v1.1.3

CHANGELOG.md

@@ -1,3 +1,21 @@
+# 1.1.3
+- `M4B (with chapters)` generation is faster now, as it directly generates `m4b` files instead of converting from `wav`, which significantly reduces processing time.
+- Better sleep state handling for Linux.
+- The app window now tries to fit the screen if its height would exceed the available display area.
+- Fixed issue where the app would not restart properly on Windows.
+- Fixed last sentence/subtitle entry timing in generated subtitles, the end time of the final subtitle entry now correctly matches the end of the audio chunk, preventing zero or invalid timings at the end.
+
+# 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.

README.md

@@ -153,6 +153,8 @@ Here’s 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'>

WINDOWS_INSTALL.bat

@@ -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

abogen/VERSION

@@ -1 +1 @@
-1.1.1
+1.1.3

abogen/book_handler.py

@@ -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
@@ -1911,5 +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()

abogen/constants.py

@@ -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)",

abogen/gui.py

@@ -75,6 +75,7 @@ from abogen.constants import (
     VOICES_INTERNAL,
     SUPPORTED_LANGUAGES_FOR_SUBTITLE_GENERATION,
     COLORS,
+    SUBTITLE_FORMATS,
 )
 import threading
 from abogen.voice_formula_gui import VoiceFormulaDialog
@@ -124,6 +125,18 @@ class IconProvider(QFileIconProvider):
         return super().icon(fileInfo)
 
 
+LOG_COLOR_MAP = {
+    True: COLORS["GREEN"],
+    False: COLORS["RED"],
+    "red": COLORS["RED"],
+    "green": COLORS["GREEN"],
+    "orange": COLORS["ORANGE"],
+    "blue": COLORS["BLUE"],
+    "grey": COLORS["LIGHT_DISABLED"],
+    None: COLORS["LIGHT_DISABLED"],
+}
+
+
 class InputBox(QLabel):
     # Define CSS styles as class constants
     STYLE_DEFAULT = f"border:2px dashed #aaa; border-radius:5px; padding:20px; background:{COLORS['BLUE_BG']}; min-height:100px;"
@@ -646,7 +659,7 @@ class abogen(QWidget):
             None  # Add new variable to track the displayed file path
         )
         # Max log lines
-        self.log_window_max_lines = self.config.get("log_window_max_lines", 5000)
+        self.log_window_max_lines = self.config.get("log_window_max_lines", 2000)
         self.selected_chapters = set()
         self.last_opened_book_path = None  # Track the last opened book path
         self.last_output_path = None
@@ -700,9 +713,6 @@ class abogen(QWidget):
         self.queued_items = []
         self.current_queue_index = 0
 
-        # Log lines
-        self._log_lines = []
-
         self.initUI()
         self.speed_slider.setValue(int(self.config.get("speed", 1.00) * 100))
         self.update_speed_label()
@@ -753,7 +763,11 @@ class abogen(QWidget):
         self.setWindowTitle(f"{PROGRAM_NAME} v{VERSION}")
         screen = QApplication.primaryScreen().geometry()
         width, height = 500, 800
-        x, y = (screen.width() - width) // 2, (screen.height() - height) // 2
+        x = (screen.width() - width) // 2
+        # If desired height is larger than screen, fit to screen height
+        if height > screen.height() - 65:
+            height = screen.height() - 100  # Leave a margin for window borders
+        y = max((screen.height() - height) // 2, 0)
         self.setGeometry(x, y, width, height)
         outer_layout = QVBoxLayout()
         outer_layout.setContentsMargins(15, 15, 15, 15)
@@ -785,6 +799,7 @@ class abogen(QWidget):
         container_layout.addWidget(self.queue_row_widget)
         self.log_text = QTextEdit(self)
         self.log_text.setReadOnly(True)
+        self.log_text.setUndoRedoEnabled(False)
         self.log_text.setFrameStyle(QTextEdit.NoFrame)
         self.log_text.setStyleSheet("QTextEdit { border: none; }")
         self.log_text.hide()
@@ -944,14 +959,7 @@ class abogen(QWidget):
         self.subtitle_format_combo.setSizePolicy(
             QSizePolicy.Expanding, QSizePolicy.Fixed
         )
-        subtitle_formats = [
+        for value, text in 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)"),
-        ]
-        for value, text in subtitle_formats:
             self.subtitle_format_combo.addItem(text, value)
         subtitle_format = self.config.get("subtitle_format", "ass_centered_narrow")
         idx = self.subtitle_format_combo.findData(subtitle_format)
@@ -1458,16 +1466,12 @@ class abogen(QWidget):
         self.input_box.hide()
         self.log_text.show()
         self.log_text.clear()
-        self._log_lines = []
         QApplication.processEvents()
 
     def restore_input_box(self):
         self.log_text.hide()
         self.input_box.show()
 
-    def color_text(self, message, color):
-        return f'<span style="color:{color};">{message.replace(chr(10), "<br>")}</span><br>'
-
     def update_log(self, message):
         # Use signal-based approach for thread-safe logging
         if QThread.currentThread() != QApplication.instance().thread():
@@ -1479,38 +1483,36 @@ class abogen(QWidget):
         self._update_log_main_thread(message)
 
     def _update_log_main_thread(self, message):
-        if not hasattr(self, "_log_lines"):
+        txt = self.log_text
-            self._log_lines = []
+        sb = txt.verticalScrollBar()
+        at_bottom = sb.value() == sb.maximum()
 
-        # Always produce a single HTML line (with color if tuple)
+        cursor = txt.textCursor()
+        cursor.movePosition(QTextCursor.End)
+
+        fmt = cursor.charFormat()
         if isinstance(message, tuple):
             text, spec = message
-            color = "green" if spec is True else ("red" if spec is False else spec)
+            fmt.setForeground(QColor(LOG_COLOR_MAP.get(spec, COLORS["LIGHT_DISABLED"])))
-            html_line = self.color_text(text, color)
         else:
-            html_line = f"{str(message).replace(chr(10), '<br>')}<br>"
+            text = str(message)
+            fmt.clearForeground()
+        cursor.setCharFormat(fmt)
+        cursor.insertText(text + "\n")
 
-        self._log_lines.append(html_line)
+        doc = txt.document()
+        excess = doc.blockCount() - self.log_window_max_lines
+        if excess > 0:
+            start = doc.findBlockByNumber(0).position()
+            end = doc.findBlockByNumber(excess).position()
+            trim_cursor = QTextCursor(doc)
+            trim_cursor.setPosition(start)
+            trim_cursor.setPosition(end, QTextCursor.KeepAnchor)
+            trim_cursor.removeSelectedText()
 
-        # Check if user is at bottom BEFORE inserting
-        sb = self.log_text.verticalScrollBar()
-        at_bottom = sb.value() >= sb.maximum() - 2
-
-        # Trim buffer if needed
-        if len(self._log_lines) > self.log_window_max_lines:
-            self._log_lines = self._log_lines[-self.log_window_max_lines :]
-            self.log_text.clear()
-            for line in self._log_lines:
-                self.log_text.insertHtml(line)
-        else:
-            self.log_text.insertHtml(html_line)
-
-        # If user was at bottom before, scroll to new bottom
         if at_bottom:
             sb.setValue(sb.maximum())
 
-        QApplication.processEvents()
-
     def _get_queue_progress_format(self, value=None):
         """Return the progress bar format string for queue mode."""
         if (
@@ -1889,7 +1891,7 @@ class abogen(QWidget):
                 f"<span style='color:{COLORS['GREEN']}; font-weight:bold;'>{idx}) {os.path.basename(item.file_name)}</span><br>"
                 f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Language:</span> {item.lang_code}<br>"
                 f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Voice:</span> {item.voice}<br>"
-                f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Speed:</span> {item.speed * 100:.1f}%<br>"
+                f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Speed:</span> {item.speed}<br>"
                 f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Characters:</span> {item.total_char_count}<br>"
                 f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Input:</span> {item.file_name}<br>"
                 f"<span style='color:{COLORS['LIGHT_DISABLED']};'>Output:</span> {output}</span>"
@@ -1988,7 +1990,6 @@ class abogen(QWidget):
         else:
             # More items in queue: clear log and reload for next item
             self.log_text.clear()
-            self._log_lines = []
             QApplication.processEvents()
 
         # Start new queued item, if we're using a queued conversion
@@ -2032,7 +2033,6 @@ class abogen(QWidget):
         self.progress_bar.hide()
         self.restore_input_box()
         self.log_text.clear()
-        self._log_lines = []
 
         # Use displayed_file_path instead of selected_file for EPUBs or PDFs
         display_path = (
@@ -2444,7 +2444,6 @@ class abogen(QWidget):
             self.finish_widget.hide()
             self.restore_input_box()
             self.log_text.clear()
-            self._log_lines = []
             display_path = (
                 self.displayed_file_path
                 if self.displayed_file_path
@@ -2484,6 +2483,16 @@ class abogen(QWidget):
         self.config["use_gpu"] = self.use_gpu
         save_config(self.config)
 
+    def cleanup_conversion_thread(self):
+        # Stop conversion thread
+        if (
+            hasattr(self, "conversion_thread")
+            and self.conversion_thread is not None
+            and self.conversion_thread.isRunning()
+        ):
+            self.conversion_thread.cancel()
+            self.conversion_thread.wait()
+
     def closeEvent(self, event):
         if self.is_converting:
             box = QMessageBox(self)
@@ -2495,16 +2504,12 @@ class abogen(QWidget):
             box.setStandardButtons(QMessageBox.Yes | QMessageBox.No)
             box.setDefaultButton(QMessageBox.No)
             if box.exec_() == QMessageBox.Yes:
-                if (
+                self.cleanup_conversion_thread()
-                    hasattr(self, "conversion_thread")
-                    and self.conversion_thread.isRunning()
-                ):
-                    self.conversion_thread.cancel()
-                    self.conversion_thread.wait()
                 event.accept()
             else:
                 event.ignore()
         else:
+            self.cleanup_conversion_thread()
             event.accept()
 
     def show_chapter_options_dialog(self, chapter_count):
@@ -2779,6 +2784,17 @@ class abogen(QWidget):
         # Add seperator"
         menu.addSeparator()
 
+        # Add "Disable Kokoro's internet access" option
+        disable_kokoro_action = QAction("Disable Kokoro's internet access", self)
+        disable_kokoro_action.setCheckable(True)
+        disable_kokoro_action.setChecked(
+            self.config.get("disable_kokoro_internet", False)
+        )
+        disable_kokoro_action.triggered.connect(
+            lambda checked: self.toggle_kokoro_internet_access(checked)
+        )
+        menu.addAction(disable_kokoro_action)
+
         # Add check for updates option
         check_updates_action = QAction("Check for updates at startup", self)
         check_updates_action.setCheckable(True)
@@ -2786,6 +2802,11 @@ class abogen(QWidget):
         check_updates_action.triggered.connect(self.toggle_check_updates)
         menu.addAction(check_updates_action)
 
+        # Add "Reset to default settings" option
+        reset_defaults_action = QAction("Reset to default settings", self)
+        reset_defaults_action.triggered.connect(self.reset_to_default_settings)
+        menu.addAction(reset_defaults_action)
+
         # Add about action
         about_action = QAction("About", self)
         about_action.triggered.connect(self.show_about_dialog)
@@ -2798,6 +2819,73 @@ class abogen(QWidget):
         self.config["replace_single_newlines"] = enabled
         save_config(self.config)
 
+    def restart_app(self):
+        from PyQt5.QtCore import QProcess
+        import sys
+        exe = sys.executable
+        args = sys.argv
+
+        # On Windows, use .exe if available
+        if platform.system() == "Windows":
+            script_path = args[0]
+            if not script_path.lower().endswith('.exe'):
+                exe_path = os.path.splitext(script_path)[0] + '.exe'
+                if os.path.exists(exe_path):
+                    args[0] = exe_path
+
+        QProcess.startDetached(exe, args)
+        QApplication.quit()
+
+    def toggle_kokoro_internet_access(self, disabled):
+        if disabled:
+            message = (
+                "Disabling Kokoro's internet access will block downloads of models and voices from Hugging Face Hub. "
+                "This can make processing faster when there is no internet connection, since no requests will be made. "
+                "The app needs to restart to apply this change.\n\nDo you want to continue?"
+            )
+        else:
+            message = (
+                "Enabling Kokoro's internet access will allow it to download models and voices from Hugging Face Hub. "
+                "The app needs to restart to apply this change.\n\nDo you want to continue?"
+            )
+        reply = QMessageBox.question(
+            self,
+            "Restart Required",
+            message,
+            QMessageBox.Yes | QMessageBox.No,
+            QMessageBox.No,
+        )
+        if reply == QMessageBox.Yes:
+            self.config["disable_kokoro_internet"] = disabled
+            save_config(self.config)
+            try:
+                self.restart_app()
+            except Exception as e:
+                QMessageBox.critical(
+                    self, "Restart Failed", f"Failed to restart the application:\n{e}"
+                )
+
+    def reset_to_default_settings(self):
+        reply = QMessageBox.question(
+            self,
+            "Reset Settings",
+            "This will reset all settings to their default values and restart the application.\n\nDo you want to continue?",
+            QMessageBox.Yes | QMessageBox.No,
+            QMessageBox.No,
+        )
+        if reply == QMessageBox.Yes:
+            from abogen.utils import get_user_config_path
+
+            config_path = get_user_config_path()
+            try:
+                if os.path.exists(config_path):
+                    os.remove(config_path)
+                self.restart_app()
+            except Exception as e:
+                QMessageBox.critical(
+                    self, "Reset Error", f"Could not reset settings:\n{e}"
+                )
+
     def reveal_config_in_explorer(self):
         """Open the configuration file location in file explorer."""
         from abogen.utils import get_user_config_path

abogen/main.py

@@ -1,20 +1,36 @@
+from json import load
 import os
 import sys
 import platform
+import atexit
 from PyQt5.QtWidgets import QApplication
 from PyQt5.QtGui import QIcon
 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, prevent_sleep_end
+
+# 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
 os.environ["MIOPEN_FIND_MODE"] = "FAST"
 os.environ["MIOPEN_CONV_PRECISE_ROCM_TUNING"] = "0"
 
+# Reset sleep states
+atexit.register(prevent_sleep_end)
+
 # Ensure sys.stdout and sys.stderr are valid in GUI mode
 if sys.stdout is None:
     sys.stdout = open(os.devnull, "w")

abogen/queue_manager_gui.py

@@ -246,6 +246,7 @@ 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)
             widget = QueueListItemWidget(file_name, char_count)
@@ -472,9 +473,9 @@ class QueueManager(QDialog):
                 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."
@@ -493,9 +494,9 @@ class QueueManager(QDialog):
                 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."

abogen/utils.py

@@ -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")
 
 
@@ -246,6 +244,7 @@ def get_gpu_acceleration(enabled):
 
 
 def prevent_sleep_start():
+    from abogen.constants import PROGRAM_NAME
     system = platform.system()
     if system == "Windows":
         import ctypes
@@ -256,11 +255,15 @@ def prevent_sleep_start():
     elif system == "Darwin":
         _sleep_procs["Darwin"] = create_process(["caffeinate"])
     elif system == "Linux":
-        # use a sleep that never exits so inhibition stays active
+        # Add program name and reason for inhibition
+        program_name = PROGRAM_NAME
+        reason = "Prevent sleep during abogen process"
         _sleep_procs["Linux"] = create_process(
             [
                 "systemd-inhibit",
-                "--what=handle-lid-switch:sleep",
+                f"--who={program_name}",
+                f"--why={reason}",
+                "--what=sleep",
                 "--mode=block",
                 "sleep",
                 "infinity",