10 Commits
16 changed files with 1373 additions and 806 deletions
+15
View File
@@ -1,3 +1,18 @@
# v1.1.2
- Now you can play the audio files while they are processing.
- Audio and subtitle files are now written directly to disk during generation, which significantly reduces memory usage.
- Added a better logic for detecting chapters from the epub, mentioned by @jefro108 in #33
- Added a new option: `Reset to default settings`, allowing users to reset all settings to their default values.
- Added a new option: `Disable Kokoro's internet access`. This lets you prevent Kokoro from downloading models or voices from HuggingFace Hub, which can help avoid long waiting times if your computer is offline.
- HuggingFace Hub telemetry is now disabled by default for improved privacy. (HuggingFace Hub is used by Kokoro to download its models)
- Potential fix for #37 and #38, where the program was becoming slow while processing large files.
- Fixed `Open folder` and `Open file` buttons in the queue manager GUI.
- Improvements in code structure.
# v1.1.1
- Fixed adding wrong file in queue for EPUB and PDF files, ensuring the correct file is added to the queue.
- Reformatted the code using Black.
# v1.1.0
- Added queue system for processing multiple items, allowing users to add multiple files and process them in a queue, mentioned by @jborza in #30 (Special thanks to @jborza for implementing this feature in PR #35)
- Added a feature that allows selecting multiple items in book handler (in right click menu) by @jborza in #31, that fixes #28
+2
View File
@@ -153,6 +153,8 @@ Heres Abogen in action: in this demo, it processes 3,000 characters of tex
| **Open temp directory** | Opens the temporary directory where converted text files are stored. |
| **Clear temporary files** | Deletes temporary files created during the conversion or preview. |
| **Check for updates at startup** | Automatically checks for updates when the program starts. |
| **Disable Kokoro's internet access** | Prevents Kokoro from downloading models or voices from HuggingFace Hub, useful for offline use. |
| **Reset to default settings** | Resets all settings to their default values. |
## `Voice Mixer`
<img title="Abogen Voice Mixer" src='https://raw.githubusercontent.com/denizsafak/abogen/refs/heads/main/demo/voice_mixer.png'>
+2 -2
View File
@@ -230,7 +230,7 @@ if errorlevel 1 (
:: Install setup requirements
echo Installing setup requirements...
%PYTHON_CONSOLE_PATH% -m pip install --upgrade setuptools setuptools-scm wheel sphinx hatchling --no-warn-script-location
%PYTHON_CONSOLE_PATH% -m pip install --upgrade setuptools setuptools-scm wheel sphinx hatchling editables --no-warn-script-location
if errorlevel 1 (
echo Failed to install setup requirements.
pause
@@ -250,7 +250,7 @@ if errorlevel 1 (
echo Checking and installing project dependencies...
if exist %PYPROJECT_FILE% (
echo Installing project from pyproject.toml...
%PYTHON_CONSOLE_PATH% -m pip install . --no-warn-script-location
%PYTHON_CONSOLE_PATH% -m pip install -e . --no-warn-script-location
if errorlevel 1 (
echo Failed to install from pyproject.toml.
pause
+1 -1
View File
@@ -1 +1 @@
1.1.0
1.1.2
+81 -32
View File
@@ -22,7 +22,7 @@ from PyQt5.QtWidgets import (
QLabel,
)
from PyQt5.QtCore import Qt
from utils import clean_text, calculate_text_length
from abogen.utils import clean_text, calculate_text_length
import os
import logging # Add logging
import urllib.parse
@@ -240,20 +240,17 @@ class HandlerDialog(QDialog):
tag.decompose()
text = clean_text(soup.get_text()).strip()
if text:
# Use doc_href as the identifier
self.content_texts[doc_href] = text
self.content_lengths[doc_href] = len(text)
# Create a synthetic TOC entry
title = f"Chapter {i+1}: {doc_href}"
# Try to get a better title from <h1> or <title>
h1 = soup.find("h1")
if h1 and h1.get_text(strip=True):
title = h1.get_text(strip=True)
else:
title_tag = soup.find("title")
if title_tag and title_tag.get_text(strip=True):
title = title_tag.get_text(strip=True)
title = None
if soup.title and soup.title.string:
title = soup.title.string.strip()
elif (h1 := soup.find("h1")) and h1.get_text(strip=True):
title = h1.get_text(strip=True)
if not title:
title = f"Untitled Chapter {i+1}"
synthetic_toc.append(
(epub.Link(doc_href, title, doc_href), [])
) # Wrap in tuple and empty list for compatibility
@@ -997,10 +994,21 @@ class HandlerDialog(QDialog):
bookmark_item.setData(0, Qt.UserRole, page_id)
# only allow checking if this chapter has content
if self.content_lengths.get(page_id, 0) > 0:
bookmark_item.setFlags(bookmark_item.flags() | Qt.ItemIsUserCheckable)
bookmark_item.setCheckState(0, Qt.Checked if page_id in self.checked_chapters else Qt.Unchecked)
bookmark_item.setFlags(
bookmark_item.flags() | Qt.ItemIsUserCheckable
)
bookmark_item.setCheckState(
0,
(
Qt.Checked
if page_id in self.checked_chapters
else Qt.Unchecked
),
)
else:
bookmark_item.setFlags(bookmark_item.flags() & ~Qt.ItemIsUserCheckable)
bookmark_item.setFlags(
bookmark_item.flags() & ~Qt.ItemIsUserCheckable
)
# map for uncategorized pages
self.bookmark_items_map[page_num] = bookmark_item
@@ -1027,10 +1035,21 @@ class HandlerDialog(QDialog):
page_item.setData(0, Qt.UserRole, page_id)
# only allow checking if this sub-page has content
if self.content_lengths.get(page_id, 0) > 0:
page_item.setFlags(page_item.flags() | Qt.ItemIsUserCheckable)
page_item.setCheckState(0, Qt.Checked if page_id in self.checked_chapters else Qt.Unchecked)
page_item.setFlags(
page_item.flags() | Qt.ItemIsUserCheckable
)
page_item.setCheckState(
0,
(
Qt.Checked
if page_id in self.checked_chapters
else Qt.Unchecked
),
)
else:
page_item.setFlags(page_item.flags() & ~Qt.ItemIsUserCheckable)
page_item.setFlags(
page_item.flags() & ~Qt.ItemIsUserCheckable
)
added_pages.add(sub_page_num)
@@ -1038,11 +1057,15 @@ class HandlerDialog(QDialog):
covered_pages = set(added_pages)
# attach any pages without direct bookmarks under nearest preceding chapter
uncategorized_pages = [i for i in range(len(self.pdf_doc)) if i not in covered_pages]
uncategorized_pages = [
i for i in range(len(self.pdf_doc)) if i not in covered_pages
]
for page_num in uncategorized_pages:
# find nearest previous bookmark
prev_nums = [n for n in sorted(self.bookmark_items_map) if n < page_num]
parent_item = self.bookmark_items_map[prev_nums[-1]] if prev_nums else self.treeWidget
parent_item = (
self.bookmark_items_map[prev_nums[-1]] if prev_nums else self.treeWidget
)
page_id = f"page_{page_num+1}"
title = f"Page {page_num+1}"
text = self.content_texts.get(page_id, "").strip()
@@ -1055,7 +1078,9 @@ class HandlerDialog(QDialog):
# only allow checking if uncategorized page has content
if self.content_lengths.get(page_id, 0) > 0:
page_item.setFlags(page_item.flags() | Qt.ItemIsUserCheckable)
page_item.setCheckState(0, Qt.Checked if page_id in self.checked_chapters else Qt.Unchecked)
page_item.setCheckState(
0, Qt.Checked if page_id in self.checked_chapters else Qt.Unchecked
)
else:
page_item.setFlags(page_item.flags() & ~Qt.ItemIsUserCheckable)
@@ -1081,7 +1106,9 @@ class HandlerDialog(QDialog):
# only allow checking if standalone page has content
if self.content_lengths.get(page_id, 0) > 0:
page_item.setFlags(page_item.flags() | Qt.ItemIsUserCheckable)
page_item.setCheckState(0, Qt.Checked if page_id in self.checked_chapters else Qt.Unchecked)
page_item.setCheckState(
0, Qt.Checked if page_id in self.checked_chapters else Qt.Unchecked
)
else:
page_item.setFlags(page_item.flags() & ~Qt.ItemIsUserCheckable)
@@ -1501,7 +1528,9 @@ class HandlerDialog(QDialog):
authors_text = ", ".join(self.book_metadata["authors"])
html_content += f"<p style='text-align: center; font-style: italic;'>By {authors_text}</p>"
if self.book_metadata["publisher"] or self.book_metadata.get("publication_year"):
if self.book_metadata["publisher"] or self.book_metadata.get(
"publication_year"
):
pub_info = []
if self.book_metadata["publisher"]:
pub_info.append(f"Published by {self.book_metadata['publisher']}")
@@ -1543,7 +1572,9 @@ class HandlerDialog(QDialog):
try:
author_items = self.book.get_metadata("DC", "creator")
if author_items:
metadata["authors"] = [author[0] for author in author_items if len(author) > 0]
metadata["authors"] = [
author[0] for author in author_items if len(author) > 0
]
except Exception as e:
logging.warning(f"Error extracting author metadata: {e}")
@@ -1567,7 +1598,7 @@ class HandlerDialog(QDialog):
if date_items and len(date_items) > 0:
date_str = date_items[0][0]
# Try to extract just the year from the date string
year_match = re.search(r'\b(19|20)\d{2}\b', date_str)
year_match = re.search(r"\b(19|20)\d{2}\b", date_str)
if year_match:
metadata["publication_year"] = year_match.group(0)
else:
@@ -1607,12 +1638,12 @@ class HandlerDialog(QDialog):
# Try to extract publication date from PDF metadata
if "creationDate" in pdf_info:
date_str = pdf_info["creationDate"]
year_match = re.search(r'D:(\d{4})', date_str)
year_match = re.search(r"D:(\d{4})", date_str)
if year_match:
metadata["publication_year"] = year_match.group(1)
elif "modDate" in pdf_info:
date_str = pdf_info["modDate"]
year_match = re.search(r'D:(\d{4})', date_str)
year_match = re.search(r"D:(\d{4})", date_str)
if year_match:
metadata["publication_year"] = year_match.group(1)
@@ -1644,11 +1675,15 @@ class HandlerDialog(QDialog):
authors = metadata.get("authors") or ["Unknown"]
authors_text = ", ".join(authors)
album_artist = authors_text or "Unknown"
year = metadata.get("publication_year") or current_year # Use publication year if available
year = (
metadata.get("publication_year") or current_year
) # Use publication year if available
# Count chapters/pages
total_chapters = len(self.checked_chapters)
chapter_text = f"{total_chapters} {'Chapters' if self.file_type == 'epub' else 'Pages'}"
chapter_text = (
f"{total_chapters} {'Chapters' if self.file_type == 'epub' else 'Pages'}"
)
# Format metadata tags
metadata_tags = [
@@ -1658,7 +1693,7 @@ class HandlerDialog(QDialog):
f"<<METADATA_YEAR:{year}>>",
f"<<METADATA_ALBUM_ARTIST:{album_artist}>>",
f"<<METADATA_COMPOSER:Narrator>>",
f"<<METADATA_GENRE:Audiobook>>"
f"<<METADATA_GENRE:Audiobook>>",
]
return "\n".join(metadata_tags)
@@ -1730,7 +1765,10 @@ class HandlerDialog(QDialog):
if text:
all_content.append(text)
included_text_ids.add(page_id)
return metadata_tags + "\n\n" + "\n\n".join(all_content), all_checked_identifiers
return (
metadata_tags + "\n\n" + "\n\n".join(all_content),
all_checked_identifiers,
)
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -1788,7 +1826,10 @@ class HandlerDialog(QDialog):
included_text_ids.add(identifier)
iterator += 1
return metadata_tags + "\n\n" + "\n\n".join([t[1] for t in section_titles]), all_checked_identifiers
return (
metadata_tags + "\n\n" + "\n\n".join([t[1] for t in section_titles]),
all_checked_identifiers,
)
def on_save_chapters_changed(self, state):
self.save_chapters_separately = bool(state)
@@ -1867,5 +1908,13 @@ class HandlerDialog(QDialog):
def closeEvent(self, event):
if self.pdf_doc is not None:
try:
if hasattr(self.pdf_doc, "is_closed"):
if not self.pdf_doc.is_closed:
self.pdf_doc.close()
else:
# Fallback: try/except close
self.pdf_doc.close()
except Exception:
pass
event.accept()
+8
View File
@@ -10,6 +10,13 @@ VERSION = get_version()
# Settings
CHAPTER_OPTIONS_COUNTDOWN = 30 # Countdown seconds for chapter options
SUBTITLE_FORMATS = [
("srt", "SRT (standard)"),
("ass_wide", "ASS (wide)"),
("ass_narrow", "ASS (narrow)"),
("ass_centered_wide", "ASS (centered wide)"),
("ass_centered_narrow", "ASS (centered narrow)"),
]
# Language description mapping
LANGUAGE_DESCRIPTIONS = {
@@ -109,6 +116,7 @@ SAMPLE_VOICE_TEXTS = {
COLORS = {
"BLUE": "#007dff",
"RED": "#c0392b",
"ORANGE": "#FFA500",
"GREEN": "#42ad4a",
"GREEN_BG": "rgba(66, 173, 73, 0.1)",
"GREEN_BG_HOVER": "rgba(66, 173, 73, 0.15)",
+528 -408
View File
File diff suppressed because it is too large Load Diff
+421 -159
View File
File diff suppressed because it is too large Load Diff
+10 -1
View File
@@ -1,16 +1,20 @@
log_callback = None
show_warning_signal_emitter = None # Renamed for clarity
def set_log_callback(cb):
global log_callback
log_callback = cb
def set_show_warning_signal_emitter(emitter): # Renamed for clarity
global show_warning_signal_emitter
show_warning_signal_emitter = emitter
from huggingface_hub import hf_hub_download
def tracked_hf_hub_download(*args, **kwargs):
try:
local_kwargs = dict(kwargs)
@@ -22,7 +26,10 @@ def tracked_hf_hub_download(*args, **kwargs):
if filename.endswith(".pth"):
msg = f"\nDownloading model '{filename}' from Hugging Face ({repo_id}). This may take a while. Please wait..."
if show_warning_signal_emitter: # Check if the emitter is set
show_warning_signal_emitter.emit("Downloading Model", f"Downloading model '{filename}' from Hugging Face repository '{repo_id}'. This may take a while, please wait.")
show_warning_signal_emitter.emit(
"Downloading Model",
f"Downloading model '{filename}' from Hugging Face repository '{repo_id}'. This may take a while, please wait.",
)
else:
msg = f"\nDownloading '{filename}' from Hugging Face ({repo_id}). Please wait..."
if log_callback:
@@ -32,5 +39,7 @@ def tracked_hf_hub_download(*args, **kwargs):
print(msg, flush=True)
return hf_hub_download(*args, **kwargs)
import huggingface_hub
huggingface_hub.hf_hub_download = tracked_hf_hub_download
+3 -1
View File
@@ -1,18 +1,20 @@
import gpustat
def check():
try:
stats = gpustat.new_query()
except Exception:
return False
nvidia_keywords = ['nvidia', 'rtx', 'gtx', 'quadro', 'tesla', 'titan', 'mx']
nvidia_keywords = ["nvidia", "rtx", "gtx", "quadro", "tesla", "titan", "mx"]
for gpu in stats.gpus:
name = gpu.name.lower()
if any(keyword in name for keyword in nvidia_keywords):
return True
return False
if __name__ == "__main__":
stats = gpustat.new_query()
for gpu in stats.gpus:
+14 -1
View File
@@ -1,3 +1,4 @@
from json import load
import os
import sys
import platform
@@ -7,8 +8,19 @@ from PyQt5.QtCore import qInstallMessageHandler, QtMsgType
# Add the directory to Python path
sys.path.insert(0, os.path.join(os.path.dirname(__file__)))
from abogen.utils import get_resource_path, load_config
# Set Hugging Face Hub environment variables
os.environ["HF_HUB_DISABLE_TELEMETRY"] = "1" # Disable Hugging Face telemetry
os.environ["HF_HUB_ETAG_TIMEOUT"] = "10" # Metadata request timeout (seconds)
os.environ["HF_HUB_DOWNLOAD_TIMEOUT"] = "10" # File download timeout (seconds)
os.environ["HF_HUB_DISABLE_SYMLINKS_WARNING"] = "1" # Disable symlinks warning
if load_config().get("disable_kokoro_internet", False):
print("INFO: Kokoro's internet access is disabled.")
os.environ["HF_HUB_OFFLINE"] = "1" # Disable Hugging Face Hub internet access
from abogen.gui import abogen
from abogen.utils import get_resource_path
from abogen.constants import PROGRAM_NAME, VERSION
# Set environment variables for AMD ROCm
@@ -25,6 +37,7 @@ if sys.stderr is None:
if platform.system() == "Darwin" and platform.processor() == "arm":
os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"
# Custom message handler to filter out specific Qt warnings
def qt_message_handler(mode, context, message):
if "Wayland does not support QWindow::requestActivate()" in message:
+118 -55
View File
@@ -20,6 +20,7 @@ from abogen.constants import COLORS
from copy import deepcopy
from PyQt5.QtGui import QFontMetrics
class ElidedLabel(QLabel):
def __init__(self, text, parent=None):
super().__init__(text, parent)
@@ -40,6 +41,7 @@ class ElidedLabel(QLabel):
def fullText(self):
return self._full_text
class QueueListItemWidget(QWidget):
def __init__(self, file_name, char_count):
super().__init__()
@@ -47,6 +49,7 @@ class QueueListItemWidget(QWidget):
layout.setContentsMargins(12, 0, 6, 0)
layout.setSpacing(0)
import os
name_label = ElidedLabel(os.path.basename(file_name))
char_label = QLabel(f"Chars: {char_count}")
char_label.setStyleSheet(f"color: {COLORS['LIGHT_DISABLED']};")
@@ -56,6 +59,7 @@ class QueueListItemWidget(QWidget):
layout.addWidget(char_label, 0)
self.setLayout(layout)
class DroppableQueueListWidget(QListWidget):
def __init__(self, parent_dialog):
super().__init__()
@@ -73,7 +77,7 @@ class DroppableQueueListWidget(QListWidget):
def dragEnterEvent(self, event):
if event.mimeData().hasUrls():
for url in event.mimeData().urls():
if url.isLocalFile() and url.toLocalFile().lower().endswith('.txt'):
if url.isLocalFile() and url.toLocalFile().lower().endswith(".txt"):
self.drag_overlay.resize(self.size())
self.drag_overlay.setVisible(True)
event.acceptProposedAction()
@@ -84,7 +88,7 @@ class DroppableQueueListWidget(QListWidget):
def dragMoveEvent(self, event):
if event.mimeData().hasUrls():
for url in event.mimeData().urls():
if url.isLocalFile() and url.toLocalFile().lower().endswith('.txt'):
if url.isLocalFile() and url.toLocalFile().lower().endswith(".txt"):
event.acceptProposedAction()
return
event.ignore()
@@ -96,7 +100,11 @@ class DroppableQueueListWidget(QListWidget):
def dropEvent(self, event):
self.drag_overlay.setVisible(False)
if event.mimeData().hasUrls():
file_paths = [url.toLocalFile() for url in event.mimeData().urls() if url.isLocalFile() and url.toLocalFile().lower().endswith('.txt')]
file_paths = [
url.toLocalFile()
for url in event.mimeData().urls()
if url.isLocalFile() and url.toLocalFile().lower().endswith(".txt")
]
if file_paths:
self.parent_dialog.add_files_from_paths(file_paths)
event.acceptProposedAction()
@@ -107,14 +115,17 @@ class DroppableQueueListWidget(QListWidget):
def resizeEvent(self, event):
super().resizeEvent(event)
if hasattr(self, 'drag_overlay'):
if hasattr(self, "drag_overlay"):
self.drag_overlay.resize(self.size())
class QueueManager(QDialog):
def __init__(self, parent, queue: list, title="Queue Manager", size=(600, 700)):
super().__init__()
self.queue = queue
self._original_queue = deepcopy(queue) # Store a deep copy of the original queue
self._original_queue = deepcopy(
queue
) # Store a deep copy of the original queue
self.parent = parent
layout = QVBoxLayout()
layout.setContentsMargins(15, 15, 15, 15) # set main layout margins
@@ -141,10 +152,12 @@ class QueueManager(QDialog):
# Overlay label for empty queue
self.empty_overlay = QLabel(
"Drag and drop your text files here or use the 'Add files' button.",
self.listwidget
self.listwidget,
)
self.empty_overlay.setAlignment(Qt.AlignCenter)
self.empty_overlay.setStyleSheet(f"color: {COLORS['LIGHT_DISABLED']}; background: transparent; padding: 20px;")
self.empty_overlay.setStyleSheet(
f"color: {COLORS['LIGHT_DISABLED']}; background: transparent; padding: 20px;"
)
self.empty_overlay.setWordWrap(True)
self.empty_overlay.setAttribute(Qt.WA_TransparentForMouseEvents, True)
self.empty_overlay.hide()
@@ -208,13 +221,14 @@ class QueueManager(QDialog):
file_name = item.file_name
display_name = file_name
import os
if os.path.sep in file_name:
display_name = os.path.basename(file_name)
# Get icon for the file
icon = icon_provider.icon(QFileInfo(file_name))
list_item = QListWidgetItem()
# Set tooltip with detailed info
output_folder = getattr(item, 'output_folder', '')
output_folder = getattr(item, "output_folder", "")
tooltip = (
f"<b>Path:</b> {file_name}<br>"
f"<b>Language:</b> {getattr(item, 'lang_code', '')}<br>"
@@ -222,7 +236,7 @@ class QueueManager(QDialog):
f"<b>Voice:</b> {getattr(item, 'voice', '')}<br>"
f"<b>Save Option:</b> {getattr(item, 'save_option', '')}<br>"
)
if output_folder not in (None, '', 'None'):
if output_folder not in (None, "", "None"):
tooltip += f"<b>Output Folder:</b> {output_folder}<br>"
tooltip += (
f"<b>Subtitle Mode:</b> {getattr(item, 'subtitle_mode', '')}<br>"
@@ -232,8 +246,9 @@ class QueueManager(QDialog):
)
list_item.setToolTip(tooltip)
list_item.setIcon(icon)
list_item.setData(Qt.UserRole, file_name)
# Use custom widget for display
char_count = getattr(item, 'total_char_count', 0)
char_count = getattr(item, "total_char_count", 0)
widget = QueueListItemWidget(file_name, char_count)
self.listwidget.addItem(list_item)
self.listwidget.setItemWidget(list_item, widget)
@@ -244,6 +259,7 @@ class QueueManager(QDialog):
if not items:
return
from PyQt5.QtWidgets import QMessageBox
# Remove by index to ensure correct mapping
rows = sorted([self.listwidget.row(item) for item in items], reverse=True)
# Warn user if removing multiple files
@@ -253,7 +269,7 @@ class QueueManager(QDialog):
"Confirm Remove",
f"Are you sure you want to remove {len(rows)} selected items from the queue?",
QMessageBox.Yes | QMessageBox.No,
QMessageBox.No
QMessageBox.No,
)
if reply != QMessageBox.Yes:
return
@@ -265,19 +281,22 @@ class QueueManager(QDialog):
def clear_queue(self):
from PyQt5.QtWidgets import QMessageBox
if len(self.queue) > 1:
reply = QMessageBox.question(
self,
"Confirm Clear Queue",
f"Are you sure you want to clear {len(self.queue)} items from the queue?",
QMessageBox.Yes | QMessageBox.No,
QMessageBox.No
QMessageBox.No,
)
if reply != QMessageBox.Yes:
return
self.queue.clear()
self.listwidget.clear()
self.empty_overlay.resize(self.listwidget.size()) # Ensure overlay is sized correctly
self.empty_overlay.resize(
self.listwidget.size()
) # Ensure overlay is sized correctly
self.empty_overlay.show() # Show the overlay when queue is empty
self.update_button_states()
@@ -290,56 +309,74 @@ class QueueManager(QDialog):
parent = self.parent
if parent is not None:
# lang_code: use parent's get_voice_formula and get_selected_lang
if hasattr(parent, 'get_voice_formula') and hasattr(parent, 'get_selected_lang'):
if hasattr(parent, "get_voice_formula") and hasattr(
parent, "get_selected_lang"
):
voice_formula = parent.get_voice_formula()
attrs['lang_code'] = parent.get_selected_lang(voice_formula)
attrs['voice'] = voice_formula
attrs["lang_code"] = parent.get_selected_lang(voice_formula)
attrs["voice"] = voice_formula
else:
attrs['lang_code'] = getattr(parent, 'selected_lang', '')
attrs['voice'] = getattr(parent, 'selected_voice', '')
attrs["lang_code"] = getattr(parent, "selected_lang", "")
attrs["voice"] = getattr(parent, "selected_voice", "")
# speed
if hasattr(parent, 'speed_slider'):
attrs['speed'] = parent.speed_slider.value() / 100.0
if hasattr(parent, "speed_slider"):
attrs["speed"] = parent.speed_slider.value() / 100.0
else:
attrs['speed'] = getattr(parent, 'speed', 1.0)
attrs["speed"] = getattr(parent, "speed", 1.0)
# save_option
attrs['save_option'] = getattr(parent, 'save_option', '')
attrs["save_option"] = getattr(parent, "save_option", "")
# output_folder
attrs['output_folder'] = getattr(parent, 'selected_output_folder', '')
attrs["output_folder"] = getattr(parent, "selected_output_folder", "")
# subtitle_mode
if hasattr(parent, 'get_actual_subtitle_mode'):
attrs['subtitle_mode'] = parent.get_actual_subtitle_mode()
if hasattr(parent, "get_actual_subtitle_mode"):
attrs["subtitle_mode"] = parent.get_actual_subtitle_mode()
else:
attrs['subtitle_mode'] = getattr(parent, 'subtitle_mode', '')
attrs["subtitle_mode"] = getattr(parent, "subtitle_mode", "")
# output_format
attrs['output_format'] = getattr(parent, 'selected_format', '')
attrs["output_format"] = getattr(parent, "selected_format", "")
# total_char_count
attrs['total_char_count'] = getattr(parent, 'char_count', '')
attrs["total_char_count"] = getattr(parent, "char_count", "")
# replace_single_newlines
attrs['replace_single_newlines'] = getattr(parent, 'replace_single_newlines', False)
attrs["replace_single_newlines"] = getattr(
parent, "replace_single_newlines", False
)
else:
# fallback: empty values
attrs = {k: '' for k in [
'lang_code', 'speed', 'voice', 'save_option',
'output_folder', 'subtitle_mode', 'output_format', 'total_char_count', 'replace_single_newlines']}
attrs = {
k: ""
for k in [
"lang_code",
"speed",
"voice",
"save_option",
"output_folder",
"subtitle_mode",
"output_format",
"total_char_count",
"replace_single_newlines",
]
}
return attrs
def add_files_from_paths(self, file_paths):
from abogen.utils import calculate_text_length
from PyQt5.QtWidgets import QMessageBox
import os
current_attrs = self.get_current_attributes()
duplicates = []
for file_path in file_paths:
class QueueItem:
pass
item = QueueItem()
item.file_name = file_path
for attr, value in current_attrs.items():
setattr(item, attr, value)
# Read file content and calculate total_char_count using calculate_text_length
try:
with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
with open(file_path, "r", encoding="utf-8", errors="ignore") as f:
file_content = f.read()
item.total_char_count = calculate_text_length(file_content)
except Exception:
@@ -348,16 +385,26 @@ class QueueManager(QDialog):
is_duplicate = False
for queued_item in self.queue:
if (
getattr(queued_item, 'file_name', None) == getattr(item, 'file_name', None) and
getattr(queued_item, 'lang_code', None) == getattr(item, 'lang_code', None) and
getattr(queued_item, 'speed', None) == getattr(item, 'speed', None) and
getattr(queued_item, 'voice', None) == getattr(item, 'voice', None) and
getattr(queued_item, 'save_option', None) == getattr(item, 'save_option', None) and
getattr(queued_item, 'output_folder', None) == getattr(item, 'output_folder', None) and
getattr(queued_item, 'subtitle_mode', None) == getattr(item, 'subtitle_mode', None) and
getattr(queued_item, 'output_format', None) == getattr(item, 'output_format', None) and
getattr(queued_item, 'total_char_count', None) == getattr(item, 'total_char_count', None) and
getattr(queued_item, 'replace_single_newlines', False) == getattr(item, 'replace_single_newlines', False)
getattr(queued_item, "file_name", None)
== getattr(item, "file_name", None)
and getattr(queued_item, "lang_code", None)
== getattr(item, "lang_code", None)
and getattr(queued_item, "speed", None)
== getattr(item, "speed", None)
and getattr(queued_item, "voice", None)
== getattr(item, "voice", None)
and getattr(queued_item, "save_option", None)
== getattr(item, "save_option", None)
and getattr(queued_item, "output_folder", None)
== getattr(item, "output_folder", None)
and getattr(queued_item, "subtitle_mode", None)
== getattr(item, "subtitle_mode", None)
and getattr(queued_item, "output_format", None)
== getattr(item, "output_format", None)
and getattr(queued_item, "total_char_count", None)
== getattr(item, "total_char_count", None)
and getattr(queued_item, "replace_single_newlines", False)
== getattr(item, "replace_single_newlines", False)
):
is_duplicate = True
break
@@ -369,7 +416,7 @@ class QueueManager(QDialog):
QMessageBox.warning(
self,
"Duplicate Item(s)",
f"Skipping {len(duplicates)} file(s) with the same attributes, already in the queue."
f"Skipping {len(duplicates)} file(s) with the same attributes, already in the queue.",
)
self.process_queue()
self.update_button_states()
@@ -377,20 +424,23 @@ class QueueManager(QDialog):
def add_more_files(self):
from PyQt5.QtWidgets import QFileDialog
from abogen.utils import calculate_text_length # import the function
# Only allow .txt files
files, _ = QFileDialog.getOpenFileNames(self, "Select .txt files", "", "Text Files (*.txt)")
files, _ = QFileDialog.getOpenFileNames(
self, "Select .txt files", "", "Text Files (*.txt)"
)
if not files:
return
self.add_files_from_paths(files)
def resizeEvent(self, event):
super().resizeEvent(event)
if hasattr(self, 'empty_overlay'):
if hasattr(self, "empty_overlay"):
self.empty_overlay.resize(self.listwidget.size())
def update_button_states(self):
# Enable Remove if at least one item is selected, else disable
if hasattr(self, 'remove_button'):
if hasattr(self, "remove_button"):
selected_count = len(self.listwidget.selectedItems())
self.remove_button.setEnabled(selected_count > 0)
if selected_count > 1:
@@ -398,7 +448,7 @@ class QueueManager(QDialog):
else:
self.remove_button.setText("Remove selected")
# Disable Clear if queue is empty
if hasattr(self, 'clear_button'):
if hasattr(self, "clear_button"):
self.clear_button.setEnabled(bool(self.queue))
def show_context_menu(self, pos):
@@ -406,6 +456,7 @@ class QueueManager(QDialog):
from PyQt5.QtGui import QDesktopServices
from PyQt5.QtCore import QUrl
import os
global_pos = self.listwidget.viewport().mapToGlobal(pos)
selected_items = self.listwidget.selectedItems()
menu = QMenu(self)
@@ -417,35 +468,45 @@ class QueueManager(QDialog):
# Add Open file action
open_file_action = QAction("Open file", self)
def open_file():
from PyQt5.QtWidgets import QMessageBox
item = selected_items[0]
display_name = item.text()
file_path = item.data(Qt.UserRole)
for q in self.queue:
if os.path.basename(q.file_name) == display_name:
if q.file_name == file_path:
if not os.path.exists(q.file_name):
QMessageBox.warning(self, "File Not Found", f"The file does not exist.")
QMessageBox.warning(
self, "File Not Found", f"The file does not exist."
)
return
QDesktopServices.openUrl(QUrl.fromLocalFile(q.file_name))
break
open_file_action.triggered.connect(open_file)
menu.addAction(open_file_action)
# Add Go to folder action
go_to_folder_action = QAction("Go to folder", self)
def go_to_folder():
from PyQt5.QtWidgets import QMessageBox
item = selected_items[0]
display_name = item.text()
file_path = item.data(Qt.UserRole)
for q in self.queue:
if os.path.basename(q.file_name) == display_name:
if q.file_name == file_path:
if not os.path.exists(q.file_name):
QMessageBox.warning(self, "File Not Found", f"The file does not exist.")
QMessageBox.warning(
self, "File Not Found", f"The file does not exist."
)
return
folder = os.path.dirname(q.file_name)
if os.path.exists(folder):
QDesktopServices.openUrl(QUrl.fromLocalFile(folder))
break
go_to_folder_action.triggered.connect(go_to_folder)
menu.addAction(go_to_folder_action)
@@ -466,6 +527,7 @@ class QueueManager(QDialog):
def reject(self):
# Cancel: restore original queue
from PyQt5.QtWidgets import QMessageBox
# Warn if user changed a lot (e.g., more than 1 items difference)
original_count = len(self._original_queue)
current_count = len(self.queue)
@@ -475,7 +537,7 @@ class QueueManager(QDialog):
"Confirm Cancel",
f"Are you sure you want to cancel and discard all changes?",
QMessageBox.Yes | QMessageBox.No,
QMessageBox.No
QMessageBox.No,
)
if reply != QMessageBox.Yes:
return
@@ -485,6 +547,7 @@ class QueueManager(QDialog):
def keyPressEvent(self, event):
from PyQt5.QtCore import Qt
if event.key() == Qt.Key_Delete:
self.remove_item()
else:
+1
View File
@@ -1,6 +1,7 @@
# represents a queued item - book, chapters, voice, etc.
from dataclasses import dataclass
@dataclass
class QueuedItem:
file_name: str
+19 -7
View File
@@ -7,8 +7,6 @@ import subprocess
import re
from threading import Thread
# suppress warnings and disable HF hub symlink warnings
os.environ["HF_HUB_DISABLE_SYMLINKS_WARNING"] = "1"
warnings.filterwarnings("ignore")
@@ -105,15 +103,17 @@ def clean_text(text, *args, **kwargs):
default_encoding = sys.getfilesystemencoding()
def create_process(cmd, stdin=None, text=True, capture_output=False):
import logging
logger = logging.getLogger(__name__)
# Configure root logger to output to console if not already configured
root = logging.getLogger()
if not root.handlers:
handler = logging.StreamHandler(sys.stdout)
formatter = logging.Formatter('%(message)s')
formatter = logging.Formatter("%(message)s")
handler.setFormatter(formatter)
root.addHandler(handler)
root.setLevel(logging.INFO)
@@ -156,6 +156,7 @@ def create_process(cmd, stdin=None, text=True, capture_output=False):
# Stream output to console in real-time if not capturing
if proc.stdout and not capture_output:
def _stream_output(stream):
if text:
# For text mode, read character by character for real-time output
@@ -174,7 +175,9 @@ def create_process(cmd, stdin=None, text=True, capture_output=False):
break
try:
# Try to decode binary data for display
sys.stdout.write(chunk.decode(default_encoding, errors='replace'))
sys.stdout.write(
chunk.decode(default_encoding, errors="replace")
)
sys.stdout.flush()
except Exception:
pass
@@ -220,6 +223,7 @@ def get_gpu_acceleration(enabled):
try:
import torch
from torch.cuda import is_available
if not enabled:
return "CUDA GPU available but using CPU.", False
if is_available():
@@ -227,7 +231,11 @@ def get_gpu_acceleration(enabled):
# Gather more diagnostic info if CUDA is not available
try:
cuda_devices = torch.cuda.device_count()
cuda_error = torch.cuda.get_device_name(0) if cuda_devices > 0 else "No devices found"
cuda_error = (
torch.cuda.get_device_name(0)
if cuda_devices > 0
else "No devices found"
)
except Exception as e:
cuda_error = str(e)
return f"CUDA GPU is not available. Using CPU. ({cuda_error})", False
@@ -239,6 +247,7 @@ def prevent_sleep_start():
system = platform.system()
if system == "Windows":
import ctypes
ctypes.windll.kernel32.SetThreadExecutionState(
0x80000000 | 0x00000001 | 0x00000040
)
@@ -246,19 +255,22 @@ def prevent_sleep_start():
_sleep_procs["Darwin"] = create_process(["caffeinate"])
elif system == "Linux":
# use a sleep that never exits so inhibition stays active
_sleep_procs["Linux"] = create_process([
_sleep_procs["Linux"] = create_process(
[
"systemd-inhibit",
"--what=handle-lid-switch:sleep",
"--mode=block",
"sleep",
"infinity",
])
]
)
def prevent_sleep_end():
system = platform.system()
if system == "Windows":
import ctypes
ctypes.windll.kernel32.SetThreadExecutionState(0x80000000) # ES_CONTINUOUS
elif system in ("Darwin", "Linux") and _sleep_procs[system]:
try:
+15 -5
View File
@@ -230,13 +230,15 @@ class VoiceMixer(QWidget):
appstyle = QApplication.instance().style().objectName().lower()
if appstyle != "windowsvista":
# Set custom groove color for disabled state using COLORS["GREY_BACKGROUND"]
self.slider.setStyleSheet(f"""
self.slider.setStyleSheet(
f"""
QSlider::groove:vertical:disabled {{
background: {COLORS.get("GREY_BACKGROUND")};
width: 4px;
border-radius: 4px;
}}
""")
"""
)
# Connect controls
self.slider.valueChanged.connect(lambda val: self.spin_box.setValue(val / 100))
@@ -330,7 +332,9 @@ class VoiceFormulaDialog(QDialog):
self._original_mixed_voice_state = None
if parent is not None:
self._original_profile_name = getattr(parent, "selected_profile_name", None)
self._original_mixed_voice_state = getattr(parent, "mixed_voice_state", None)
self._original_mixed_voice_state = getattr(
parent, "mixed_voice_state", None
)
profiles = load_profiles()
self._virtual_new_profile = False
if not profiles:
@@ -369,7 +373,9 @@ class VoiceFormulaDialog(QDialog):
self.profile_list = QListWidget()
self.profile_list.setSelectionMode(QListWidget.SingleSelection)
self.profile_list.setSelectionBehavior(QListWidget.SelectRows)
self.profile_list.setStyleSheet("QListWidget::item:selected { background: palette(highlight); color: palette(highlighted-text); }")
self.profile_list.setStyleSheet(
"QListWidget::item:selected { background: palette(highlight); color: palette(highlighted-text); }"
)
icon = QIcon(get_resource_path("abogen.assets", "profile.png"))
if self._virtual_new_profile:
item = QListWidgetItem(icon, "New profile")
@@ -976,6 +982,7 @@ class VoiceFormulaDialog(QDialog):
def _parse_rgba_to_qcolor(self, rgba_str):
from PyQt5.QtCore import Qt
from PyQt5.QtGui import QColor
"""Helper to convert 'rgba(R,G,B,A_float)' string to QColor."""
match = re.match(r"rgba\((\d+),\s*(\d+),\s*(\d+),\s*([\d.]+)\)", rgba_str)
if match:
@@ -1372,6 +1379,7 @@ class VoiceFormulaDialog(QDialog):
def update_profile_list_colors(self):
from PyQt5.QtCore import Qt
profiles = load_profiles()
for i in range(self.profile_list.count()):
item = self.profile_list.item(i)
@@ -1383,7 +1391,9 @@ class VoiceFormulaDialog(QDialog):
color = self._parse_rgba_to_qcolor(COLORS.get("YELLOW_BACKGROUND"))
item.setData(Qt.BackgroundRole, color)
else:
item.setData(Qt.BackgroundRole, self.profile_list.palette().base().color())
item.setData(
Qt.BackgroundRole, self.profile_list.palette().base().color()
)
weights = profiles.get(name, {}).get("voices", [])
total = 0
if isinstance(weights, list):
+1
View File
@@ -1,6 +1,7 @@
import re
from abogen.constants import VOICES_INTERNAL
# Calls parsing and loads the voice to gpu or cpu
def get_new_voice(pipeline, formula, use_gpu):
try: