mirror of
https://github.com/denizsafak/abogen.git
synced 2026-09-20 19:50:59 +02:00
Artem Akymenko
2b70b9ca45
Language enum expanded from 9 to 33 languages:
- Added 24 new ISO 639-1 languages: AR, BG, CS, DA, DE, EL, ET, FI,
HR, HU, ID, KO, LT, LV, NL, PL, RO, RU, SK, SL, SV, TR, UK, VI
- Updated display_name, is_cjk (added KO)
Supertonic language mapping (32 languages, no ZH):
- engine.py: _SUPERTONIC_LANG_MAP, engine_language(), supported_languages()
- __init__.py: create_engine() passes config.language to pipeline
- pipeline.py: __init__() accepts language, resolves to ISO code;
__call__() passes lang= to TTS.synthesize()
total_steps propagation:
- tts_segments(): +total_steps param, conditionally passed to backend
- synthesize_text(): +total_steps param
- run_tts_segment_loop(): +total_steps param
- executor: all 5 synthesize_text() calls pass total_steps
Integration:
- pipeline_factory: create_pipeline_for_job() passes language to supertonic
- preview path: create_pipeline('supertonic', language=language)
Tests updated to accept total_steps in FakeBackend.__call__
138 lines
4.1 KiB
Python
138 lines
4.1 KiB
Python
"""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(language: Any = None) -> Any:
|
|
"""Lazy-load SuperTonic dependencies and create pipeline."""
|
|
from plugins.supertonic.pipeline import SupertonicPipeline
|
|
|
|
return SupertonicPipeline(
|
|
language=language,
|
|
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(language=config.language)
|
|
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
|