuRequirement."""
def test_defaults(self) -> None:
gpu = GpuRequirement()
assert gpu.required is False
assert gpu.type is None
assert gpu.memory is None
def test_required_gpu(self) -> None:
gpu = GpuRequirement(required=True, type="cuda", memory=8.0)
assert gpu.required is True
class TestModelManifestContract:
"""Contract tests for ModelManifest."""
def test_required_fields(self) -> None:
m = ModelManifest(id="xtts_v2", name="XTTS v2", size="2GB")
assert m.id == "xtts_v2"
assert m.name == "XTTS v2"
assert m.size == "2GB"
"""Contract tests for plugin manifest types.
These tests verify that manifest types satisfy the architectural requirements:
- All required fields are present
- api_version follows semver format
- capabilities are properly defined
- engine manifest describes the engine correctly
"""
import re
import pytest
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
EnumOption,
GpuRequirement,
ModelManifest,
ParameterManifest,
PluginManifest,
RequirementManifest,
VoiceManifest,
VoiceSourceManifest,
)
class TestPluginManifestContract:
"""Contract tests for PluginManifest."""
def test_required_fields(self) -> None:
manifest = PluginManifest(
id="test-plugin",
name="Test Plugin",
version="1.0.0",
api_version="1.0",
description="A test plugin",
author="Test Author",
)
assert manifest.id == "test-plugin"
assert manifest.name == "Test Plugin"
assert manifest.version == "1.0.0"
assert manifest.api_version == "1.0"
assert manifest.description == "A test plugin"
assert manifest.author == "Test Author"
def test_api_version_semver_format(self) -> None:
"""Architecture spec: api_version format is semver (MAJOR.MINOR)."""
valid_versions = ["1.0", "2.1", "10.5"]
for version in valid_versions:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version=version,
description="Test",
author="Test",
)
assert re.match(r"^\d+\.\d+$", manifest.api_version)
def test_capabilities_default_empty(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
)
assert manifest.capabilities == ()
def test_capabilities_tuple(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
capabilities=("voice_list", "preview"),
)
assert "voice_list" in manifest.capabilities
assert "preview" in manifest.capabilities
def test_requires_default(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
)
assert isinstance(manifest.requires, RequirementManifest)
def test_engine_default(self) -> None:
manifest = PluginManifest(
id="test",
name="Test",
version="1.0.0",
api_version="1.0",
description="Test",
author="Test",
)
assert isinstance(manifest.engine, EngineManifest)
class TestEngineManifestContract:
"""Contract tests for EngineManifest."""
def test_required_fields(self) -> None:
manifest = EngineManifest(
voiceSources=(
VoiceSourceManifest(id="builtin", name="Builtin", type="list"),
),
parameters=(
ParameterManifest(
id="speed", name="Speed", description="Speed", type="float", default=1.0
),
),
audioFormats=(AudioFormatManifest(mime="audio/wav", extension="wav"),),
)
assert len(manifest.voiceSources) == 1
assert len(manifest.parameters) == 1
assert len(manifest.audioFormats) == 1
def test_defaults_empty(self) -> None:
manifest = EngineManifest()
assert manifest.voiceSources == ()
assert manifest.parameters == ()
assert manifest.audioFormats == ()
class TestVoiceSourceManifestContract:
"""Contract tests for VoiceSourceManifest."""
def test_required_fields(self) -> None:
vs = VoiceSourceManifest(id="builtin", name="Builtin", type="list")
assert vs.id == "builtin"
assert vs.name == "Builtin"
assert vs.type == "list"
def test_valid_types(self) -> None:
valid_types = ["list", "speaker_id", "clone", "blend", "generate", "none"]
for vtype in valid_types:
vs = VoiceSourceManifest(id="test", name="Test", type=vtype)
assert vs.type == vtype
def test_config_optional(self) -> None:
vs = VoiceSourceManifest(id="test", name="Test", type="list")
assert vs.config is None
def test_config_any(self) -> None:
config = {"voices": ["af_nova", "af_sky"]}
vs = VoiceSourceManifest(id="test", name="Test", type="list", config=config)
assert vs.config == config
class TestVoiceManifestContract:
"""Contract tests for VoiceManifest."""
def test_required_fields(self) -> None:
v = VoiceManifest(id="af_nova", name="Nova")
assert v.id == "af_nova"
assert v.name == "Nova"
def test_tags_default_empty(self) -> None:
v = VoiceManifest(id="af_nova", name="Nova")
assert v.tags == ()
def test_tags_tuple(self) -> None:
v = VoiceManifest(id="af_nova", name="Nova", tags=("en", "female"))
assert "en" in v.tags
assert "female" in v.tags
class TestParameterManifestContract:
"""Contract tests for ParameterManifest."""
def test_required_fields(self) -> None:
p = ParameterManifest(
id="speed", name="Speed", description="Speech speed", type="float", default=1.0
)
assert p.id == "speed"
assert p.name == "Speed"
assert p.description == "Speech speed"
assert p.type == "float"
assert p.default == 1.0
def test_valid_types(self) -> None:
valid_types = ["float", "int", "string", "boolean", "enum"]
for ptype in valid_types:
p = ParameterManifest(
id="test", name="Test", description="Test", type=ptype, default=None
)
assert p.type == ptype
def test_optional_numeric_bounds(self) -> None:
p = ParameterManifest(
id="speed",
name="Speed",
description="Speed",
type="float",
default=1.0,
min=0.5,
max=2.0,
step=0.1,
)
assert p.min == 0.5
assert p.max == 2.0
assert p.step == 0.1
def test_enum_options(self) -> None:
options = (
EnumOption(value="low", label="Low"),
EnumOption(value="high", label="High"),
)
p = ParameterManifest(
id="quality",
name="Quality",
description="Quality",
type="enum",
default="low",
options=options,
)
assert len(p.options) == 2
assert p.options[0].value == "low"
class TestAudioFormatManifestContract:
"""Contract tests for AudioFormatManifest."""
def test_required_fields(self) -> None:
af = AudioFormatManifest(mime="audio/wav", extension="wav")
assert af.mime == "audio/wav"
assert af.extension == "wav"
class TestEnumOptionContract:
"""Contract tests for EnumOption."""
def test_required_fields(self) -> None:
opt = EnumOption(value="low", label="Low Quality")
assert opt.value == "low"
assert opt.label == "Low Quality"
class TestRequirementManifestContract:
"""Contract tests for RequirementManifest."""
def test_defaults(self) -> None:
req = RequirementManifest()
assert req.gpu is None
assert req.memory is None
assert req.internet is None
def test_with_gpu(self) -> None:
gpu = GpuRequirement(required=True, type="cuda", memory=8.0)
req = RequirementManifest(gpu=gpu)
assert req.gpu.required is True
assert req.gpu.type == "cuda"
assert req.gpu.memory == 8.0
def test_with_internet(self) -> None:
req = RequirementManifest(internet=True)
assert req.internet is True
class TestGpuRequirementContract:
"""Contract tests for GpuRequirement."""
def test_defaults(self) -> None:
gpu = GpuRequirement()
assert gpu.required is False
assert gpu.type is None
assert gpu.memory is None
def test_required_gpu(self) -> None:
gpu = GpuRequirement(required=True, type="cuda", memory=8.0)
assert gpu.required is True
class TestModelManifestContract:
"""Contract tests for ModelManifest."""
def test_required_fields(self) -> None:
m = ModelManifest(id="xtts_v2", name="XTTS v2", size="2GB")
assert m.id == "xtts_v2"
assert m.name == "XTTS v2"
assert m.size == "2GB"
tests/contracts/test_plugin_contract.py
+146 -146
View File
@@ -1,146 +1,146 @@
"""Contract tests for plugin contract.
These tests verify that plugin modules satisfy the architectural requirements:
- Must export PLUGIN_MANIFEST: PluginManifest
- Must export MODEL_REQUIREMENTS: list[ModelManifest]
- Must export create_engine: Callable[[HostContext, Path | None, EngineConfig], Engine]
- create_engine() must be atomic
"""
import logging
from pathlib import Path
from typing import Any
import pytest
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import EngineManifest, ModelManifest, PluginManifest
from abogen.tts_plugin.plugin import Plugin
from abogen.tts_plugin.types import EngineConfig
from .conftest import FakeEngine
class FakePluginModule:
"""Stub plugin module that satisfies the plugin contract."""
PLUGIN_MANIFEST = PluginManifest(
id="fake-plugin",
name="Fake Plugin",
version="1.0.0",
api_version="1.0",
description="A fake plugin for testing",
author="Test Author",
capabilities=(),
engine=EngineManifest(),
)
MODEL_REQUIREMENTS: list[ModelManifest] = []
@staticmethod
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
return FakeEngine()
class TestPluginProtocolContract:
"""Contract tests for the Plugin protocol."""
def test_plugin_is_protocol(self) -> None:
assert hasattr(Plugin, "__protocol_attrs__")
class TestPluginExportsContract:
"""Contract tests for required plugin exports."""
def test_plugin_has_plugin_manifest(self) -> None:
"""Architecture spec: Plugin must export PLUGIN_MANIFEST."""
assert hasattr(FakePluginModule, "PLUGIN_MANIFEST")
assert isinstance(FakePluginModule.PLUGIN_MANIFEST, PluginManifest)
def test_plugin_has_model_requirements(self) -> None:
"""Architecture spec: Plugin must export MODEL_REQUIREMENTS."""
assert hasattr(FakePluginModule, "MODEL_REQUIREMENTS")
assert isinstance(FakePluginModule.MODEL_REQUIREMENTS, list)
def test_plugin_has_create_engine(self) -> None:
"""Architecture spec: Plugin must export create_engine."""
assert hasattr(FakePluginModule, "create_engine")
assert callable(FakePluginModule.create_engine)
def test_plugin_manifest_required_fields(self) -> None:
"""Architecture spec: PluginManifest has required fields."""
manifest = FakePluginModule.PLUGIN_MANIFEST
assert manifest.id
assert manifest.name
assert manifest.version
assert manifest.api_version
assert manifest.description
assert manifest.author
def test_plugin_manifest_capabilities_is_tuple(self) -> None:
manifest = FakePluginModule.PLUGIN_MANIFEST
assert isinstance(manifest.capabilities, tuple)
class TestCreateEngineContract:
"""Contract tests for create_engine() function."""
def test_create_engine_returns_engine(self) -> None:
"""Architecture spec: create_engine() returns Engine."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, None, EngineConfig())
assert isinstance(engine, Engine)
def test_create_engine_atomic(self) -> None:
"""Architecture spec: create_engine() is atomic (all-or-nothing)."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
def test_create_engine_with_none_model_path(self) -> None:
"""Architecture spec: model_path can be None for cloud/no-model engines."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
def test_create_engine_with_model_path(self) -> None:
"""Architecture spec: model_path is Path | None."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, Path("/models/test"), EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
class TestModelRequirementsContract:
"""Contract tests for MODEL_REQUIREMENTS."""
def test_model_requirements_is_list(self) -> None:
assert isinstance(FakePluginModule.MODEL_REQUIREMENTS, list)
def test_model_requirements_contains_model_manifests(self) -> None:
"""If non-empty, each item must be a ModelManifest."""
for req in FakePluginModule.MODEL_REQUIREMENTS:
assert isinstance(req, ModelManifest)
"""Contract tests for plugin contract.
These tests verify that plugin modules satisfy the architectural requirements:
- Must export PLUGIN_MANIFEST: PluginManifest
- Must export MODEL_REQUIREMENTS: list[ModelManifest]
- Must export create_engine: Callable[[HostContext, Path | None, EngineConfig], Engine]
- create_engine() must be atomic
"""
import logging
from pathlib import Path
from typing import Any
import pytest
from abogen.tts_plugin.engine import Engine
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import EngineManifest, ModelManifest, PluginManifest
from abogen.tts_plugin.plugin import Plugin
from abogen.tts_plugin.types import EngineConfig
from .conftest import FakeEngine
class FakePluginModule:
"""Stub plugin module that satisfies the plugin contract."""
PLUGIN_MANIFEST = PluginManifest(
id="fake-plugin",
name="Fake Plugin",
version="1.0.0",
api_version="1.0",
description="A fake plugin for testing",
author="Test Author",
capabilities=(),
engine=EngineManifest(),
)
MODEL_REQUIREMENTS: list[ModelManifest] = []
@staticmethod
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
return FakeEngine()
class TestPluginProtocolContract:
"""Contract tests for the Plugin protocol."""
def test_plugin_is_protocol(self) -> None:
assert hasattr(Plugin, "__protocol_attrs__")
class TestPluginExportsContract:
"""Contract tests for required plugin exports."""
def test_plugin_has_plugin_manifest(self) -> None:
"""Architecture spec: Plugin must export PLUGIN_MANIFEST."""
assert hasattr(FakePluginModule, "PLUGIN_MANIFEST")
assert isinstance(FakePluginModule.PLUGIN_MANIFEST, PluginManifest)
def test_plugin_has_model_requirements(self) -> None:
"""Architecture spec: Plugin must export MODEL_REQUIREMENTS."""
assert hasattr(FakePluginModule, "MODEL_REQUIREMENTS")
assert isinstance(FakePluginModule.MODEL_REQUIREMENTS, list)
def test_plugin_has_create_engine(self) -> None:
"""Architecture spec: Plugin must export create_engine."""
assert hasattr(FakePluginModule, "create_engine")
assert callable(FakePluginModule.create_engine)
def test_plugin_manifest_required_fields(self) -> None:
"""Architecture spec: PluginManifest has required fields."""
manifest = FakePluginModule.PLUGIN_MANIFEST
assert manifest.id
assert manifest.name
assert manifest.version
assert manifest.api_version
assert manifest.description
assert manifest.author
def test_plugin_manifest_capabilities_is_tuple(self) -> None:
manifest = FakePluginModule.PLUGIN_MANIFEST
assert isinstance(manifest.capabilities, tuple)
class TestCreateEngineContract:
"""Contract tests for create_engine() function."""
def test_create_engine_returns_engine(self) -> None:
"""Architecture spec: create_engine() returns Engine."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, None, EngineConfig())
assert isinstance(engine, Engine)
def test_create_engine_atomic(self) -> None:
"""Architecture spec: create_engine() is atomic (all-or-nothing)."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
def test_create_engine_with_none_model_path(self) -> None:
"""Architecture spec: model_path can be None for cloud/no-model engines."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
def test_create_engine_with_model_path(self) -> None:
"""Architecture spec: model_path is Path | None."""
ctx = HostContext(
config_dir=Path("/tmp/test"),
logger=logging.getLogger("test"),
http_client=type("FakeClient", (), {"get": lambda self, **kw: None, "post": lambda self, **kw: None})(),
)
engine = FakePluginModule.create_engine(ctx, Path("/models/test"), EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
class TestModelRequirementsContract:
"""Contract tests for MODEL_REQUIREMENTS."""
def test_model_requirements_is_list(self) -> None:
assert isinstance(FakePluginModule.MODEL_REQUIREMENTS, list)
def test_model_requirements_contains_model_manifests(self) -> None:
"""If non-empty, each item must be a ModelManifest."""
for req in FakePluginModule.MODEL_REQUIREMENTS:
assert isinstance(req, ModelManifest)
tests/contracts/test_plugin_manager_contract.py
+274 -274
View File
@@ -1,274 +1,274 @@
"""Integration tests for Plugin Manager and direct utility functions."""
import pytest
from unittest.mock import MagicMock, patch
from abogen.tts_plugin.plugin_manager import PluginManager, get_plugin_manager, reset_plugin_manager
from abogen.tts_plugin.utils import Pipeline, create_pipeline
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.types import SynthesisRequest, SynthesizedAudio, AudioFormat
class FakeEngine:
"""Fake Engine for testing."""
def __init__(self, **kwargs):
self.kwargs = kwargs
self._disposed = False
def createSession(self):
return FakeEngineSession()
def dispose(self):
self._disposed = True
@property
def manifest(self):
return MagicMock()
class FakeEngineSession:
"""Fake EngineSession for testing."""
def __init__(self):
self._disposed = False
def synthesize(self, request):
# Return fake audio
import numpy as np
from abogen.tts_plugin.types import AudioFormat, Duration, SynthesizedAudio
audio = np.zeros(1000, dtype=np.float32)
return SynthesizedAudio(
data=audio.tobytes(),
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.04167), # 1000 samples / 24000 Hz
)
def dispose(self):
self._disposed = True
class TestPluginManager:
"""Test PluginManager functionality."""
def test_plugin_manager_creation(self):
"""PluginManager can be created."""
manager = PluginManager()
assert manager is not None
def test_plugin_manager_list_discovers_plugins(self):
"""PluginManager discovers plugins from plugins directory."""
manager = PluginManager()
manager.discover("plugins")
plugins = manager.list_plugins()
# Should discover kokoro plugin if it exists
assert isinstance(plugins, list)
def test_plugin_manager_has_plugin_after_discover(self):
"""PluginManager reports plugins after discovery."""
manager = PluginManager()
manager.discover("plugins")
# kokoro plugin should be discovered if plugins/kokoro exists
# This is expected behavior
assert isinstance(manager._plugins, dict)
def test_plugin_manager_get_plugin_after_discover(self):
"""PluginManager returns plugin info after discovery."""
manager = PluginManager()
manager.discover("plugins")
# kokoro plugin should be discovered if plugins/kokoro exists
assert isinstance(manager._plugins, dict)
def test_plugin_manager_create_engine_not_found(self):
"""PluginManager raises KeyError for unknown plugins."""
manager = PluginManager()
with pytest.raises(KeyError, match="Plugin not found"):
manager.create_engine("nonexistent")
def test_plugin_manager_discover_with_empty_dir(self):
"""PluginManager handles missing plugins directory."""
manager = PluginManager()
manager.discover("/nonexistent/path")
plugins = manager.list_plugins()
assert plugins == []
def test_global_plugin_manager_singleton(self):
"""Global PluginManager is a singleton."""
reset_plugin_manager()
manager1 = get_plugin_manager()
manager2 = get_plugin_manager()
assert manager1 is manager2
reset_plugin_manager()
def test_reset_plugin_manager(self):
"""reset_plugin_manager clears the singleton."""
manager1 = get_plugin_manager()
reset_plugin_manager()
manager2 = get_plugin_manager()
assert manager1 is not manager2
reset_plugin_manager()
class TestPipeline:
"""Test Pipeline functionality."""
def test_pipeline_creation(self):
"""Pipeline can be created."""
engine = FakeEngine()
backend = Pipeline(engine)
assert backend is not None
def test_pipeline_callable(self):
"""Pipeline is callable like old TTSBackend."""
engine = FakeEngine()
backend = Pipeline(engine)
# Should be callable
assert callable(backend)
def test_pipeline_synthesize(self):
"""Pipeline can synthesize text."""
engine = FakeEngine()
backend = Pipeline(engine)
# Call the backend
segments = list(backend("Hello world", voice="default", speed=1.0))
# Should return at least one segment
assert len(segments) >= 1
# Segment should have graphemes and audio
segment = segments[0]
assert hasattr(segment, "graphemes")
assert hasattr(segment, "audio")
assert segment.graphemes == "Hello world"
def test_pipeline_dispose(self):
"""Pipeline can be disposed."""
engine = FakeEngine()
backend = Pipeline(engine)
# Create a session by calling
list(backend("test"))
# Dispose should not raise
backend.dispose()
# Double dispose should be safe
backend.dispose()
class TestCreatePipelineCompat:
"""Test create_pipeline utility function."""
def test_create_pipeline_returns_callable(self):
"""create_pipeline returns a callable backend."""
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
# Mock the plugin manager
with patch("abogen.tts_plugin.utils.get_plugin_manager") as mock_get_manager:
mock_manager = MagicMock()
mock_get_manager.return_value = mock_manager
mock_engine = FakeEngine()
mock_manager.create_engine.return_value = mock_engine
backend = create_pipeline("kokoro", lang_code="a", device="cpu")
assert callable(backend)
mock_manager.create_engine.assert_called_once()
call_args = mock_manager.create_engine.call_args
assert call_args.args[0] == "kokoro"
assert isinstance(call_args.kwargs["context"], HostContext)
assert call_args.kwargs["model_path"] is None
assert isinstance(call_args.kwargs["config"], EngineConfig)
assert call_args.kwargs["config"].device == "cpu"
assert call_args.kwargs["config"].lang_code == "a"
def test_create_pipeline_raises_for_unknown_plugin(self):
"""create_pipeline raises KeyError for unknown plugins."""
with patch("abogen.tts_plugin.utils.get_plugin_manager") as mock_get_manager:
mock_manager = MagicMock()
mock_get_manager.return_value = mock_manager
mock_manager.create_engine.side_effect = KeyError("Plugin not found")
with pytest.raises(KeyError):
create_pipeline("nonexistent")
class TestPluginManagerWithFakePlugins:
"""Test PluginManager with fake plugin loading."""
def test_plugin_manager_create_engine_from_plugin(self):
"""PluginManager creates engine from loaded plugin."""
manager = PluginManager()
# Manually add a fake plugin
def fake_create_engine(**kwargs):
return FakeEngine(**kwargs)
manager._plugins["fake"] = {
"manifest": MagicMock(),
"create_engine": fake_create_engine,
}
manager._loaded = True
# Create engine
engine = manager.create_engine("fake", param="value")
assert isinstance(engine, FakeEngine)
assert engine.kwargs == {"param": "value"}
def test_plugin_manager_get_or_create_engine(self):
"""PluginManager caches engines."""
manager = PluginManager()
call_count = 0
def fake_create_engine(**kwargs):
nonlocal call_count
call_count += 1
return FakeEngine(**kwargs)
manager._plugins["fake"] = {
"manifest": MagicMock(),
"create_engine": fake_create_engine,
}
manager._loaded = True
# Get engine twice
engine1 = manager.get_or_create_engine("fake")
engine2 = manager.get_or_create_engine("fake")
# Should be same instance
assert engine1 is engine2
assert call_count == 1
def test_plugin_manager_dispose_all(self):
"""PluginManager disposes all cached engines."""
manager = PluginManager()
def fake_create_engine(**kwargs):
return FakeEngine(**kwargs)
manager._plugins["fake"] = {
"manifest": MagicMock(),
"create_engine": fake_create_engine,
}
manager._loaded = True
# Create and cache engines
engine1 = manager.get_or_create_engine("fake")
engine2 = manager.get_or_create_engine("fake")
# Dispose all
manager.dispose_all()
# Engines should be disposed
assert engine1._disposed is True
assert engine2._disposed is True
# Cache should be empty
assert len(manager._engines) == 0
"""Integration tests for Plugin Manager and direct utility functions."""
import pytest
from unittest.mock import MagicMock, patch
from abogen.tts_plugin.plugin_manager import PluginManager, get_plugin_manager, reset_plugin_manager
from abogen.tts_plugin.utils import Pipeline, create_pipeline
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.types import SynthesisRequest, SynthesizedAudio, AudioFormat
class FakeEngine:
"""Fake Engine for testing."""
def __init__(self, **kwargs):
self.kwargs = kwargs
self._disposed = False
def createSession(self):
return FakeEngineSession()
def dispose(self):
self._disposed = True
@property
def manifest(self):
return MagicMock()
class FakeEngineSession:
"""Fake EngineSession for testing."""
def __init__(self):
self._disposed = False
def synthesize(self, request):
# Return fake audio
import numpy as np
from abogen.tts_plugin.types import AudioFormat, Duration, SynthesizedAudio
audio = np.zeros(1000, dtype=np.float32)
return SynthesizedAudio(
data=audio.tobytes(),
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=0.04167), # 1000 samples / 24000 Hz
)
def dispose(self):
self._disposed = True
class TestPluginManager:
"""Test PluginManager functionality."""
def test_plugin_manager_creation(self):
"""PluginManager can be created."""
manager = PluginManager()
assert manager is not None
def test_plugin_manager_list_discovers_plugins(self):
"""PluginManager discovers plugins from plugins directory."""
manager = PluginManager()
manager.discover("plugins")
plugins = manager.list_plugins()
# Should discover kokoro plugin if it exists
assert isinstance(plugins, list)
def test_plugin_manager_has_plugin_after_discover(self):
"""PluginManager reports plugins after discovery."""
manager = PluginManager()
manager.discover("plugins")
# kokoro plugin should be discovered if plugins/kokoro exists
# This is expected behavior
assert isinstance(manager._plugins, dict)
def test_plugin_manager_get_plugin_after_discover(self):
"""PluginManager returns plugin info after discovery."""
manager = PluginManager()
manager.discover("plugins")
# kokoro plugin should be discovered if plugins/kokoro exists
assert isinstance(manager._plugins, dict)
def test_plugin_manager_create_engine_not_found(self):
"""PluginManager raises KeyError for unknown plugins."""
manager = PluginManager()
with pytest.raises(KeyError, match="Plugin not found"):
manager.create_engine("nonexistent")
def test_plugin_manager_discover_with_empty_dir(self):
"""PluginManager handles missing plugins directory."""
manager = PluginManager()
manager.discover("/nonexistent/path")
plugins = manager.list_plugins()
assert plugins == []
def test_global_plugin_manager_singleton(self):
"""Global PluginManager is a singleton."""
reset_plugin_manager()
manager1 = get_plugin_manager()
manager2 = get_plugin_manager()
assert manager1 is manager2
reset_plugin_manager()
def test_reset_plugin_manager(self):
"""reset_plugin_manager clears the singleton."""
manager1 = get_plugin_manager()
reset_plugin_manager()
manager2 = get_plugin_manager()
assert manager1 is not manager2
reset_plugin_manager()
class TestPipeline:
"""Test Pipeline functionality."""
def test_pipeline_creation(self):
"""Pipeline can be created."""
engine = FakeEngine()
backend = Pipeline(engine)
assert backend is not None
def test_pipeline_callable(self):
"""Pipeline is callable like old TTSBackend."""
engine = FakeEngine()
backend = Pipeline(engine)
# Should be callable
assert callable(backend)
def test_pipeline_synthesize(self):
"""Pipeline can synthesize text."""
engine = FakeEngine()
backend = Pipeline(engine)
# Call the backend
segments = list(backend("Hello world", voice="default", speed=1.0))
# Should return at least one segment
assert len(segments) >= 1
# Segment should have graphemes and audio
segment = segments[0]
assert hasattr(segment, "graphemes")
assert hasattr(segment, "audio")
assert segment.graphemes == "Hello world"
def test_pipeline_dispose(self):
"""Pipeline can be disposed."""
engine = FakeEngine()
backend = Pipeline(engine)
# Create a session by calling
list(backend("test"))
# Dispose should not raise
backend.dispose()
# Double dispose should be safe
backend.dispose()
class TestCreatePipelineCompat:
"""Test create_pipeline utility function."""
def test_create_pipeline_returns_callable(self):
"""create_pipeline returns a callable backend."""
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.types import EngineConfig
# Mock the plugin manager
with patch("abogen.tts_plugin.utils.get_plugin_manager") as mock_get_manager:
mock_manager = MagicMock()
mock_get_manager.return_value = mock_manager
mock_engine = FakeEngine()
mock_manager.create_engine.return_value = mock_engine
backend = create_pipeline("kokoro", lang_code="a", device="cpu")
assert callable(backend)
mock_manager.create_engine.assert_called_once()
call_args = mock_manager.create_engine.call_args
assert call_args.args[0] == "kokoro"
assert isinstance(call_args.kwargs["context"], HostContext)
assert call_args.kwargs["model_path"] is None
assert isinstance(call_args.kwargs["config"], EngineConfig)
assert call_args.kwargs["config"].device == "cpu"
assert call_args.kwargs["config"].lang_code == "a"
def test_create_pipeline_raises_for_unknown_plugin(self):
"""create_pipeline raises KeyError for unknown plugins."""
with patch("abogen.tts_plugin.utils.get_plugin_manager") as mock_get_manager:
mock_manager = MagicMock()
mock_get_manager.return_value = mock_manager
mock_manager.create_engine.side_effect = KeyError("Plugin not found")
with pytest.raises(KeyError):
create_pipeline("nonexistent")
class TestPluginManagerWithFakePlugins:
"""Test PluginManager with fake plugin loading."""
def test_plugin_manager_create_engine_from_plugin(self):
"""PluginManager creates engine from loaded plugin."""
manager = PluginManager()
# Manually add a fake plugin
def fake_create_engine(**kwargs):
return FakeEngine(**kwargs)
manager._plugins["fake"] = {
"manifest": MagicMock(),
"create_engine": fake_create_engine,
}
manager._loaded = True
# Create engine
engine = manager.create_engine("fake", param="value")
assert isinstance(engine, FakeEngine)
assert engine.kwargs == {"param": "value"}
def test_plugin_manager_get_or_create_engine(self):
"""PluginManager caches engines."""
manager = PluginManager()
call_count = 0
def fake_create_engine(**kwargs):
nonlocal call_count
call_count += 1
return FakeEngine(**kwargs)
manager._plugins["fake"] = {
"manifest": MagicMock(),
"create_engine": fake_create_engine,
}
manager._loaded = True
# Get engine twice
engine1 = manager.get_or_create_engine("fake")
engine2 = manager.get_or_create_engine("fake")
# Should be same instance
assert engine1 is engine2
assert call_count == 1
def test_plugin_manager_dispose_all(self):
"""PluginManager disposes all cached engines."""
manager = PluginManager()
def fake_create_engine(**kwargs):
return FakeEngine(**kwargs)
manager._plugins["fake"] = {
"manifest": MagicMock(),
"create_engine": fake_create_engine,
}
manager._loaded = True
# Create and cache engines
engine1 = manager.get_or_create_engine("fake")
engine2 = manager.get_or_create_engine("fake")
# Dispose all
manager.dispose_all()
# Engines should be disposed
assert engine1._disposed is True
assert engine2._disposed is True
# Cache should be empty
assert len(manager._engines) == 0
tests/contracts/test_session_contract.py
+135 -135
View File
@@ -1,135 +1,135 @@
"""Contract tests for EngineSession protocol.
These tests verify that EngineSession implementations satisfy the architectural requirements:
- synthesize() returns SynthesizedAudio
- dispose() is idempotent
- After dispose(), synthesize() raises EngineError
- Session remains usable after synthesize() failure
"""
import pytest
from abogen.tts_plugin.engine import EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
from .conftest import FakeEngineSession
class TestEngineSessionProtocolContract:
"""Contract tests for the EngineSession protocol itself."""
def test_engine_session_is_protocol(self) -> None:
assert hasattr(EngineSession, "__protocol_attrs__")
def test_fake_session_satisfies_protocol(self) -> None:
session = FakeEngineSession()
assert isinstance(session, EngineSession)
class TestSessionSynthesizeContract:
"""Contract tests for EngineSession.synthesize()."""
def test_synthesize_returns_synthesized_audio(self) -> None:
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result, SynthesizedAudio)
def test_synthesize_returns_valid_audio_data(self) -> None:
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
assert len(result.data) > 0
assert isinstance(result.format, AudioFormat)
assert isinstance(result.duration, Duration)
class TestSessionDisposeContract:
"""Contract tests for EngineSession.dispose()."""
def test_dispose_is_idempotent(self) -> None:
"""Architecture spec: dispose() is idempotent."""
session = FakeEngineSession()
session.dispose()
session.dispose() # Should not raise
def test_dispose_never_raises(self) -> None:
"""Architecture spec: dispose() never raises exceptions."""
session = FakeEngineSession()
session.dispose() # Should not raise
class TestSessionAfterDisposeContract:
"""Contract tests for behavior after dispose()."""
def test_synthesize_after_dispose_raises(self) -> None:
"""Architecture spec: After dispose(), all methods except dispose() raise EngineError."""
session = FakeEngineSession()
session.dispose()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError):
session.synthesize(request)
class TestSessionLifecycleContract:
"""Contract tests for EngineSession lifecycle."""
def test_full_lifecycle(self) -> None:
"""Test complete session lifecycle: create -> synthesize -> dispose."""
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# Synthesize
result = session.synthesize(request)
assert isinstance(result, SynthesizedAudio)
# Dispose
session.dispose()
def test_multiple_synthesize_before_dispose(self) -> None:
"""Architecture spec: Session remains usable after synthesize() failure."""
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# Multiple synthesize calls
result1 = session.synthesize(request)
result2 = session.synthesize(request)
assert isinstance(result1, SynthesizedAudio)
assert isinstance(result2, SynthesizedAudio)
# Dispose
session.dispose()
"""Contract tests for EngineSession protocol.
These tests verify that EngineSession implementations satisfy the architectural requirements:
- synthesize() returns SynthesizedAudio
- dispose() is idempotent
- After dispose(), synthesize() raises EngineError
- Session remains usable after synthesize() failure
"""
import pytest
from abogen.tts_plugin.engine import EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
from .conftest import FakeEngineSession
class TestEngineSessionProtocolContract:
"""Contract tests for the EngineSession protocol itself."""
def test_engine_session_is_protocol(self) -> None:
assert hasattr(EngineSession, "__protocol_attrs__")
def test_fake_session_satisfies_protocol(self) -> None:
session = FakeEngineSession()
assert isinstance(session, EngineSession)
class TestSessionSynthesizeContract:
"""Contract tests for EngineSession.synthesize()."""
def test_synthesize_returns_synthesized_audio(self) -> None:
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result, SynthesizedAudio)
def test_synthesize_returns_valid_audio_data(self) -> None:
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
assert len(result.data) > 0
assert isinstance(result.format, AudioFormat)
assert isinstance(result.duration, Duration)
class TestSessionDisposeContract:
"""Contract tests for EngineSession.dispose()."""
def test_dispose_is_idempotent(self) -> None:
"""Architecture spec: dispose() is idempotent."""
session = FakeEngineSession()
session.dispose()
session.dispose() # Should not raise
def test_dispose_never_raises(self) -> None:
"""Architecture spec: dispose() never raises exceptions."""
session = FakeEngineSession()
session.dispose() # Should not raise
class TestSessionAfterDisposeContract:
"""Contract tests for behavior after dispose()."""
def test_synthesize_after_dispose_raises(self) -> None:
"""Architecture spec: After dispose(), all methods except dispose() raise EngineError."""
session = FakeEngineSession()
session.dispose()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(EngineError):
session.synthesize(request)
class TestSessionLifecycleContract:
"""Contract tests for EngineSession lifecycle."""
def test_full_lifecycle(self) -> None:
"""Test complete session lifecycle: create -> synthesize -> dispose."""
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# Synthesize
result = session.synthesize(request)
assert isinstance(result, SynthesizedAudio)
# Dispose
session.dispose()
def test_multiple_synthesize_before_dispose(self) -> None:
"""Architecture spec: Session remains usable after synthesize() failure."""
session = FakeEngineSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
# Multiple synthesize calls
result1 = session.synthesize(request)
result2 = session.synthesize(request)
assert isinstance(result1, SynthesizedAudio)
assert isinstance(result2, SynthesizedAudio)
# Dispose
session.dispose()
tests/contracts/test_types_contract.py
+244 -244
View File
@@ -1,244 +1,244 @@
"""Contract tests for core domain value objects.
These tests verify that value objects satisfy the architectural requirements:
- Frozen (immutable) dataclasses
- Correct field definitions
- Proper equality behavior
"""
import pytest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class TestAudioFormatContract:
"""Contract tests for AudioFormat value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(AudioFormat, "__dataclass_params__")
assert AudioFormat.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
af = AudioFormat(mime="audio/wav", extension="wav")
assert af.mime == "audio/wav"
assert af.extension == "wav"
def test_immutability(self) -> None:
af = AudioFormat(mime="audio/wav", extension="wav")
with pytest.raises(AttributeError):
af.mime = "audio/mpeg" # type: ignore[misc]
def test_equality(self) -> None:
af1 = AudioFormat(mime="audio/wav", extension="wav")
af2 = AudioFormat(mime="audio/wav", extension="wav")
assert af1 == af2
def test_inequality(self) -> None:
af1 = AudioFormat(mime="audio/wav", extension="wav")
af2 = AudioFormat(mime="audio/mpeg", extension="mp3")
assert af1 != af2
def test_hashable(self) -> None:
af = AudioFormat(mime="audio/wav", extension="wav")
assert hash(af) == hash(AudioFormat(mime="audio/wav", extension="wav"))
class TestDurationContract:
"""Contract tests for Duration value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(Duration, "__dataclass_params__")
assert Duration.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
d = Duration(seconds=1.5)
assert d.seconds == 1.5
def test_immutability(self) -> None:
d = Duration(seconds=1.0)
with pytest.raises(AttributeError):
d.seconds = 2.0 # type: ignore[misc]
def test_equality(self) -> None:
d1 = Duration(seconds=1.0)
d2 = Duration(seconds=1.0)
assert d1 == d2
class TestVoiceSelectionContract:
"""Contract tests for VoiceSelection value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(VoiceSelection, "__dataclass_params__")
assert VoiceSelection.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
vs = VoiceSelection(source="builtin", key="af_nova")
assert vs.source == "builtin"
assert vs.key == "af_nova"
def test_payload_default_none(self) -> None:
vs = VoiceSelection(source="builtin", key="af_nova")
assert vs.payload is None
def test_payload_optional(self) -> None:
vs = VoiceSelection(source="clone", key="my_voice", payload=b"audio_data")
assert vs.payload == b"audio_data"
def test_immutability(self) -> None:
vs = VoiceSelection(source="builtin", key="af_nova")
with pytest.raises(AttributeError):
vs.source = "other" # type: ignore[misc]
class TestParameterValuesContract:
"""Contract tests for ParameterValues value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(ParameterValues, "__dataclass_params__")
assert ParameterValues.__dataclass_params__.frozen is True
def test_default_empty(self) -> None:
pv = ParameterValues()
assert pv.values == {}
def test_with_values(self) -> None:
pv = ParameterValues(values={"speed": 1.0, "pitch": 0.5})
assert pv.values["speed"] == 1.0
assert pv.values["pitch"] == 0.5
def test_immutability(self) -> None:
pv = ParameterValues(values={"speed": 1.0})
with pytest.raises(AttributeError):
pv.values = {} # type: ignore[misc]
class TestSynthesisRequestContract:
"""Contract tests for SynthesisRequest value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(SynthesisRequest, "__dataclass_params__")
assert SynthesisRequest.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
req = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
assert req.text == "Hello"
assert req.voice.source == "builtin"
assert req.format.mime == "audio/wav"
def test_immutability(self) -> None:
req = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(AttributeError):
req.text = "World" # type: ignore[misc]
class TestSynthesizedAudioContract:
"""Contract tests for SynthesizedAudio value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(SynthesizedAudio, "__dataclass_params__")
assert SynthesizedAudio.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
audio = SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
assert audio.data == b"\x00" * 100
assert audio.format.mime == "audio/wav"
assert audio.duration.seconds == 1.0
def test_immutability(self) -> None:
audio = SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
with pytest.raises(AttributeError):
audio.data = b"\x00" # type: ignore[misc]
class TestEngineConfigContract:
"""Contract tests for EngineConfig value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(EngineConfig, "__dataclass_params__")
assert EngineConfig.__dataclass_params__.frozen is True
def test_default_device(self) -> None:
config = EngineConfig()
assert config.device == "cpu"
def test_custom_device(self) -> None:
config = EngineConfig(device="cuda:0")
assert config.device == "cuda:0"
def test_default_lang_code(self) -> None:
config = EngineConfig()
assert config.lang_code == "a"
def test_custom_lang_code(self) -> None:
config = EngineConfig(lang_code="j")
assert config.lang_code == "j"
def test_immutability(self) -> None:
config = EngineConfig()
with pytest.raises(AttributeError):
config.device = "cuda:0" # type: ignore[misc]
def test_immutability_lang_code(self) -> None:
config = EngineConfig()
with pytest.raises(AttributeError):
config.lang_code = "j" # type: ignore[misc]
def test_unknown_keys_ignored_per_spec(self) -> None:
"""Architecture spec: Unknown keys are ignored (no error).
EngineConfig is frozen, so unknown keys cannot be set after creation.
This test verifies the default behavior matches the spec.
"""
config = EngineConfig()
assert config.device == "cpu"
def test_plugins_may_ignore_irrelevant_fields(self) -> None:
"""Architecture Amendment #1: Plugins ignore unsupported fields.
EngineConfig may contain fields that are not relevant to every plugin.
Plugins MUST ignore fields they do not need, not raise on them.
"""
config = EngineConfig(device="cuda:0", lang_code="j")
assert config.device == "cuda:0"
assert config.lang_code == "j"
# A plugin that only needs device simply reads config.device
# and ignores config.lang_code — this must not raise.
def test_engine_config_contains_engine_instance_configuration(self) -> None:
"""Architecture Amendment #1: EngineConfig definition.
EngineConfig contains parameters that define how a particular
Engine instance is created and that remain constant throughout
the lifetime of that Engine.
"""
config = EngineConfig(device="cpu", lang_code="a")
# Both fields are init-time, immutable, engine-scoped.
assert config.device == "cpu"
assert config.lang_code == "a"
"""Contract tests for core domain value objects.
These tests verify that value objects satisfy the architectural requirements:
- Frozen (immutable) dataclasses
- Correct field definitions
- Proper equality behavior
"""
import pytest
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
VoiceSelection,
)
class TestAudioFormatContract:
"""Contract tests for AudioFormat value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(AudioFormat, "__dataclass_params__")
assert AudioFormat.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
af = AudioFormat(mime="audio/wav", extension="wav")
assert af.mime == "audio/wav"
assert af.extension == "wav"
def test_immutability(self) -> None:
af = AudioFormat(mime="audio/wav", extension="wav")
with pytest.raises(AttributeError):
af.mime = "audio/mpeg" # type: ignore[misc]
def test_equality(self) -> None:
af1 = AudioFormat(mime="audio/wav", extension="wav")
af2 = AudioFormat(mime="audio/wav", extension="wav")
assert af1 == af2
def test_inequality(self) -> None:
af1 = AudioFormat(mime="audio/wav", extension="wav")
af2 = AudioFormat(mime="audio/mpeg", extension="mp3")
assert af1 != af2
def test_hashable(self) -> None:
af = AudioFormat(mime="audio/wav", extension="wav")
assert hash(af) == hash(AudioFormat(mime="audio/wav", extension="wav"))
class TestDurationContract:
"""Contract tests for Duration value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(Duration, "__dataclass_params__")
assert Duration.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
d = Duration(seconds=1.5)
assert d.seconds == 1.5
def test_immutability(self) -> None:
d = Duration(seconds=1.0)
with pytest.raises(AttributeError):
d.seconds = 2.0 # type: ignore[misc]
def test_equality(self) -> None:
d1 = Duration(seconds=1.0)
d2 = Duration(seconds=1.0)
assert d1 == d2
class TestVoiceSelectionContract:
"""Contract tests for VoiceSelection value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(VoiceSelection, "__dataclass_params__")
assert VoiceSelection.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
vs = VoiceSelection(source="builtin", key="af_nova")
assert vs.source == "builtin"
assert vs.key == "af_nova"
def test_payload_default_none(self) -> None:
vs = VoiceSelection(source="builtin", key="af_nova")
assert vs.payload is None
def test_payload_optional(self) -> None:
vs = VoiceSelection(source="clone", key="my_voice", payload=b"audio_data")
assert vs.payload == b"audio_data"
def test_immutability(self) -> None:
vs = VoiceSelection(source="builtin", key="af_nova")
with pytest.raises(AttributeError):
vs.source = "other" # type: ignore[misc]
class TestParameterValuesContract:
"""Contract tests for ParameterValues value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(ParameterValues, "__dataclass_params__")
assert ParameterValues.__dataclass_params__.frozen is True
def test_default_empty(self) -> None:
pv = ParameterValues()
assert pv.values == {}
def test_with_values(self) -> None:
pv = ParameterValues(values={"speed": 1.0, "pitch": 0.5})
assert pv.values["speed"] == 1.0
assert pv.values["pitch"] == 0.5
def test_immutability(self) -> None:
pv = ParameterValues(values={"speed": 1.0})
with pytest.raises(AttributeError):
pv.values = {} # type: ignore[misc]
class TestSynthesisRequestContract:
"""Contract tests for SynthesisRequest value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(SynthesisRequest, "__dataclass_params__")
assert SynthesisRequest.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
req = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
assert req.text == "Hello"
assert req.voice.source == "builtin"
assert req.format.mime == "audio/wav"
def test_immutability(self) -> None:
req = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="af_nova"),
parameters=ParameterValues(),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
with pytest.raises(AttributeError):
req.text = "World" # type: ignore[misc]
class TestSynthesizedAudioContract:
"""Contract tests for SynthesizedAudio value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(SynthesizedAudio, "__dataclass_params__")
assert SynthesizedAudio.__dataclass_params__.frozen is True
def test_required_fields(self) -> None:
audio = SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
assert audio.data == b"\x00" * 100
assert audio.format.mime == "audio/wav"
assert audio.duration.seconds == 1.0
def test_immutability(self) -> None:
audio = SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
with pytest.raises(AttributeError):
audio.data = b"\x00" # type: ignore[misc]
class TestEngineConfigContract:
"""Contract tests for EngineConfig value object."""
def test_is_frozen_dataclass(self) -> None:
assert hasattr(EngineConfig, "__dataclass_params__")
assert EngineConfig.__dataclass_params__.frozen is True
def test_default_device(self) -> None:
config = EngineConfig()
assert config.device == "cpu"
def test_custom_device(self) -> None:
config = EngineConfig(device="cuda:0")
assert config.device == "cuda:0"
def test_default_lang_code(self) -> None:
config = EngineConfig()
assert config.lang_code == "a"
def test_custom_lang_code(self) -> None:
config = EngineConfig(lang_code="j")
assert config.lang_code == "j"
def test_immutability(self) -> None:
config = EngineConfig()
with pytest.raises(AttributeError):
config.device = "cuda:0" # type: ignore[misc]
def test_immutability_lang_code(self) -> None:
config = EngineConfig()
with pytest.raises(AttributeError):
config.lang_code = "j" # type: ignore[misc]
def test_unknown_keys_ignored_per_spec(self) -> None:
"""Architecture spec: Unknown keys are ignored (no error).
EngineConfig is frozen, so unknown keys cannot be set after creation.
This test verifies the default behavior matches the spec.
"""
config = EngineConfig()
assert config.device == "cpu"
def test_plugins_may_ignore_irrelevant_fields(self) -> None:
"""Architecture Amendment #1: Plugins ignore unsupported fields.
EngineConfig may contain fields that are not relevant to every plugin.
Plugins MUST ignore fields they do not need, not raise on them.
"""
config = EngineConfig(device="cuda:0", lang_code="j")
assert config.device == "cuda:0"
assert config.lang_code == "j"
# A plugin that only needs device simply reads config.device
# and ignores config.lang_code — this must not raise.
def test_engine_config_contains_engine_instance_configuration(self) -> None:
"""Architecture Amendment #1: EngineConfig definition.
EngineConfig contains parameters that define how a particular
Engine instance is created and that remain constant throughout
the lifetime of that Engine.
"""
config = EngineConfig(device="cpu", lang_code="a")
# Both fields are init-time, immutable, engine-scoped.
assert config.device == "cpu"
assert config.lang_code == "a"
tests/plugins/fake_plugin/__init__.py
+92 -92
View File
@@ -1,92 +1,92 @@
"""Fake plugin for testing the plugin loader.
This is a minimal valid plugin that satisfies the Plugin API contract.
It does NOT perform any real TTS synthesis.
"""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
PluginManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
)
class FakeSession:
"""Minimal EngineSession implementation for testing."""
def __init__(self) -> None:
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
raise EngineError("Session disposed")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def dispose(self) -> None:
self._disposed = True
class FakeEngine:
"""Minimal Engine implementation for testing."""
def __init__(self) -> None:
self._disposed = False
def createSession(self) -> EngineSession:
if self._disposed:
raise EngineError("Engine disposed")
return FakeSession()
def dispose(self) -> None:
self._disposed = True
PLUGIN_MANIFEST = PluginManifest(
id="fake_plugin",
name="Fake Plugin",
version="1.0.0",
api_version="1.0",
description="A fake plugin for testing",
author="Test Author",
capabilities=(),
engine=EngineManifest(
voiceSources=(
VoiceSourceManifest(id="builtin", name="Builtin", type="list"),
),
audioFormats=(
AudioFormatManifest(mime="audio/wav", extension="wav"),
),
),
)
MODEL_REQUIREMENTS: list[Any] = []
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create a fake engine instance."""
return FakeEngine()
"""Fake plugin for testing the plugin loader.
This is a minimal valid plugin that satisfies the Plugin API contract.
It does NOT perform any real TTS synthesis.
"""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.errors import EngineError
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.manifest import (
AudioFormatManifest,
EngineManifest,
PluginManifest,
VoiceSourceManifest,
)
from abogen.tts_plugin.types import (
AudioFormat,
Duration,
EngineConfig,
ParameterValues,
SynthesisRequest,
SynthesizedAudio,
)
class FakeSession:
"""Minimal EngineSession implementation for testing."""
def __init__(self) -> None:
self._disposed = False
def synthesize(self, request: SynthesisRequest) -> SynthesizedAudio:
if self._disposed:
raise EngineError("Session disposed")
return SynthesizedAudio(
data=b"\x00" * 100,
format=AudioFormat(mime="audio/wav", extension="wav"),
duration=Duration(seconds=1.0),
)
def dispose(self) -> None:
self._disposed = True
class FakeEngine:
"""Minimal Engine implementation for testing."""
def __init__(self) -> None:
self._disposed = False
def createSession(self) -> EngineSession:
if self._disposed:
raise EngineError("Engine disposed")
return FakeSession()
def dispose(self) -> None:
self._disposed = True
PLUGIN_MANIFEST = PluginManifest(
id="fake_plugin",
name="Fake Plugin",
version="1.0.0",
api_version="1.0",
description="A fake plugin for testing",
author="Test Author",
capabilities=(),
engine=EngineManifest(
voiceSources=(
VoiceSourceManifest(id="builtin", name="Builtin", type="list"),
),
audioFormats=(
AudioFormatManifest(mime="audio/wav", extension="wav"),
),
),
)
MODEL_REQUIREMENTS: list[Any] = []
def create_engine(
context: HostContext,
model_path: Path | None,
config: EngineConfig,
) -> Engine:
"""Create a fake engine instance."""
return FakeEngine()
tests/plugins/import_error/__init__.py
+18 -18
View File
@@ -1,18 +1,18 @@
"""Invalid plugin: raises ImportError during import."""
from __future__ import annotations
# This plugin intentionally raises an ImportError
raise ImportError("Simulated import error for testing")
# The following code will never be reached, but is here for documentation
from abogen.tts_plugin.manifest import PluginManifest
PLUGIN_MANIFEST = PluginManifest(
id="import_error",
name="Import Error Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin that fails to import",
author="Test Author",
)
"""Invalid plugin: raises ImportError during import."""
from __future__ import annotations
# This plugin intentionally raises an ImportError
raise ImportError("Simulated import error for testing")
# The following code will never be reached, but is here for documentation
from abogen.tts_plugin.manifest import PluginManifest
PLUGIN_MANIFEST = PluginManifest(
id="import_error",
name="Import Error Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin that fails to import",
author="Test Author",
)
tests/plugins/invalid_api_version/__init__.py
+29 -29
View File
@@ -1,29 +1,29 @@
"""Invalid plugin: incompatible api_version (major version mismatch)."""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import EngineConfig
# api_version "2.0" has major version 2, but host expects major version 1
PLUGIN_MANIFEST = PluginManifest(
id="invalid_api_version",
name="Invalid API Version Plugin",
version="1.0.0",
api_version="2.0", # Major version mismatch!
description="Plugin with incompatible api_version",
author="Test Author",
)
MODEL_REQUIREMENTS: list[Any] = []
def create_engine(
context: Any,
model_path: Path | None,
config: EngineConfig,
) -> Any:
raise NotImplementedError("This plugin is invalid")
"""Invalid plugin: incompatible api_version (major version mismatch)."""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import EngineConfig
# api_version "2.0" has major version 2, but host expects major version 1
PLUGIN_MANIFEST = PluginManifest(
id="invalid_api_version",
name="Invalid API Version Plugin",
version="1.0.0",
api_version="2.0", # Major version mismatch!
description="Plugin with incompatible api_version",
author="Test Author",
)
MODEL_REQUIREMENTS: list[Any] = []
def create_engine(
context: Any,
model_path: Path | None,
config: EngineConfig,
) -> Any:
raise NotImplementedError("This plugin is invalid")
tests/plugins/invalid_capabilities/__init__.py
+29 -29
View File
@@ -1,29 +1,29 @@
"""Invalid plugin: unknown capabilities."""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import EngineConfig
PLUGIN_MANIFEST = PluginManifest(
id="invalid_capabilities",
name="Invalid Capabilities Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin with unknown capabilities",
author="Test Author",
capabilities=("voice_list", "unknown_capability", "another_unknown"),
)
MODEL_REQUIREMENTS: list[Any] = []
def create_engine(
context: Any,
model_path: Path | None,
config: EngineConfig,
) -> Any:
raise NotImplementedError("This plugin is invalid")
"""Invalid plugin: unknown capabilities."""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import EngineConfig
PLUGIN_MANIFEST = PluginManifest(
id="invalid_capabilities",
name="Invalid Capabilities Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin with unknown capabilities",
author="Test Author",
capabilities=("voice_list", "unknown_capability", "another_unknown"),
)
MODEL_REQUIREMENTS: list[Any] = []
def create_engine(
context: Any,
model_path: Path | None,
config: EngineConfig,
) -> Any:
raise NotImplementedError("This plugin is invalid")
tests/plugins/missing_create_engine/__init__.py
+18 -18
View File
@@ -1,18 +1,18 @@
"""Invalid plugin: missing create_engine function."""
from __future__ import annotations
from abogen.tts_plugin.manifest import PluginManifest
PLUGIN_MANIFEST = PluginManifest(
id="missing_create_engine",
name="Missing Create Engine Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin missing create_engine",
author="Test Author",
)
MODEL_REQUIREMENTS: list = []
# This plugin intentionally does NOT export create_engine
"""Invalid plugin: missing create_engine function."""
from __future__ import annotations
from abogen.tts_plugin.manifest import PluginManifest
PLUGIN_MANIFEST = PluginManifest(
id="missing_create_engine",
name="Missing Create Engine Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin missing create_engine",
author="Test Author",
)
MODEL_REQUIREMENTS: list = []
# This plugin intentionally does NOT export create_engine
tests/plugins/missing_manifest/__init__.py
+10 -10
View File
@@ -1,10 +1,10 @@
"""Invalid plugin: missing PLUGIN_MANIFEST."""
from __future__ import annotations
# This plugin intentionally does NOT export PLUGIN_MANIFEST
MODEL_REQUIREMENTS: list = []
def create_engine(context, model_path, config):
raise NotImplementedError("This plugin is invalid")
"""Invalid plugin: missing PLUGIN_MANIFEST."""
from __future__ import annotations
# This plugin intentionally does NOT export PLUGIN_MANIFEST
MODEL_REQUIREMENTS: list = []
def create_engine(context, model_path, config):
raise NotImplementedError("This plugin is invalid")
tests/plugins/missing_model_requirements/__init__.py
+28 -28
View File
@@ -1,28 +1,28 @@
"""Invalid plugin: missing MODEL_REQUIREMENTS."""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import EngineConfig
PLUGIN_MANIFEST = PluginManifest(
id="missing_model_requirements",
name="Missing Model Requirements Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin missing MODEL_REQUIREMENTS",
author="Test Author",
)
# This plugin intentionally does NOT export MODEL_REQUIREMENTS
def create_engine(
context: Any,
model_path: Path | None,
config: EngineConfig,
) -> Any:
raise NotImplementedError("This plugin is invalid")
"""Invalid plugin: missing MODEL_REQUIREMENTS."""
from __future__ import annotations
from pathlib import Path
from typing import Any
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import EngineConfig
PLUGIN_MANIFEST = PluginManifest(
id="missing_model_requirements",
name="Missing Model Requirements Plugin",
version="1.0.0",
api_version="1.0",
description="Plugin missing MODEL_REQUIREMENTS",
author="Test Author",
)
# This plugin intentionally does NOT export MODEL_REQUIREMENTS
def create_engine(
context: Any,
model_path: Path | None,
config: EngineConfig,
) -> Any:
raise NotImplementedError("This plugin is invalid")
tests/test_behavioral_regression.py
+1088 -1088
View File
File diff suppressed because it is too large Load Diff
tests/test_conversion_voice_resolution.py
+52 -52
View File
@@ -1,52 +1,52 @@
from types import SimpleNamespace
from typing import cast
from abogen.tts_plugin.utils import get_voices
from abogen.webui.conversion_runner import (
_chapter_voice_spec,
_chunk_voice_spec,
_collect_required_voice_ids,
)
from abogen.webui.service import Job
def _sample_job(formula: str) -> Job:
return cast(
Job,
SimpleNamespace(
voice="__custom_mix",
speakers={
"narrator": {
"resolved_voice": formula,
}
},
chapters=[],
chunks=[{}],
),
)
def test_chapter_voice_spec_uses_resolved_formula():
formula = "af_nova*0.7+am_liam*0.3"
job = _sample_job(formula)
assert _chapter_voice_spec(job, None) == formula
def test_chunk_voice_fallback_uses_resolved_formula():
formula = "af_nova*0.7+am_liam*0.3"
job = _sample_job(formula)
result = _chunk_voice_spec(job, {}, "")
assert result == formula
def test_voice_collection_includes_formula_components():
formula = "af_nova*0.7+am_liam*0.3"
job = _sample_job(formula)
voices = _collect_required_voice_ids(job)
assert {"af_nova", "am_liam"}.issubset(voices)
assert voices.issuperset(get_voices("kokoro"))
from types import SimpleNamespace
from typing import cast
from abogen.tts_plugin.utils import get_voices
from abogen.webui.conversion_runner import (
_chapter_voice_spec,
_chunk_voice_spec,
_collect_required_voice_ids,
)
from abogen.webui.service import Job
def _sample_job(formula: str) -> Job:
return cast(
Job,
SimpleNamespace(
voice="__custom_mix",
speakers={
"narrator": {
"resolved_voice": formula,
}
},
chapters=[],
chunks=[{}],
),
)
def test_chapter_voice_spec_uses_resolved_formula():
formula = "af_nova*0.7+am_liam*0.3"
job = _sample_job(formula)
assert _chapter_voice_spec(job, None) == formula
def test_chunk_voice_fallback_uses_resolved_formula():
formula = "af_nova*0.7+am_liam*0.3"
job = _sample_job(formula)
result = _chunk_voice_spec(job, {}, "")
assert result == formula
def test_voice_collection_includes_formula_components():
formula = "af_nova*0.7+am_liam*0.3"
job = _sample_job(formula)
voices = _collect_required_voice_ids(job)
assert {"af_nova", "am_liam"}.issubset(voices)
assert voices.issuperset(get_voices("kokoro"))
tests/test_kokoro_plugin.py
+196 -196
View File
@@ -1,196 +1,196 @@
"""Tests for the Kokoro TTS Plugin.
These tests verify that the Kokoro plugin:
- Loads correctly through the Plugin Loader
- Has a valid manifest
- Creates a valid Engine
- Satisfies the Engine/EngineSession contract (via EngineContractMixin)
- Implements VoiceLister capability
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Any
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.loader import load_plugin_from_dir
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import (
AudioFormat,
EngineConfig,
ParameterValues,
SynthesisRequest,
VoiceSelection,
)
from tests.contracts.engine_contract import EngineContractMixin
# ──────────────────────────────────────────────────────────────
# Helpers
# ──────────────────────────────────────────────────────────────
def _kokoro_available() -> bool:
try:
from kokoro import KPipeline # type: ignore[import-not-found]
return True
except ImportError:
return False
def _make_mock_engine() -> Any:
from plugins.kokoro.engine import KokoroEngine
class MockPipeline:
def __call__(self, text, voice, speed, split_pattern=None):
class MockSegment:
def __init__(self):
self.audio = MockAudio()
class MockAudio:
def numpy(self):
import numpy as np
return np.zeros(24000, dtype="float32")
return [MockSegment()]
engine = KokoroEngine(MockPipeline())
# Override listVoices for testing (real engine reads from manifest)
from abogen.tts_plugin.manifest import VoiceManifest
engine.listVoices = lambda source_id: [
VoiceManifest(id="test_voice_1", name="Test Voice 1", tags=("en",)),
VoiceManifest(id="test_voice_2", name="Test Voice 2", tags=("es",)),
]
return engine
# ──────────────────────────────────────────────────────────────
# Fixtures
# ──────────────────────────────────────────────────────────────
@pytest.fixture
def kokoro_plugin_dir() -> Path:
return Path(__file__).parent.parent / "plugins" / "kokoro"
@pytest.fixture
def host_context(tmp_path: Path) -> HostContext:
class FakeHttpClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
return HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=FakeHttpClient(),
)
@pytest.fixture
def engine() -> Engine:
return _make_mock_engine()
# ──────────────────────────────────────────────────────────────
# Plugin Loading Tests
# ──────────────────────────────────────────────────────────────
class TestKokoroPluginLoading:
def test_plugin_loads_successfully(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
assert result.manifest is not None
assert result.create_engine is not None
def test_plugin_has_valid_manifest(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
manifest = result.manifest
assert isinstance(manifest, PluginManifest)
assert manifest.id == "kokoro"
assert manifest.name == "Kokoro"
assert manifest.api_version == "1.0"
def test_plugin_has_model_requirements(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
assert result.model_requirements is not None
assert isinstance(result.model_requirements, tuple)
def test_plugin_manifest_capabilities(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
assert "voice_list" in result.manifest.capabilities
def test_plugin_manifest_engine(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
engine_manifest = result.manifest.engine
assert len(engine_manifest.voiceSources) > 0
assert len(engine_manifest.audioFormats) > 0
assert len(engine_manifest.parameters) > 0
# ──────────────────────────────────────────────────────────────
# Engine Creation (real backend, skipped if not installed)
# ──────────────────────────────────────────────────────────────
class TestKokoroEngineCreation:
@pytest.mark.skipif(not _kokoro_available(), reason="Kokoro not installed")
def test_create_engine(self, kokoro_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
@pytest.mark.skipif(not _kokoro_available(), reason="Kokoro not installed")
def test_engine_satisfies_protocol(self, kokoro_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
# ──────────────────────────────────────────────────────────────
# Engine / Session Contract (inherited from base)
# ──────────────────────────────────────────────────────────────
class TestKokoroEngineContract(EngineContractMixin):
"""Every test from EngineContractMixin runs against KokoroEngine."""
@pytest.fixture
def default_voice(self) -> str:
return "af_nova"
# ──────────────────────────────────────────────────────────────
# VoiceLister Tests
# ──────────────────────────────────────────────────────────────
class TestKokoroVoiceLister:
def test_list_voices(self) -> None:
engine = _make_mock_engine()
voices = engine.listVoices("builtin")
assert len(voices) > 0
assert all(hasattr(v, "id") for v in voices)
assert all(hasattr(v, "name") for v in voices)
engine.dispose()
def test_voices_have_tags(self) -> None:
engine = _make_mock_engine()
voices = engine.listVoices("builtin")
for voice in voices:
assert isinstance(voice.tags, tuple)
assert len(voice.tags) > 0
engine.dispose()
"""Tests for the Kokoro TTS Plugin.
These tests verify that the Kokoro plugin:
- Loads correctly through the Plugin Loader
- Has a valid manifest
- Creates a valid Engine
- Satisfies the Engine/EngineSession contract (via EngineContractMixin)
- Implements VoiceLister capability
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Any
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.loader import load_plugin_from_dir
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import (
AudioFormat,
EngineConfig,
ParameterValues,
SynthesisRequest,
VoiceSelection,
)
from tests.contracts.engine_contract import EngineContractMixin
# ──────────────────────────────────────────────────────────────
# Helpers
# ──────────────────────────────────────────────────────────────
def _kokoro_available() -> bool:
try:
from kokoro import KPipeline # type: ignore[import-not-found]
return True
except ImportError:
return False
def _make_mock_engine() -> Any:
from plugins.kokoro.engine import KokoroEngine
class MockPipeline:
def __call__(self, text, voice, speed, split_pattern=None):
class MockSegment:
def __init__(self):
self.audio = MockAudio()
class MockAudio:
def numpy(self):
import numpy as np
return np.zeros(24000, dtype="float32")
return [MockSegment()]
engine = KokoroEngine(MockPipeline())
# Override listVoices for testing (real engine reads from manifest)
from abogen.tts_plugin.manifest import VoiceManifest
engine.listVoices = lambda source_id: [
VoiceManifest(id="test_voice_1", name="Test Voice 1", tags=("en",)),
VoiceManifest(id="test_voice_2", name="Test Voice 2", tags=("es",)),
]
return engine
# ──────────────────────────────────────────────────────────────
# Fixtures
# ──────────────────────────────────────────────────────────────
@pytest.fixture
def kokoro_plugin_dir() -> Path:
return Path(__file__).parent.parent / "plugins" / "kokoro"
@pytest.fixture
def host_context(tmp_path: Path) -> HostContext:
class FakeHttpClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
return HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=FakeHttpClient(),
)
@pytest.fixture
def engine() -> Engine:
return _make_mock_engine()
# ──────────────────────────────────────────────────────────────
# Plugin Loading Tests
# ──────────────────────────────────────────────────────────────
class TestKokoroPluginLoading:
def test_plugin_loads_successfully(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
assert result.manifest is not None
assert result.create_engine is not None
def test_plugin_has_valid_manifest(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
manifest = result.manifest
assert isinstance(manifest, PluginManifest)
assert manifest.id == "kokoro"
assert manifest.name == "Kokoro"
assert manifest.api_version == "1.0"
def test_plugin_has_model_requirements(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
assert result.model_requirements is not None
assert isinstance(result.model_requirements, tuple)
def test_plugin_manifest_capabilities(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
assert "voice_list" in result.manifest.capabilities
def test_plugin_manifest_engine(self, kokoro_plugin_dir: Path) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
engine_manifest = result.manifest.engine
assert len(engine_manifest.voiceSources) > 0
assert len(engine_manifest.audioFormats) > 0
assert len(engine_manifest.parameters) > 0
# ──────────────────────────────────────────────────────────────
# Engine Creation (real backend, skipped if not installed)
# ──────────────────────────────────────────────────────────────
class TestKokoroEngineCreation:
@pytest.mark.skipif(not _kokoro_available(), reason="Kokoro not installed")
def test_create_engine(self, kokoro_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
@pytest.mark.skipif(not _kokoro_available(), reason="Kokoro not installed")
def test_engine_satisfies_protocol(self, kokoro_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(kokoro_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
# ──────────────────────────────────────────────────────────────
# Engine / Session Contract (inherited from base)
# ──────────────────────────────────────────────────────────────
class TestKokoroEngineContract(EngineContractMixin):
"""Every test from EngineContractMixin runs against KokoroEngine."""
@pytest.fixture
def default_voice(self) -> str:
return "af_nova"
# ──────────────────────────────────────────────────────────────
# VoiceLister Tests
# ──────────────────────────────────────────────────────────────
class TestKokoroVoiceLister:
def test_list_voices(self) -> None:
engine = _make_mock_engine()
voices = engine.listVoices("builtin")
assert len(voices) > 0
assert all(hasattr(v, "id") for v in voices)
assert all(hasattr(v, "name") for v in voices)
engine.dispose()
def test_voices_have_tags(self) -> None:
engine = _make_mock_engine()
voices = engine.listVoices("builtin")
for voice in voices:
assert isinstance(voice.tags, tuple)
assert len(voice.tags) > 0
engine.dispose()
tests/test_preview_applies_manual_overrides.py
+60 -60
View File
@@ -1,60 +1,60 @@
from abogen.webui.routes.utils import preview
def test_preview_applies_manual_override_before_normalization(monkeypatch):
# Don't run real TTS/normalization; just exercise the override stage by
# forcing provider=kokoro and then stubbing normalize_for_pipeline.
monkeypatch.setattr(preview, "get_preview_pipeline", lambda language, device: None)
# Stub normalize_for_pipeline to be identity; we only care that overrides run.
class _Norm:
@staticmethod
def normalize_for_pipeline(text):
return text
monkeypatch.setitem(
__import__("sys").modules, "abogen.kokoro_text_normalization", _Norm
)
# And stub the kokoro pipeline path so generate_preview_audio won't proceed.
# We'll instead validate by calling the override logic through generate_preview_audio
# with provider=supertonic and stub create_backend to capture input.
captured = {}
class DummyPipeline:
def __init__(self, **kwargs):
pass
def __call__(self, text, **kwargs):
captured["text"] = text
return iter(())
from abogen.tts_plugin import utils
original_create_pipeline = utils.create_pipeline
def _mock_create_pipeline(backend_id, **kwargs):
if backend_id == "supertonic":
return DummyPipeline(**kwargs)
return original_create_pipeline(backend_id, **kwargs)
monkeypatch.setattr(utils, "create_pipeline", _mock_create_pipeline)
try:
preview.generate_preview_audio(
text="He said Unfu*k loudly.",
voice_spec="M1",
language="en",
speed=1.0,
use_gpu=False,
tts_provider="supertonic",
manual_overrides=[{"token": "Unfu*k", "pronunciation": "Unfuck"}],
)
except Exception:
# generate_preview_audio will raise because no audio chunks; that's fine.
pass
assert "text" in captured
assert "Unfuck" in captured["text"]
assert "Unfu*k" not in captured["text"]
from abogen.webui.routes.utils import preview
def test_preview_applies_manual_override_before_normalization(monkeypatch):
# Don't run real TTS/normalization; just exercise the override stage by
# forcing provider=kokoro and then stubbing normalize_for_pipeline.
monkeypatch.setattr(preview, "get_preview_pipeline", lambda language, device: None)
# Stub normalize_for_pipeline to be identity; we only care that overrides run.
class _Norm:
@staticmethod
def normalize_for_pipeline(text):
return text
monkeypatch.setitem(
__import__("sys").modules, "abogen.kokoro_text_normalization", _Norm
)
# And stub the kokoro pipeline path so generate_preview_audio won't proceed.
# We'll instead validate by calling the override logic through generate_preview_audio
# with provider=supertonic and stub create_backend to capture input.
captured = {}
class DummyPipeline:
def __init__(self, **kwargs):
pass
def __call__(self, text, **kwargs):
captured["text"] = text
return iter(())
from abogen.tts_plugin import utils
original_create_pipeline = utils.create_pipeline
def _mock_create_pipeline(backend_id, **kwargs):
if backend_id == "supertonic":
return DummyPipeline(**kwargs)
return original_create_pipeline(backend_id, **kwargs)
monkeypatch.setattr(utils, "create_pipeline", _mock_create_pipeline)
try:
preview.generate_preview_audio(
text="He said Unfu*k loudly.",
voice_spec="M1",
language="en",
speed=1.0,
use_gpu=False,
tts_provider="supertonic",
manual_overrides=[{"token": "Unfu*k", "pronunciation": "Unfuck"}],
)
except Exception:
# generate_preview_audio will raise because no audio chunks; that's fine.
pass
assert "text" in captured
assert "Unfuck" in captured["text"]
assert "Unfu*k" not in captured["text"]
tests/test_supertonic_plugin.py
+281 -281
View File
@@ -1,281 +1,281 @@
"""Tests for the SuperTonic TTS Plugin.
These tests verify that the SuperTonic plugin:
- Loads correctly through the Plugin Loader
- Has a valid manifest
- Creates a valid Engine
- Satisfies the Engine/EngineSession contract (via EngineContractMixin)
- Implements VoiceLister capability
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Any
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.loader import load_plugin_from_dir
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import (
AudioFormat,
EngineConfig,
ParameterValues,
SynthesisRequest,
VoiceSelection,
)
from tests.contracts.engine_contract import EngineContractMixin
# ──────────────────────────────────────────────────────────────
# Helpers
# ──────────────────────────────────────────────────────────────
def _supertonic_available() -> bool:
try:
from supertonic import TTS # type: ignore[import-not-found]
return True
except ImportError:
return False
def _make_mock_engine() -> Any:
from plugins.supertonic.engine import SuperTonicEngine
class MockSegment:
def __init__(self):
import numpy as np
self.audio = np.zeros(24000, dtype="float32")
class MockPipeline:
sample_rate = 24000
def __call__(self, text, voice, speed, split_pattern=None, total_steps=None):
return [MockSegment()]
engine = SuperTonicEngine(MockPipeline())
# Override listVoices for testing (real engine reads from manifest)
from abogen.tts_plugin.manifest import VoiceManifest
original_list_voices = engine.listVoices
def mock_list_voices(source_id):
if engine._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return [
VoiceManifest(id="M1", name="Male 1", tags=("male",)),
VoiceManifest(id="M2", name="Male 2", tags=("male",)),
VoiceManifest(id="M3", name="Male 3", tags=("male",)),
VoiceManifest(id="M4", name="Male 4", tags=("male",)),
VoiceManifest(id="M5", name="Male 5", tags=("male",)),
VoiceManifest(id="F1", name="Female 1", tags=("female",)),
VoiceManifest(id="F2", name="Female 2", tags=("female",)),
VoiceManifest(id="F3", name="Female 3", tags=("female",)),
VoiceManifest(id="F4", name="Female 4", tags=("female",)),
VoiceManifest(id="F5", name="Female 5", tags=("female",)),
]
engine.listVoices = mock_list_voices
return engine
# ──────────────────────────────────────────────────────────────
# Fixtures
# ──────────────────────────────────────────────────────────────
@pytest.fixture
def supertonic_plugin_dir() -> Path:
return Path(__file__).parent.parent / "plugins" / "supertonic"
@pytest.fixture
def host_context(tmp_path: Path) -> HostContext:
class FakeHttpClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
return HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=FakeHttpClient(),
)
@pytest.fixture
def engine() -> Engine:
return _make_mock_engine()
# ──────────────────────────────────────────────────────────────
# Plugin Loading Tests
# ──────────────────────────────────────────────────────────────
class TestSuperTonicPluginLoading:
def test_plugin_loads_successfully(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
assert result.manifest is not None
assert result.create_engine is not None
def test_plugin_has_valid_manifest(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
manifest = result.manifest
assert isinstance(manifest, PluginManifest)
assert manifest.id == "supertonic"
assert manifest.name == "SuperTonic"
assert manifest.api_version == "1.0"
def test_plugin_has_model_requirements(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
assert result.model_requirements is not None
assert isinstance(result.model_requirements, tuple)
def test_plugin_manifest_capabilities(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
assert "voice_list" in result.manifest.capabilities
def test_plugin_manifest_engine(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
engine_manifest = result.manifest.engine
assert len(engine_manifest.voiceSources) > 0
assert len(engine_manifest.audioFormats) > 0
assert len(engine_manifest.parameters) > 0
# ──────────────────────────────────────────────────────────────
# Engine Creation (real backend, skipped if not installed)
# ──────────────────────────────────────────────────────────────
class TestSuperTonicEngineCreation:
@pytest.mark.skipif(not _supertonic_available(), reason="SuperTonic not installed")
def test_create_engine(self, supertonic_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
@pytest.mark.skipif(not _supertonic_available(), reason="SuperTonic not installed")
def test_engine_satisfies_protocol(self, supertonic_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
# ──────────────────────────────────────────────────────────────
# Engine / Session Contract (inherited from base)
# ──────────────────────────────────────────────────────────────
class TestSuperTonicEngineContract(EngineContractMixin):
"""Every test from EngineContractMixin runs against SuperTonicEngine."""
@pytest.fixture
def default_voice(self) -> str:
return "M1"
# ──────────────────────────────────────────────────────────────
# VoiceLister Tests
# ──────────────────────────────────────────────────────────────
class TestSuperTonicVoiceLister:
def test_list_voices(self) -> None:
engine = _make_mock_engine()
voices = engine.listVoices("builtin")
assert len(voices) == 10
assert all(hasattr(v, "id") for v in voices)
assert all(hasattr(v, "name") for v in voices)
engine.dispose()
def test_voices_have_tags(self) -> None:
engine = _make_mock_engine()
for voice in engine.listVoices("builtin"):
assert isinstance(voice.tags, tuple)
assert len(voice.tags) > 0
engine.dispose()
def test_male_voices_have_male_tag(self) -> None:
engine = _make_mock_engine()
for v in engine.listVoices("builtin"):
if v.id.startswith("M"):
assert "male" in v.tags
engine.dispose()
def test_female_voices_have_female_tag(self) -> None:
engine = _make_mock_engine()
for v in engine.listVoices("builtin"):
if v.id.startswith("F"):
assert "female" in v.tags
engine.dispose()
def test_list_voices_after_dispose_raises(self) -> None:
from abogen.tts_plugin.errors import EngineError
engine = _make_mock_engine()
engine.dispose()
with pytest.raises(EngineError):
engine.listVoices("builtin")
# ──────────────────────────────────────────────────────────────
# SuperTonic-specific parameter tests
# ──────────────────────────────────────────────────────────────
class TestSuperTonicParameters:
def test_speed_parameter(self) -> None:
engine = _make_mock_engine()
session = engine.createSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="M1"),
parameters=ParameterValues(values={"speed": 1.5}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
session.dispose()
engine.dispose()
def test_total_steps_parameter(self) -> None:
engine = _make_mock_engine()
session = engine.createSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="M1"),
parameters=ParameterValues(values={"total_steps": 10}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
session.dispose()
engine.dispose()
def test_default_parameters(self) -> None:
engine = _make_mock_engine()
session = engine.createSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="M1"),
parameters=ParameterValues(values={}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
session.dispose()
engine.dispose()
"""Tests for the SuperTonic TTS Plugin.
These tests verify that the SuperTonic plugin:
- Loads correctly through the Plugin Loader
- Has a valid manifest
- Creates a valid Engine
- Satisfies the Engine/EngineSession contract (via EngineContractMixin)
- Implements VoiceLister capability
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Any
import pytest
from abogen.tts_plugin.engine import Engine, EngineSession
from abogen.tts_plugin.host_context import HostContext
from abogen.tts_plugin.loader import load_plugin_from_dir
from abogen.tts_plugin.manifest import PluginManifest
from abogen.tts_plugin.types import (
AudioFormat,
EngineConfig,
ParameterValues,
SynthesisRequest,
VoiceSelection,
)
from tests.contracts.engine_contract import EngineContractMixin
# ──────────────────────────────────────────────────────────────
# Helpers
# ──────────────────────────────────────────────────────────────
def _supertonic_available() -> bool:
try:
from supertonic import TTS # type: ignore[import-not-found]
return True
except ImportError:
return False
def _make_mock_engine() -> Any:
from plugins.supertonic.engine import SuperTonicEngine
class MockSegment:
def __init__(self):
import numpy as np
self.audio = np.zeros(24000, dtype="float32")
class MockPipeline:
sample_rate = 24000
def __call__(self, text, voice, speed, split_pattern=None, total_steps=None):
return [MockSegment()]
engine = SuperTonicEngine(MockPipeline())
# Override listVoices for testing (real engine reads from manifest)
from abogen.tts_plugin.manifest import VoiceManifest
original_list_voices = engine.listVoices
def mock_list_voices(source_id):
if engine._disposed:
from abogen.tts_plugin.errors import EngineError
raise EngineError("Engine disposed")
return [
VoiceManifest(id="M1", name="Male 1", tags=("male",)),
VoiceManifest(id="M2", name="Male 2", tags=("male",)),
VoiceManifest(id="M3", name="Male 3", tags=("male",)),
VoiceManifest(id="M4", name="Male 4", tags=("male",)),
VoiceManifest(id="M5", name="Male 5", tags=("male",)),
VoiceManifest(id="F1", name="Female 1", tags=("female",)),
VoiceManifest(id="F2", name="Female 2", tags=("female",)),
VoiceManifest(id="F3", name="Female 3", tags=("female",)),
VoiceManifest(id="F4", name="Female 4", tags=("female",)),
VoiceManifest(id="F5", name="Female 5", tags=("female",)),
]
engine.listVoices = mock_list_voices
return engine
# ──────────────────────────────────────────────────────────────
# Fixtures
# ──────────────────────────────────────────────────────────────
@pytest.fixture
def supertonic_plugin_dir() -> Path:
return Path(__file__).parent.parent / "plugins" / "supertonic"
@pytest.fixture
def host_context(tmp_path: Path) -> HostContext:
class FakeHttpClient:
def get(self, url: str, **kwargs: object) -> object:
return None
def post(self, url: str, **kwargs: object) -> object:
return None
return HostContext(
config_dir=tmp_path,
logger=logging.getLogger("test"),
http_client=FakeHttpClient(),
)
@pytest.fixture
def engine() -> Engine:
return _make_mock_engine()
# ──────────────────────────────────────────────────────────────
# Plugin Loading Tests
# ──────────────────────────────────────────────────────────────
class TestSuperTonicPluginLoading:
def test_plugin_loads_successfully(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
assert result.manifest is not None
assert result.create_engine is not None
def test_plugin_has_valid_manifest(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
manifest = result.manifest
assert isinstance(manifest, PluginManifest)
assert manifest.id == "supertonic"
assert manifest.name == "SuperTonic"
assert manifest.api_version == "1.0"
def test_plugin_has_model_requirements(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
assert result.model_requirements is not None
assert isinstance(result.model_requirements, tuple)
def test_plugin_manifest_capabilities(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
assert "voice_list" in result.manifest.capabilities
def test_plugin_manifest_engine(self, supertonic_plugin_dir: Path) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
engine_manifest = result.manifest.engine
assert len(engine_manifest.voiceSources) > 0
assert len(engine_manifest.audioFormats) > 0
assert len(engine_manifest.parameters) > 0
# ──────────────────────────────────────────────────────────────
# Engine Creation (real backend, skipped if not installed)
# ──────────────────────────────────────────────────────────────
class TestSuperTonicEngineCreation:
@pytest.mark.skipif(not _supertonic_available(), reason="SuperTonic not installed")
def test_create_engine(self, supertonic_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
@pytest.mark.skipif(not _supertonic_available(), reason="SuperTonic not installed")
def test_engine_satisfies_protocol(self, supertonic_plugin_dir: Path, host_context: HostContext) -> None:
result = load_plugin_from_dir(supertonic_plugin_dir)
assert result.success is True
engine = result.create_engine(host_context, None, EngineConfig())
assert isinstance(engine, Engine)
engine.dispose()
# ──────────────────────────────────────────────────────────────
# Engine / Session Contract (inherited from base)
# ──────────────────────────────────────────────────────────────
class TestSuperTonicEngineContract(EngineContractMixin):
"""Every test from EngineContractMixin runs against SuperTonicEngine."""
@pytest.fixture
def default_voice(self) -> str:
return "M1"
# ──────────────────────────────────────────────────────────────
# VoiceLister Tests
# ──────────────────────────────────────────────────────────────
class TestSuperTonicVoiceLister:
def test_list_voices(self) -> None:
engine = _make_mock_engine()
voices = engine.listVoices("builtin")
assert len(voices) == 10
assert all(hasattr(v, "id") for v in voices)
assert all(hasattr(v, "name") for v in voices)
engine.dispose()
def test_voices_have_tags(self) -> None:
engine = _make_mock_engine()
for voice in engine.listVoices("builtin"):
assert isinstance(voice.tags, tuple)
assert len(voice.tags) > 0
engine.dispose()
def test_male_voices_have_male_tag(self) -> None:
engine = _make_mock_engine()
for v in engine.listVoices("builtin"):
if v.id.startswith("M"):
assert "male" in v.tags
engine.dispose()
def test_female_voices_have_female_tag(self) -> None:
engine = _make_mock_engine()
for v in engine.listVoices("builtin"):
if v.id.startswith("F"):
assert "female" in v.tags
engine.dispose()
def test_list_voices_after_dispose_raises(self) -> None:
from abogen.tts_plugin.errors import EngineError
engine = _make_mock_engine()
engine.dispose()
with pytest.raises(EngineError):
engine.listVoices("builtin")
# ──────────────────────────────────────────────────────────────
# SuperTonic-specific parameter tests
# ──────────────────────────────────────────────────────────────
class TestSuperTonicParameters:
def test_speed_parameter(self) -> None:
engine = _make_mock_engine()
session = engine.createSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="M1"),
parameters=ParameterValues(values={"speed": 1.5}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
session.dispose()
engine.dispose()
def test_total_steps_parameter(self) -> None:
engine = _make_mock_engine()
session = engine.createSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="M1"),
parameters=ParameterValues(values={"total_steps": 10}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
session.dispose()
engine.dispose()
def test_default_parameters(self) -> None:
engine = _make_mock_engine()
session = engine.createSession()
request = SynthesisRequest(
text="Hello",
voice=VoiceSelection(source="builtin", key="M1"),
parameters=ParameterValues(values={}),
format=AudioFormat(mime="audio/wav", extension="wav"),
)
result = session.synthesize(request)
assert isinstance(result.data, bytes)
session.dispose()
engine.dispose()
tests/test_voice_cache.py
+69 -69
View File
@@ -1,69 +1,69 @@
from types import SimpleNamespace
from typing import cast
import pytest
from abogen.tts_plugin.utils import get_voices
from abogen.voice_cache import (
LocalEntryNotFoundError,
_CACHED_VOICES,
ensure_voice_assets,
)
from abogen.webui.conversion_runner import _collect_required_voice_ids
from abogen.webui.service import Job
@pytest.fixture(autouse=True)
def clear_voice_cache():
_CACHED_VOICES.clear()
yield
_CACHED_VOICES.clear()
def test_ensure_voice_assets_downloads_missing(monkeypatch):
recorded = []
cached = set()
def fake_download(**kwargs):
filename = kwargs["filename"]
if kwargs.get("local_files_only"):
if filename in cached:
return f"/tmp/{filename}"
raise LocalEntryNotFoundError(f"{filename} missing")
recorded.append(filename)
cached.add(filename)
return f"/tmp/{filename}"
monkeypatch.setattr("abogen.voice_cache.hf_hub_download", fake_download)
downloaded, errors = ensure_voice_assets(["af_nova", "am_liam"])
assert downloaded == {"af_nova", "am_liam"}
assert errors == {}
assert set(recorded) == {"voices/af_nova.pt", "voices/am_liam.pt"}
recorded.clear()
downloaded_again, errors_again = ensure_voice_assets(["af_nova"])
assert downloaded_again == set()
assert errors_again == {}
assert recorded == []
def test_collect_required_voice_ids_includes_all():
job = SimpleNamespace(
voice="af_nova",
chapters=[{"voice_formula": "af_nova*0.7+am_liam*0.3"}],
chunks=[{"voice": "am_michael"}],
speakers={
"hero": {"voice_formula": "af_nova*0.6+am_liam*0.4"},
"narrator": {"voice": "af_nova"},
},
)
voices = _collect_required_voice_ids(cast(Job, job))
assert {"af_nova", "am_liam", "am_michael"}.issubset(voices)
assert voices.issuperset(get_voices("kokoro"))
from types import SimpleNamespace
from typing import cast
import pytest
from abogen.tts_plugin.utils import get_voices
from abogen.voice_cache import (
LocalEntryNotFoundError,
_CACHED_VOICES,
ensure_voice_assets,
)
from abogen.webui.conversion_runner import _collect_required_voice_ids
from abogen.webui.service import Job
@pytest.fixture(autouse=True)
def clear_voice_cache():
_CACHED_VOICES.clear()
yield
_CACHED_VOICES.clear()
def test_ensure_voice_assets_downloads_missing(monkeypatch):
recorded = []
cached = set()
def fake_download(**kwargs):
filename = kwargs["filename"]
if kwargs.get("local_files_only"):
if filename in cached:
return f"/tmp/{filename}"
raise LocalEntryNotFoundError(f"{filename} missing")
recorded.append(filename)
cached.add(filename)
return f"/tmp/{filename}"
monkeypatch.setattr("abogen.voice_cache.hf_hub_download", fake_download)
downloaded, errors = ensure_voice_assets(["af_nova", "am_liam"])
assert downloaded == {"af_nova", "am_liam"}
assert errors == {}
assert set(recorded) == {"voices/af_nova.pt", "voices/am_liam.pt"}
recorded.clear()
downloaded_again, errors_again = ensure_voice_assets(["af_nova"])
assert downloaded_again == set()
assert errors_again == {}
assert recorded == []
def test_collect_required_voice_ids_includes_all():
job = SimpleNamespace(
voice="af_nova",
chapters=[{"voice_formula": "af_nova*0.7+am_liam*0.3"}],
chunks=[{"voice": "am_michael"}],
speakers={
"hero": {"voice_formula": "af_nova*0.6+am_liam*0.4"},
"narrator": {"voice": "af_nova"},
},
)
voices = _collect_required_voice_ids(cast(Job, job))
assert {"af_nova", "am_liam", "am_michael"}.issubset(voices)
assert voices.issuperset(get_voices("kokoro"))