1.1.7

Add word-by-word karaoke highlighting option

.github/workflows/test_publish.yml

@@ -5,7 +5,7 @@ on:
   #release:
   #  types: [published]
   # Build only
-  #push:
+  #push: it 
   #  branches: [main]
   # TODO - enable build on pull requests if build times can be reduced
   # pull_request:
@@ -22,7 +22,7 @@ jobs:
 
     - name: Login to Github Container Registry
       # Only if we need to push an image
-      if: ${{ github.event_name == 'release' && github.event.action == 'published' }}
+      # if: ${{ github.event_name == 'release' && github.event.action == 'published' }}
       uses: docker/login-action@v3
       with:
           registry: ghcr.io
@@ -51,11 +51,13 @@ jobs:
         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,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: ${{ 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 || 'temp' }}
+        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

CHANGELOG.md

@@ -1,3 +1,16 @@
+# 1.1.7
+- Added MPS GPU acceleration support for Silicon Mac, mentioned in https://github.com/denizsafak/abogen/issues/32#issuecomment-3155902040 by @jefro108. **Please read the [Mac](https://github.com/denizsafak/abogen?tab=readme-ov-file#mac) section in the documentation again, as it requires additional configuration.**
+- Added word-by-word karaoke highlighting feature by @robmckinnon in PR #65
+- Fixed sleep inhibition error occurring on some Linux systems that do not use systemd, mentioned in #67 by @hendrack
+- Improvements in code and documentation.
+
+# 1.1.6
+- Improved EPUB chapter detection: Now reliably detects chapters from NAV HTML (TOC) files, even in non-standard EPUBs, fixes the issue mentioned by @jefro108 in #33
+- Fixed SRT subtitle numbering issue, mentioned by @page-muncher in #41
+- Fixed missing chapter contents issue in some EPUB files.
+- Windows installer script now prompts the user to install the CUDA version of PyTorch even if no NVIDIA GPU is detected.
+- Abogen now includes Mandarin Chinese (misaki[zh]) by default; manual installation is no longer required.
+
 # 1.1.5
 - Changed the temporary directory path to user's cache directory, which is more appropriate for storing cache files and avoids issues with unintended cleanup.
 - Fixed the isssue where extra metadata information was not being saved to M4B files when they have no chapters, ensuring that all metadata is correctly written to the output file.
@@ -23,7 +36,7 @@
 - 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.
+- cPotential 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.
 

README.md

@@ -12,7 +12,8 @@ Abogen is a powerful text-to-speech conversion tool that makes it easy to turn e
 <img title="Abogen Main" src='https://raw.githubusercontent.com/denizsafak/abogen/refs/heads/main/demo/abogen.png' width="380"> <img title="Abogen Processing" src='https://raw.githubusercontent.com/denizsafak/abogen/refs/heads/main/demo/abogen2.png' width="380">
 
 ## Demo
-https://github.com/user-attachments/assets/cb66512d-0a52-48c3-bda4-f1e6a03fb8d6
+
+https://github.com/user-attachments/assets/094ba3df-7d66-494a-bc31-0e4b41d0b865
 
 > This demo was generated in just 5 seconds, producing ∼1 minute of audio with perfectly synced subtitles. To create a similar video, see [the demo guide](https://github.com/denizsafak/abogen/tree/main/demo).
 
@@ -60,6 +61,10 @@ source venv/bin/activate
 
 # Install abogen
 pip3 install abogen
+
+# For Silicon Mac (M1, M2 etc.)
+# After installing abogen, we need to install Kokoro's development version which includes MPS support.
+pip3 install git+https://github.com/hexgrad/kokoro.git
 ```
 ### Linux
 ```bash
@@ -129,7 +134,7 @@ Here’s Abogen in action: in this demo, it processes ∼3,000 characters of tex
 | **Select Voice** | First letter of the language code (e.g., `a` for American English, `b` for British English, etc.), second letter is for `m` for male and `f` for female. |
 | **Voice mixer** | Create custom voices by mixing different voice models with a profile system. See [Voice Mixer](#voice-mixer) for more details. |
 | **Voice preview** | Listen to the selected voice before processing. |
-| **Generate subtitles** | `Disabled`, `Sentence`, `Sentence + Comma`, `1 word`, `2 words`, `3 words`, etc. (Represents the number of words in each subtitle entry) |
+| **Generate subtitles** | `Disabled`, `Sentence`, `Sentence + Comma`, `Sentence + Highlighting`, `1 word`, `2 words`, `3 words`, etc. (Represents the number of words in each subtitle entry) |
 | **Output voice format** | `.WAV`, `.FLAC`, `.MP3`, `.OPUS (best compression)` and `M4B (with chapters)` (Special thanks to [@jborza](https://github.com/jborza) for chapter support in PR [#10](https://github.com/denizsafak/abogen/pull/10)) |
 | **Output subtitle format** | Configures the subtitle format as `SRT (standard)`, `ASS (wide)`, `ASS (narrow)`, `ASS (centered wide)`, or `ASS (centered narrow)`. |
 | **Replace single newlines with spaces** | Replaces single newlines with spaces in the text. This is useful for texts that have imaginary line breaks. |
@@ -156,6 +161,8 @@ Here’s Abogen in action: in this demo, it processes ∼3,000 characters of tex
 | **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. |
 
+> Special thanks to [@robmckinnon](https://github.com/robmckinnon) for adding Sentence + Highlighting feature in PR [#65](https://github.com/denizsafak/abogen/pull/65)
+
 ## `Voice Mixer`
 <img title="Abogen Voice Mixer" src='https://raw.githubusercontent.com/denizsafak/abogen/refs/heads/main/demo/voice_mixer.png'>
 
@@ -221,6 +228,9 @@ Similar to chapter markers, it is possible to add metadata tags for `M4B` files.
 ```
 For a complete list of supported languages and voices, refer to Kokoro's [VOICES.md](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/VOICES.md). To listen to sample audio outputs, see [SAMPLES.md](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/SAMPLES.md).
 
+> [!NOTE]
+> Japanese audio may require additional configuration. Please check [#56](https://github.com/denizsafak/abogen/issues/56) for more information.
+
 ## `MPV Config`
 I highly recommend using [MPV](https://mpv.io/installation/) to play your audio files, as it supports displaying subtitles even without a video track. Here's my `mpv.conf`:
 ```

WINDOWS_INSTALL.bat

@@ -1,5 +1,5 @@
 @echo off
-setlocal
+setlocal EnableDelayedExpansion
 cd /d "%~dp0"
 
 :: Set misaki language
@@ -23,7 +23,6 @@ set CUDA_VERSION=128
 for /f "delims=: tokens=*" %%A in ('findstr /b ::: "%~f0"') do @echo(%%A
 
 set CURRENT_DIR="%CD%"
-setlocal enabledelayedexpansion
 set NAME=abogen
 set PROJECTFOLDER=abogen
 set RUN=python_embedded\Scripts\abogen.exe
@@ -264,7 +263,7 @@ if exist %PYPROJECT_FILE% (
 :: Install misaki again if MISAKI_LANG is not set to "en"
 if "%MISAKI_LANG%" NEQ "en" (
     echo Configuring language pack: %MISAKI_LANG%
-    %PYTHON_CONSOLE_PATH% -m pip install misaki[lang] --upgrade --no-warn-script-location
+    %PYTHON_CONSOLE_PATH% -m pip install misaki[%MISAKI_LANG%] --upgrade --no-warn-script-location
     if errorlevel 1 (
         echo Failed to install misaki language pack.
         pause
@@ -276,6 +275,7 @@ if "%MISAKI_LANG%" NEQ "en" (
 for /f %%i in ('%PYTHON_CONSOLE_PATH% -c "from abogen.is_nvidia import check; print(check())"') do set IS_NVIDIA=%%i
 
 :: Check if torch is installed with CUDA support
+echo.
 echo Checking CUDA availability...
 if /I "%IS_NVIDIA%"=="true" (
     for /f %%i in ('%PYTHON_CONSOLE_PATH% -c "from torch.cuda import is_available; print(is_available())"') do set cuda_available=%%i
@@ -293,7 +293,26 @@ if /I "%IS_NVIDIA%"=="true" (
         echo CUDA is available on NVIDIA GPU.
     )
 ) else (
-    echo GPU is not NVIDIA. Skipping PyTorch CUDA installation.
+    echo.
+    echo Unable to detect an NVIDIA GPU in your system.
+    echo.
+    echo Do you want to install PyTorch anyway?
+    echo.
+    echo If you DO have an NVIDIA GPU, please press Y.
+    echo If you DO NOT have an NVIDIA GPU, please press N.
+    echo.
+    choice /C YN /M "Y=Yes, N=No"
+    if errorlevel 2 (
+        echo Skipping PyTorch installation.
+    ) else (
+        echo Installing PyTorch with CUDA %CUDA_VERSION% support...
+        %PYTHON_CONSOLE_PATH% -m pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu%CUDA_VERSION% --no-warn-script-location
+        if errorlevel 1 (
+            echo Failed to install PyTorch.
+            pause
+            exit /b
+        )
+    )
 )
 
 :: Ask user if they want to create a desktop shortcut

abogen/VERSION

@@ -1 +1 @@
-1.1.5
+1.1.7

abogen/book_handler.py

@@ -324,14 +324,31 @@ class HandlerDialog(QDialog):
                     f"Found NCX item via ITEM_NAVIGATION: {ncx_in_nav.get_name()}"
                 )
 
-        # 3. If still no nav_item, check for ITEM_NCX directly
+        # 3. If still no nav_item, check for NCX or fallback to NAV HTML in all ITEM_DOCUMENTs
-        if not nav_item:
+        ncx_constant = getattr(epub, "ITEM_NCX", None)
-            ncx_items = list(self.book.get_items_of_type(ebooklib.ITEM_NCX))
+        if not nav_item and ncx_constant is not None:
+            ncx_items = list(self.book.get_items_of_type(ncx_constant))
             if ncx_items:
-                nav_item = ncx_items[0]  # Take the first one
+                nav_item = ncx_items[0]
                 nav_type = "ncx"
-                logging.info(f"Found NCX item via ITEM_NCX: {ncx_items[0].get_name()}")
+                logging.info(f"Found NCX item via ITEM_NCX: {nav_item.get_name()}")
-
+        # Fallback: search all ITEM_DOCUMENTs for a NAV HTML with <nav epub:type="toc">
+        if not nav_item:
+            for item in self.book.get_items_of_type(ebooklib.ITEM_DOCUMENT):
+                try:
+                    html_content = item.get_content().decode("utf-8", errors="ignore")
+                    if "<nav" in html_content and 'epub:type="toc"' in html_content:
+                        soup = BeautifulSoup(html_content, "html.parser")
+                        nav_tag = soup.find("nav", attrs={"epub:type": "toc"})
+                        if nav_tag:
+                            nav_item = item
+                            nav_type = "html"
+                            logging.info(
+                                f"Found NAV HTML with TOC in: {item.get_name()}"
+                            )
+                            break
+                except Exception as e:
+                    continue
         # 4. If no navigation item found by any method, trigger fallback
         if not nav_item or not nav_type:
             logging.warning(
@@ -516,10 +533,14 @@ class HandlerDialog(QDialog):
                         slice_html += self.doc_content.get(intermediate_doc_href, "")
                 except Exception:
                     pass
-
+            # Fallback: if slice_html is empty, try to get the whole file's text
+            if not slice_html.strip() and current_doc_html:
+                logging.warning(
+                    f"No content found for src '{current_src}', using full file as fallback."
+                )
+                slice_html = current_doc_html
             if slice_html.strip():
                 slice_soup = BeautifulSoup(slice_html, "html.parser")
-                # Add double newlines after <p> and <div> tags
                 for tag in slice_soup.find_all(["p", "div"]):
                     tag.append("\n\n")
                 for tag in slice_soup.find_all(["sup", "sub"]):
@@ -581,6 +602,23 @@ class HandlerDialog(QDialog):
             f"Finished processing EPUB navigation. Found {len(self.content_texts)} content sections linked to TOC."
         )
 
+    def _find_doc_key(self, base_href, doc_order, doc_order_decoded):
+        """Find the best matching doc_key for a given base_href using robust matching."""
+        candidates = [
+            base_href,
+            urllib.parse.unquote(base_href),
+        ]
+        base_name = os.path.basename(base_href).lower()
+        for k in list(doc_order.keys()) + list(doc_order_decoded.keys()):
+            if os.path.basename(k).lower() == base_name:
+                candidates.append(k)
+        for candidate in candidates:
+            if candidate in doc_order:
+                return candidate, doc_order[candidate]
+            elif candidate in doc_order_decoded:
+                return candidate, doc_order_decoded[candidate]
+        return None, None
+
     def _parse_ncx_navpoint(
         self,
         nav_point,
@@ -603,27 +641,15 @@ class HandlerDialog(QDialog):
 
         if src:
             base_href, fragment = src.split("#", 1) if "#" in src else (src, None)
-            # Try both original and decoded hrefs
+            doc_key, doc_idx = self._find_doc_key(
-            doc_key = None
+                base_href, doc_order, doc_order_decoded
-            if base_href in doc_order:
+            )
-                doc_key = base_href
+            if not doc_key:
-                doc_idx = doc_order[base_href]
-            elif urllib.parse.unquote(base_href) in doc_order:
-                doc_key = urllib.parse.unquote(base_href)
-                doc_idx = doc_order[doc_key]
-            elif base_href in doc_order_decoded:
-                doc_key = base_href
-                doc_idx = doc_order_decoded[base_href]
-            elif urllib.parse.unquote(base_href) in doc_order_decoded:
-                doc_key = urllib.parse.unquote(base_href)
-                doc_idx = doc_order_decoded[doc_key]
-            else:
                 logging.warning(
-                    f"Navigation entry '{title}' points to '{base_href}', which is not in the spine or document list."
+                    f"Navigation entry '{title}' points to '{base_href}', which is not in the spine or document list (even after basename fallback)."
                 )
                 current_entry_node["has_content"] = False
-                doc_key = None
+            else:
-            if doc_key is not None:
                 position = find_position_func(doc_key, fragment)
                 entry_data = {
                     "src": src,
@@ -698,28 +724,15 @@ class HandlerDialog(QDialog):
         current_entry_node["title"] = title
         current_entry_node["src"] = src
 
+        doc_key = None
+        doc_idx = None
+        position = 0
+        fragment = None
         if src:
             base_href, fragment = src.split("#", 1) if "#" in src else (src, None)
-            # Try both original and decoded hrefs
+            doc_key, doc_idx = self._find_doc_key(
-            doc_key = None
+                base_href, doc_order, doc_order_decoded
-            if base_href in doc_order:
+            )
-                doc_key = base_href
-                doc_idx = doc_order[base_href]
-            elif urllib.parse.unquote(base_href) in doc_order:
-                doc_key = urllib.parse.unquote(base_href)
-                doc_idx = doc_order[doc_key]
-            elif base_href in doc_order_decoded:
-                doc_key = base_href
-                doc_idx = doc_order_decoded[base_href]
-            elif urllib.parse.unquote(base_href) in doc_order_decoded:
-                doc_key = urllib.parse.unquote(base_href)
-                doc_idx = doc_order_decoded[doc_key]
-            else:
-                logging.warning(
-                    f"Navigation entry '{title}' points to '{base_href}', which is not in the spine or document list."
-                )
-                current_entry_node["has_content"] = False
-                doc_key = None
             if doc_key is not None:
                 position = find_position_func(doc_key, fragment)
                 entry_data = {
@@ -731,13 +744,16 @@ class HandlerDialog(QDialog):
                 }
                 ordered_entries.append(entry_data)
                 current_entry_node["has_content"] = True
+            else:
+                logging.warning(
+                    f"Navigation entry '{title}' points to '{base_href}', which is not in the spine or document list (even after basename fallback)."
+                )
+                current_entry_node["has_content"] = False
         else:
             current_entry_node["has_content"] = False
 
-        child_ol = li_element.find("ol", recursive=False)
+        for child_ol in li_element.find_all("ol", recursive=False):
-        if child_ol:
             for child_li in child_ol.find_all("li", recursive=False):
-                # Pass find_position_func down recursively
                 self._parse_html_nav_li(
                     child_li,
                     ordered_entries,
@@ -746,12 +762,7 @@ class HandlerDialog(QDialog):
                     current_entry_node["children"],
                     find_position_func,
                 )
-
+        tree_structure_list.append(current_entry_node)
-        if title and (
-            current_entry_node.get("has_content", False)
-            or current_entry_node["children"]
-        ):
-            tree_structure_list.append(current_entry_node)
 
     def _find_position_robust(self, doc_href, fragment_id):
         if doc_href not in self.doc_content:

abogen/conversion.py

@@ -23,6 +23,7 @@ import abogen.hf_tracker as hf_tracker
 import static_ffmpeg
 import threading  # for efficient waiting
 import subprocess
+import platform
 
 
 def get_sample_voice_text(lang_code):
@@ -310,8 +311,15 @@ class ConversionThread(QThread):
 
             self.log_updated.emit("\nInitializing TTS pipeline...")
 
-            # Set device based on use_gpu setting
+            # Set device based on use_gpu setting and platform
-            device = "cuda" if self.use_gpu else "cpu"
+            if self.use_gpu:
+                if platform.system() == "Darwin" and platform.processor() == "arm":
+                    device = "mps"  # Use MPS for Apple Silicon
+                else:
+                    device = "cuda"  # Use CUDA for other platforms
+            else:
+                device = "cpu"
+
             tts = self.KPipeline(
                 lang_code=self.lang_code, repo_id="hexgrad/Kokoro-82M", device=device
             )
@@ -434,8 +442,8 @@ class ConversionThread(QThread):
                 )
                 # Only check for files with allowed extensions (extension without dot, case-insensitive)
                 clash = any(
-                    os.path.splitext(fname)[0] == f"{base_name}{suffix}" and
+                    os.path.splitext(fname)[0] == f"{base_name}{suffix}"
-                    os.path.splitext(fname)[1][1:].lower() in allowed_exts
+                    and os.path.splitext(fname)[1][1:].lower() in allowed_exts
                     for fname in os.listdir(parent_dir)
                 )
                 if not os.path.exists(chapters_out_dir_candidate) and not clash:
@@ -467,6 +475,8 @@ class ConversionThread(QThread):
                     {"chapter": chapter[0], "start": 0.0, "end": 0.0}
                     for chapter in chapters
                 ]
+                # SRT numbering fix: use a global counter
+                merged_srt_index = 1  # SRT numbering for merged file
                 # Prepare output file/ffmpeg process for merged output
                 if self.output_format in ["wav", "mp3", "flac"]:
                     merged_out_file = sf.SoundFile(
@@ -483,8 +493,8 @@ class ConversionThread(QThread):
                     merged_out_file = None
                     ffmpeg_proc = None
                     metadata_options = (
-                            self._extract_and_add_metadata_tags_to_ffmpeg_cmd()
+                        self._extract_and_add_metadata_tags_to_ffmpeg_cmd()
-                        )
+                    )
                     # Prepare ffmpeg command for m4b output
                     cmd = [
                         "ffmpeg",
@@ -556,6 +566,11 @@ class ConversionThread(QThread):
                         merged_subtitle_file.write("[Script Info]\n")
                         merged_subtitle_file.write("Title: Generated by Abogen\n")
                         merged_subtitle_file.write("ScriptType: v4.00+\n\n")
+                        # Add style definitions for karaoke highlighting
+                        if self.subtitle_mode == "Sentence + Highlighting":
+                            merged_subtitle_file.write("[V4+ Styles]\n")
+                            merged_subtitle_file.write("Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding\n")
+                            merged_subtitle_file.write("Style: Default,Arial,24,&H00FFFFFF,&H00808080,&H00000000,&H00404040,0,0,0,0,100,100,0,0,3,2,0,5,10,10,10,1\n\n")
                         merged_subtitle_file.write("[Events]\n")
                         merged_subtitle_file.write(
                             "Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text\n"
@@ -599,6 +614,7 @@ class ConversionThread(QThread):
                     {"chapter": chapter[0], "start": 0.0, "end": 0.0}
                     for chapter in chapters
                 ]
+                srt_index = 1  # SRT numbering fix for chapter-only mode
             # Instead of processing the whole text, process by chapter
             for chapter_idx, (chapter_name, chapter_text) in enumerate(chapters, 1):
                 chapter_out_path = None
@@ -677,6 +693,9 @@ class ConversionThread(QThread):
                         continue
                     # Open chapter subtitle file for incremental writing if needed
                     chapter_subtitle_file = None
+                    chapter_srt_index = (
+                        1  # Initialize SRT numbering for this chapter file
+                    )
                     if self.subtitle_mode != "Disabled":
                         subtitle_format = getattr(self, "subtitle_format", "srt")
                         file_extension = "ass" if "ass" in subtitle_format else "srt"
@@ -694,6 +713,13 @@ class ConversionThread(QThread):
                             chapter_subtitle_file.write("[Script Info]\n")
                             chapter_subtitle_file.write("Title: Generated by Abogen\n")
                             chapter_subtitle_file.write("ScriptType: v4.00+\n\n")
+
+                            # Add style definitions for karaoke highlighting
+                            if self.subtitle_mode == "Sentence + Highlighting":
+                                chapter_subtitle_file.write("[V4+ Styles]\n")
+                                chapter_subtitle_file.write("Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding\n")
+                                chapter_subtitle_file.write("Style: Default,Arial,24,&H00FFFFFF,&H00808080,&H00000000,&H00404040,0,0,0,0,100,100,0,0,3,2,0,5,10,10,10,1\n\n")
+
                             chapter_subtitle_file.write("[Events]\n")
                             chapter_subtitle_file.write(
                                 "Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text\n"
@@ -817,16 +843,18 @@ class ConversionThread(QThread):
                                     for start, end, text in new_entries:
                                         start_time = self._ass_time(start)
                                         end_time = self._ass_time(end)
+                                        # Use karaoke effect for highlighting mode
+                                        effect = "karaoke" if self.subtitle_mode == "Sentence + Highlighting" else ""
                                         merged_subtitle_file.write(
-                                            f"Dialogue: 0,{start_time},{end_time},Default,,{merged_subtitle_margin},{merged_subtitle_margin},0,,{merged_subtitle_alignment_tag}{text}\n"
+                                            f"Dialogue: 0,{start_time},{end_time},Default,,{merged_subtitle_margin},{merged_subtitle_margin},0,{effect},{merged_subtitle_alignment_tag}{text}\n"
                                         )
                                 else:
-                                    for i, (start, end, text) in enumerate(
+                                    for entry in new_entries:
-                                        new_entries, 1
+                                        start, end, text = entry
-                                    ):
                                         merged_subtitle_file.write(
-                                            f"{i}\n{self._srt_time(start)} --> {self._srt_time(end)}\n{text}\n\n"
+                                            f"{merged_srt_index}\n{self._srt_time(start)} --> {self._srt_time(end)}\n{text}\n\n"
                                         )
+                                        merged_srt_index += 1
                         # Per-chapter subtitle processing for both file and ffmpeg_proc
                         if chapter_out_file or chapter_ffmpeg_proc:
                             new_chapter_entries = []
@@ -844,16 +872,18 @@ class ConversionThread(QThread):
                                     for start, end, text in new_chapter_entries:
                                         start_time = self._ass_time(start)
                                         end_time = self._ass_time(end)
+                                        # Use karaoke effect for highlighting mode
+                                        effect = "karaoke" if self.subtitle_mode == "Sentence + Highlighting" else ""
                                         chapter_subtitle_file.write(
-                                            f"Dialogue: 0,{start_time},{end_time},Default,,{chapter_subtitle_margin},{chapter_subtitle_margin},0,,{chapter_subtitle_alignment_tag}{text}\n"
+                                            f"Dialogue: 0,{start_time},{end_time},Default,,{chapter_subtitle_margin},{chapter_subtitle_margin},0,{effect},{chapter_subtitle_alignment_tag}{text}\n"
                                         )
                                 else:
-                                    for i, (start, end, text) in enumerate(
+                                    for entry in new_chapter_entries:
-                                        new_chapter_entries, 1
+                                        start, end, text = entry
-                                    ):
                                         chapter_subtitle_file.write(
-                                            f"{i}\n{self._srt_time(start)} --> {self._srt_time(end)}\n{text}\n\n"
+                                            f"{chapter_srt_index}\n{self._srt_time(start)} --> {self._srt_time(end)}\n{text}\n\n"
                                         )
+                                        chapter_srt_index += 1
                     if merge_chapters_at_end:
                         current_time += chunk_dur
                         if chapter_out_file or chapter_ffmpeg_proc:
@@ -1146,7 +1176,61 @@ class ConversionThread(QThread):
         if not tokens_with_timestamps:
             return
 
-        if self.subtitle_mode == "Sentence" or self.subtitle_mode == "Sentence + Comma":
+        if self.subtitle_mode == "Sentence + Highlighting":
+            # Sentence-based processing with karaoke highlighting
+            separator = r"[.!?]"
+            current_sentence = []
+            word_count = 0
+
+            for token in tokens_with_timestamps:
+                current_sentence.append(token)
+                word_count += 1
+
+                # Split sentences based on separator or word count
+                if (
+                    re.search(separator, token["text"]) and token["whitespace"] == " "
+                ) or word_count >= max_subtitle_words:
+                    if current_sentence:
+                        # Create karaoke subtitle entry for this sentence
+                        start_time = current_sentence[0]["start"]
+                        end_time = current_sentence[-1]["end"]
+
+                        # Generate karaoke text with background highlighting
+                        karaoke_text = ""
+                        for t in current_sentence:
+                            # Calculate duration in centiseconds
+                            duration = t["end"] - t["start"] if t["end"] and t["start"] else 0.5
+                            duration_cs = int(duration * 100)
+                            # Add karaoke effect - relies on style's SecondaryColour for highlighting
+                            karaoke_text += f"{{\\kf{duration_cs}}}{t['text']}{t.get('whitespace', '') or ''}"
+
+                        subtitle_entries.append(
+                            (start_time, end_time, karaoke_text.strip())
+                        )
+                        current_sentence = []
+                        word_count = 0
+
+            # Add any remaining tokens as a sentence
+            if current_sentence:
+                start_time = current_sentence[0]["start"]
+                end_time = current_sentence[-1]["end"]
+
+                # Generate karaoke text for remaining tokens
+                karaoke_text = ""
+                for t in current_sentence:
+                    duration = t["end"] - t["start"] if t["end"] and t["start"] else 0.5
+                    duration_cs = int(duration * 100)
+                    karaoke_text += f"{{\\kf{duration_cs}}}{t['text']}{t.get('whitespace', '') or ''}"
+                subtitle_entries.append((start_time, end_time, karaoke_text.strip()))
+
+            # Fallback for last entry
+            if subtitle_entries and fallback_end_time is not None:
+                last_entry = subtitle_entries[-1]
+                start, end, text = last_entry
+                if end is None or end <= start or end <= 0:
+                    subtitle_entries[-1] = (start, fallback_end_time, text)
+
+        elif self.subtitle_mode == "Sentence" or self.subtitle_mode == "Sentence + Comma":
             # Define separator pattern based on mode
             separator = r"[.!?]" if self.subtitle_mode == "Sentence" else r"[.!?,]"
             current_sentence = []
@@ -1313,7 +1397,16 @@ class VoicePreviewThread(QThread):
 
         # Generate the preview and save to cache
         try:
-            device = "cuda" if self.use_gpu else "cpu"
+
+            # Set device based on use_gpu setting and platform
+            if self.use_gpu:
+                if platform.system() == "Darwin" and platform.processor() == "arm":
+                    device = "mps"  # Use MPS for Apple Silicon
+                else:
+                    device = "cuda"  # Use CUDA for other platforms
+            else:
+                device = "cpu"
+
             tts = self.kpipeline_class(
                 lang_code=self.lang_code, repo_id="hexgrad/Kokoro-82M", device=device
             )

abogen/gui.py

@@ -895,6 +895,7 @@ class abogen(QWidget):
             "Disabled: No subtitles will be generated.\n"
             "Sentence: Subtitles will be generated for each sentence.\n"
             "Sentence + Comma: Subtitles will be generated for each sentence and comma.\n"
+            "Sentence + Highlighting: Subtitles with word-by-word karaoke highlighting.\n"
             "1+ word: Subtitles will be generated for each word(s).\n\n"
             "Supported languages for subtitle generation:\n"
             + "\n".join(
@@ -902,7 +903,7 @@ class abogen(QWidget):
                 for lang in SUPPORTED_LANGUAGES_FOR_SUBTITLE_GENERATION
             )
         )
-        subtitle_options = ["Disabled", "Sentence", "Sentence + Comma"] + [
+        subtitle_options = ["Disabled", "Sentence", "Sentence + Comma", "Sentence + Highlighting"] + [
             f"{i} word" if i == 1 else f"{i} words" for i in range(1, 11)
         ]
         self.subtitle_combo.addItems(subtitle_options)
@@ -970,6 +971,25 @@ class abogen(QWidget):
             lambda i: self.set_subtitle_format(self.subtitle_format_combo.itemData(i))
         )
         subtitle_format_layout.addWidget(self.subtitle_format_combo)
+        # If subtitle mode requires highlighting, SRT is not supported. Disable SRT item
+        # and auto-switch to a compatible ASS format if SRT is currently selected.
+        try:
+            if hasattr(self, "subtitle_mode") and self.subtitle_mode == "Sentence + Highlighting":
+                idx_srt = self.subtitle_format_combo.findData("srt")
+                if idx_srt >= 0:
+                    item = self.subtitle_format_combo.model().item(idx_srt)
+                    if item is not None:
+                        item.setEnabled(False)
+                # If current selection is SRT, switch to centered narrow ASS
+                if self.subtitle_format_combo.currentData() == "srt":
+                    new_idx = self.subtitle_format_combo.findData("ass_centered_narrow")
+                    if new_idx >= 0:
+                        self.subtitle_format_combo.setCurrentIndex(new_idx)
+                        # Persist the change
+                        self.set_subtitle_format(self.subtitle_format_combo.itemData(new_idx))
+        except Exception:
+            # Fail-safe: don't crash UI if model manipulation isn't supported on some platforms
+            pass
         controls_layout.addLayout(subtitle_format_layout)
 
         # Replace single newlines dropdown (acts like checkbox)
@@ -2471,6 +2491,30 @@ class abogen(QWidget):
             self.subtitle_format_combo.setEnabled(False)
         else:
             self.subtitle_format_combo.setEnabled(True)
+        # If highlighting mode selected, SRT is not supported. Disable SRT option and
+        # switch away from it if currently selected.
+        try:
+            idx_srt = self.subtitle_format_combo.findData("srt")
+            if mode == "Sentence + Highlighting":
+                if idx_srt >= 0:
+                    item = self.subtitle_format_combo.model().item(idx_srt)
+                    if item is not None:
+                        item.setEnabled(False)
+                # If current format is SRT, switch to a compatible ASS format
+                if self.subtitle_format_combo.currentData() == "srt":
+                    new_idx = self.subtitle_format_combo.findData("ass_centered_narrow")
+                    if new_idx >= 0:
+                        self.subtitle_format_combo.setCurrentIndex(new_idx)
+                        self.set_subtitle_format(self.subtitle_format_combo.itemData(new_idx))
+            else:
+                # Re-enable SRT option when not in highlighting mode
+                if idx_srt >= 0:
+                    item = self.subtitle_format_combo.model().item(idx_srt)
+                    if item is not None:
+                        item.setEnabled(True)
+        except Exception:
+            # Ignore errors interacting with model (defensive)
+            pass
 
     def on_format_changed(self, fmt):
         self.selected_format = fmt

abogen/main.py

@@ -32,11 +32,13 @@ os.environ["MIOPEN_CONV_PRECISE_ROCM_TUNING"] = "0"
 # Reset sleep states
 atexit.register(prevent_sleep_end)
 
+
 # Also handle signals (Ctrl+C, kill, etc.)
 def _cleanup_sleep(signum, frame):
     prevent_sleep_end()
     sys.exit(0)
 
+
 signal.signal(signal.SIGINT, _cleanup_sleep)
 signal.signal(signal.SIGTERM, _cleanup_sleep)
 
@@ -50,6 +52,7 @@ 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:

abogen/utils.py

@@ -3,6 +3,7 @@ import sys
 import json
 import warnings
 import platform
+import shutil
 import subprocess
 import re
 from threading import Thread
@@ -74,24 +75,32 @@ def get_user_config_path():
         if os.path.exists(custom_dir):
             config_dir = custom_dir
         else:
-            config_dir = user_config_dir("abogen", appauthor=False, roaming=True, ensure_exists=True)
+            config_dir = user_config_dir(
+                "abogen", appauthor=False, roaming=True, ensure_exists=True
+            )
     else:
         # Windows and fallback case
-        config_dir = user_config_dir("abogen", appauthor=False, roaming=True, ensure_exists=True)
+        config_dir = user_config_dir(
+            "abogen", appauthor=False, roaming=True, ensure_exists=True
+        )
 
     return os.path.join(config_dir, "config.json")
 
+
 # Define cache path
 def get_user_cache_path(folder=None):
     from platformdirs import user_cache_dir
 
-    cache_dir = user_cache_dir("abogen", appauthor=False, opinion=True, ensure_exists=True)
+    cache_dir = user_cache_dir(
+        "abogen", appauthor=False, opinion=True, ensure_exists=True
+    )
     if folder:
         cache_dir = os.path.join(cache_dir, folder)
         # Ensure the directory exists
         os.makedirs(cache_dir, exist_ok=True)
     return cache_dir
 
+
 _sleep_procs = {"Darwin": None, "Linux": None}  # Store sleep prevention processes
 
 
@@ -231,13 +240,23 @@ def calculate_text_length(text):
 def get_gpu_acceleration(enabled):
     try:
         import torch
-        from torch.cuda import is_available
+        from torch.cuda import is_available as cuda_available
 
         if not enabled:
-            return "CUDA GPU available but using CPU.", False
+            return "GPU available but using CPU.", False
-        if is_available():
+        
+        # Check for Apple Silicon MPS
+        if platform.system() == "Darwin" and platform.processor() == "arm":
+            if torch.backends.mps.is_available():
+                return "MPS GPU available and enabled.", True
+            else:
+                return "MPS GPU not available on Apple Silicon. Using CPU.", False
+        
+        # Check for CUDA
+        if cuda_available():
             return "CUDA GPU available and enabled.", True
-        # Gather more diagnostic info if CUDA is not available
+        
+        # Gather CUDA diagnostic info if not available
         try:
             cuda_devices = torch.cuda.device_count()
             cuda_error = (
@@ -249,7 +268,7 @@ def get_gpu_acceleration(enabled):
             cuda_error = str(e)
         return f"CUDA GPU is not available. Using CPU. ({cuda_error})", False
     except Exception as e:
-        return f"Error checking CUDA GPU: {e}", False
+        return f"Error checking GPU: {e}", False
 
 
 def prevent_sleep_start():
@@ -268,17 +287,24 @@ def prevent_sleep_start():
         # Add program name and reason for inhibition
         program_name = PROGRAM_NAME
         reason = "Prevent sleep during abogen process"
-        _sleep_procs["Linux"] = create_process(
+        # Only attempt to use systemd-inhibit if it's available on the system.
-            [
+        if shutil.which("systemd-inhibit"):
-                "systemd-inhibit",
+            _sleep_procs["Linux"] = create_process(
-                f"--who={program_name}",
+                [
-                f"--why={reason}",
+                    "systemd-inhibit",
-                "--what=sleep",
+                    f"--who={program_name}",
-                "--mode=block",
+                    f"--why={reason}",
-                "sleep",
+                    "--what=sleep",
-                "infinity",
+                    "--mode=block",
-            ]
+                    "sleep",
-        )
+                    "infinity",
+                ]
+            )
+        else:
+            # Non-systemd distro or systemd tools not installed: skip inhibition rather than crash
+            print(
+                "systemd-inhibit not found: skipping sleep inhibition on this Linux system."
+            )
 
 
 def prevent_sleep_end():

demo/README.md

@@ -23,20 +23,22 @@ brew install ffmpeg
 sudo apt install ffmpeg
 ```
 
-## Create the Video (.webm)
+## Create the Video (.mp4) (Recommended)
 
-Run this FFmpeg command to create the tiny video:
+Run this FFmpeg command to create the video:
 
 ```
-ffmpeg -loop 1 -framerate 24 -i bg.jpg -i audio.wav -vf "ass=subtitle.ass" -c:v libvpx-vp9 -b:v 0 -crf 30 -c:a libopus -shortest demo.webm
+ffmpeg -loop 1 -framerate 24 -i bg.jpg -i demo.wav -vf "ass=demo.ass" -c:v libx264 -pix_fmt yuv420p -preset slow -crf 18 -c:a aac -b:a 192k -movflags +faststart -shortest demo.mp4
 ```
 
-## For Higher Quality (But Larger) Video (.mp4)
+## For Smaller Filesize (.webm)
 
-If you need better quality for distribution, use this command instead:
+If you need a smaller video file, use this command:
 
 ```
-ffmpeg -loop 1 -framerate 24 -i bg.jpg -i audio.wav -vf "ass=subtitle.ass" -c:v libx264 -preset slow -crf 18 -movflags +faststart -c:a copy -shortest demo.mp4
+ffmpeg -loop 1 -framerate 24 -i bg.jpg -i demo.wav -vf "ass=demo.ass" -c:v libvpx-vp9 -b:v 0 -crf 30 -c:a libopus -shortest demo.webm
 ```
 
-This creates an MP4 file that's compatible with more devices but larger in size.
+
+
+

demo/demo.ass

@@ -4,127 +4,127 @@ ScriptType: v4.00+
 
 [Events]
 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
-Dialogue: 0,0:00:00.27,0:00:00.83,Default,,100,100,0,,{\an5}Abogen
+Dialogue: 0,0:00:00.35,0:00:00.93,Default,,90,90,0,,{\an5}Abogen
-Dialogue: 0,0:00:00.83,0:00:00.96,Default,,100,100,0,,{\an5}is
+Dialogue: 0,0:00:00.93,0:00:01.07,Default,,90,90,0,,{\an5}is
-Dialogue: 0,0:00:00.96,0:00:01.11,Default,,100,100,0,,{\an5}your
+Dialogue: 0,0:00:01.07,0:00:01.27,Default,,90,90,0,,{\an5}your
-Dialogue: 0,0:00:01.11,0:00:01.66,Default,,100,100,0,,{\an5}all-in-one
+Dialogue: 0,0:00:01.27,0:00:01.86,Default,,90,90,0,,{\an5}all-in-one
-Dialogue: 0,0:00:01.66,0:00:02.04,Default,,100,100,0,,{\an5}tool
+Dialogue: 0,0:00:01.86,0:00:02.22,Default,,90,90,0,,{\an5}tool
-Dialogue: 0,0:00:02.04,0:00:02.17,Default,,100,100,0,,{\an5}for
+Dialogue: 0,0:00:02.22,0:00:02.35,Default,,90,90,0,,{\an5}for
-Dialogue: 0,0:00:02.17,0:00:02.53,Default,,100,100,0,,{\an5}turning
+Dialogue: 0,0:00:02.35,0:00:02.73,Default,,90,90,0,,{\an5}turning
-Dialogue: 0,0:00:02.53,0:00:03.22,Default,,100,100,0,,{\an5}text
+Dialogue: 0,0:00:02.73,0:00:03.13,Default,,90,90,0,,{\an5}text
-Dialogue: 0,0:00:03.22,0:00:03.45,Default,,100,100,0,,{\an5}into
+Dialogue: 0,0:00:03.13,0:00:03.35,Default,,90,90,0,,{\an5}into
-Dialogue: 0,0:00:03.45,0:00:03.75,Default,,100,100,0,,{\an5}something
+Dialogue: 0,0:00:03.35,0:00:03.70,Default,,90,90,0,,{\an5}something
-Dialogue: 0,0:00:03.75,0:00:03.85,Default,,100,100,0,,{\an5}you
+Dialogue: 0,0:00:03.70,0:00:03.79,Default,,90,90,0,,{\an5}you
-Dialogue: 0,0:00:03.85,0:00:04.04,Default,,100,100,0,,{\an5}can
+Dialogue: 0,0:00:03.79,0:00:03.98,Default,,90,90,0,,{\an5}can
-Dialogue: 0,0:00:04.04,0:00:04.32,Default,,100,100,0,,{\an5}see
+Dialogue: 0,0:00:03.98,0:00:04.25,Default,,90,90,0,,{\an5}see
-Dialogue: 0,0:00:04.32,0:00:04.45,Default,,100,100,0,,{\an5}and
+Dialogue: 0,0:00:04.25,0:00:04.37,Default,,90,90,0,,{\an5}and
-Dialogue: 0,0:00:04.45,0:00:05.61,Default,,100,100,0,,{\an5}hear.
+Dialogue: 0,0:00:04.37,0:00:05.37,Default,,90,90,0,,{\an5}hear.
-Dialogue: 0,0:00:05.61,0:00:06.02,Default,,100,100,0,,{\an5}Easily
+Dialogue: 0,0:00:05.37,0:00:05.79,Default,,90,90,0,,{\an5}Easily
-Dialogue: 0,0:00:06.02,0:00:06.48,Default,,100,100,0,,{\an5}convert
+Dialogue: 0,0:00:05.79,0:00:06.23,Default,,90,90,0,,{\an5}convert
-Dialogue: 0,0:00:06.48,0:00:07.31,Default,,100,100,0,,{\an5}ePub,
+Dialogue: 0,0:00:06.23,0:00:07.11,Default,,90,90,0,,{\an5}ePub,
-Dialogue: 0,0:00:07.31,0:00:08.40,Default,,100,100,0,,{\an5}PDF,
+Dialogue: 0,0:00:07.11,0:00:08.17,Default,,90,90,0,,{\an5}PDF,
-Dialogue: 0,0:00:08.40,0:00:08.55,Default,,100,100,0,,{\an5}or
+Dialogue: 0,0:00:08.17,0:00:08.33,Default,,90,90,0,,{\an5}or
-Dialogue: 0,0:00:08.55,0:00:08.90,Default,,100,100,0,,{\an5}text
+Dialogue: 0,0:00:08.33,0:00:08.67,Default,,90,90,0,,{\an5}text
-Dialogue: 0,0:00:08.90,0:00:09.21,Default,,100,100,0,,{\an5}file
+Dialogue: 0,0:00:08.67,0:00:08.98,Default,,90,90,0,,{\an5}file
-Dialogue: 0,0:00:09.21,0:00:09.46,Default,,100,100,0,,{\an5}into
+Dialogue: 0,0:00:08.98,0:00:09.23,Default,,90,90,0,,{\an5}into
-Dialogue: 0,0:00:09.46,0:00:10.34,Default,,100,100,0,,{\an5}audio
+Dialogue: 0,0:00:09.23,0:00:09.91,Default,,90,90,0,,{\an5}audio
-Dialogue: 0,0:00:10.34,0:00:10.56,Default,,100,100,0,,{\an5}with
+Dialogue: 0,0:00:09.91,0:00:10.09,Default,,90,90,0,,{\an5}with
-Dialogue: 0,0:00:10.56,0:00:11.17,Default,,100,100,0,,{\an5}subtitles
+Dialogue: 0,0:00:10.09,0:00:10.69,Default,,90,90,0,,{\an5}subtitles
-Dialogue: 0,0:00:11.17,0:00:11.32,Default,,100,100,0,,{\an5}that
+Dialogue: 0,0:00:10.69,0:00:10.87,Default,,90,90,0,,{\an5}that
-Dialogue: 0,0:00:11.32,0:00:11.51,Default,,100,100,0,,{\an5}light
+Dialogue: 0,0:00:10.87,0:00:11.06,Default,,90,90,0,,{\an5}light
-Dialogue: 0,0:00:11.51,0:00:11.71,Default,,100,100,0,,{\an5}up
+Dialogue: 0,0:00:11.06,0:00:11.26,Default,,90,90,0,,{\an5}up
-Dialogue: 0,0:00:11.71,0:00:12.05,Default,,100,100,0,,{\an5}every
+Dialogue: 0,0:00:11.26,0:00:11.59,Default,,90,90,0,,{\an5}every
-Dialogue: 0,0:00:12.05,0:00:12.40,Default,,100,100,0,,{\an5}word
+Dialogue: 0,0:00:11.59,0:00:11.90,Default,,90,90,0,,{\an5}word
-Dialogue: 0,0:00:12.40,0:00:12.51,Default,,100,100,0,,{\an5}as
+Dialogue: 0,0:00:11.90,0:00:12.02,Default,,90,90,0,,{\an5}as
-Dialogue: 0,0:00:12.51,0:00:12.65,Default,,100,100,0,,{\an5}it’s
+Dialogue: 0,0:00:12.02,0:00:12.17,Default,,90,90,0,,{\an5}it’s
-Dialogue: 0,0:00:12.65,0:00:14.21,Default,,100,100,0,,{\an5}spoken.
+Dialogue: 0,0:00:12.17,0:00:13.46,Default,,90,90,0,,{\an5}spoken.
-Dialogue: 0,0:00:14.21,0:00:14.32,Default,,100,100,0,,{\an5}The
+Dialogue: 0,0:00:13.46,0:00:13.59,Default,,90,90,0,,{\an5}The
-Dialogue: 0,0:00:14.32,0:00:14.62,Default,,100,100,0,,{\an5}voice
+Dialogue: 0,0:00:13.59,0:00:13.92,Default,,90,90,0,,{\an5}voice
-Dialogue: 0,0:00:14.62,0:00:14.77,Default,,100,100,0,,{\an5}you
+Dialogue: 0,0:00:13.92,0:00:14.07,Default,,90,90,0,,{\an5}you
-Dialogue: 0,0:00:14.77,0:00:15.65,Default,,100,100,0,,{\an5}hear,
+Dialogue: 0,0:00:14.07,0:00:14.86,Default,,90,90,0,,{\an5}hear,
-Dialogue: 0,0:00:15.65,0:00:15.76,Default,,100,100,0,,{\an5}and
+Dialogue: 0,0:00:14.86,0:00:14.98,Default,,90,90,0,,{\an5}and
-Dialogue: 0,0:00:15.76,0:00:15.86,Default,,100,100,0,,{\an5}the
+Dialogue: 0,0:00:14.98,0:00:15.08,Default,,90,90,0,,{\an5}the
-Dialogue: 0,0:00:15.86,0:00:16.44,Default,,100,100,0,,{\an5}subtitles
+Dialogue: 0,0:00:15.08,0:00:15.67,Default,,90,90,0,,{\an5}subtitles
-Dialogue: 0,0:00:16.44,0:00:16.58,Default,,100,100,0,,{\an5}you
+Dialogue: 0,0:00:15.67,0:00:15.81,Default,,90,90,0,,{\an5}you
-Dialogue: 0,0:00:16.58,0:00:17.03,Default,,100,100,0,,{\an5}see
+Dialogue: 0,0:00:15.81,0:00:16.13,Default,,90,90,0,,{\an5}see
-Dialogue: 0,0:00:17.03,0:00:17.14,Default,,100,100,0,,{\an5}are
+Dialogue: 0,0:00:16.13,0:00:16.27,Default,,90,90,0,,{\an5}are
-Dialogue: 0,0:00:17.14,0:00:17.44,Default,,100,100,0,,{\an5}both
+Dialogue: 0,0:00:16.27,0:00:16.57,Default,,90,90,0,,{\an5}both
-Dialogue: 0,0:00:17.44,0:00:17.91,Default,,100,100,0,,{\an5}created
+Dialogue: 0,0:00:16.57,0:00:17.03,Default,,90,90,0,,{\an5}created
-Dialogue: 0,0:00:17.91,0:00:18.96,Default,,100,100,0,,{\an5}automatically
+Dialogue: 0,0:00:17.03,0:00:17.96,Default,,90,90,0,,{\an5}automatically
-Dialogue: 0,0:00:18.96,0:00:19.14,Default,,100,100,0,,{\an5}by
+Dialogue: 0,0:00:17.96,0:00:18.14,Default,,90,90,0,,{\an5}by
-Dialogue: 0,0:00:19.14,0:00:19.64,Default,,100,100,0,,{\an5}Abogen
+Dialogue: 0,0:00:18.14,0:00:18.67,Default,,90,90,0,,{\an5}Abogen
-Dialogue: 0,0:00:19.64,0:00:20.00,Default,,100,100,0,,{\an5}using
+Dialogue: 0,0:00:18.67,0:00:19.02,Default,,90,90,0,,{\an5}using
-Dialogue: 0,0:00:20.00,0:00:21.14,Default,,100,100,0,,{\an5}Kokoro,
+Dialogue: 0,0:00:19.02,0:00:20.10,Default,,90,90,0,,{\an5}Kokoro,
-Dialogue: 0,0:00:21.14,0:00:21.28,Default,,100,100,0,,{\an5}an
+Dialogue: 0,0:00:20.10,0:00:20.27,Default,,90,90,0,,{\an5}an
-Dialogue: 0,0:00:21.28,0:00:21.88,Default,,100,100,0,,{\an5}open-weight
+Dialogue: 0,0:00:20.27,0:00:20.88,Default,,90,90,0,,{\an5}open-weight
-Dialogue: 0,0:00:21.88,0:00:22.52,Default,,100,100,0,,{\an5}TTS
+Dialogue: 0,0:00:20.88,0:00:21.55,Default,,90,90,0,,{\an5}TTS
-Dialogue: 0,0:00:22.52,0:00:23.18,Default,,100,100,0,,{\an5}model
+Dialogue: 0,0:00:21.55,0:00:21.94,Default,,90,90,0,,{\an5}model
-Dialogue: 0,0:00:23.18,0:00:23.38,Default,,100,100,0,,{\an5}with
+Dialogue: 0,0:00:21.94,0:00:22.14,Default,,90,90,0,,{\an5}with
-Dialogue: 0,0:00:23.38,0:00:24.01,Default,,100,100,0,,{\an5}82
+Dialogue: 0,0:00:22.14,0:00:22.78,Default,,90,90,0,,{\an5}82
-Dialogue: 0,0:00:24.01,0:00:24.37,Default,,100,100,0,,{\an5}million
+Dialogue: 0,0:00:22.78,0:00:23.14,Default,,90,90,0,,{\an5}million
-Dialogue: 0,0:00:24.37,0:00:25.98,Default,,100,100,0,,{\an5}parameters.
+Dialogue: 0,0:00:23.14,0:00:24.55,Default,,90,90,0,,{\an5}parameters.
-Dialogue: 0,0:00:25.98,0:00:26.47,Default,,100,100,0,,{\an5}Everything
+Dialogue: 0,0:00:24.55,0:00:25.05,Default,,90,90,0,,{\an5}Everything
-Dialogue: 0,0:00:26.47,0:00:26.77,Default,,100,100,0,,{\an5}works
+Dialogue: 0,0:00:25.05,0:00:25.35,Default,,90,90,0,,{\an5}works
-Dialogue: 0,0:00:26.77,0:00:27.01,Default,,100,100,0,,{\an5}right
+Dialogue: 0,0:00:25.35,0:00:25.58,Default,,90,90,0,,{\an5}right
-Dialogue: 0,0:00:27.01,0:00:27.11,Default,,100,100,0,,{\an5}on
+Dialogue: 0,0:00:25.58,0:00:25.68,Default,,90,90,0,,{\an5}on
-Dialogue: 0,0:00:27.11,0:00:27.22,Default,,100,100,0,,{\an5}your
+Dialogue: 0,0:00:25.68,0:00:25.80,Default,,90,90,0,,{\an5}your
-Dialogue: 0,0:00:27.22,0:00:28.23,Default,,100,100,0,,{\an5}computer,
+Dialogue: 0,0:00:25.80,0:00:26.69,Default,,90,90,0,,{\an5}computer,
-Dialogue: 0,0:00:28.23,0:00:28.37,Default,,100,100,0,,{\an5}so
+Dialogue: 0,0:00:26.69,0:00:26.87,Default,,90,90,0,,{\an5}so
-Dialogue: 0,0:00:28.37,0:00:28.53,Default,,100,100,0,,{\an5}your
+Dialogue: 0,0:00:26.87,0:00:27.06,Default,,90,90,0,,{\an5}your
-Dialogue: 0,0:00:28.53,0:00:28.94,Default,,100,100,0,,{\an5}files
+Dialogue: 0,0:00:27.06,0:00:27.47,Default,,90,90,0,,{\an5}files
-Dialogue: 0,0:00:28.94,0:00:29.23,Default,,100,100,0,,{\an5}stay
+Dialogue: 0,0:00:27.47,0:00:27.73,Default,,90,90,0,,{\an5}stay
-Dialogue: 0,0:00:29.23,0:00:29.88,Default,,100,100,0,,{\an5}private
+Dialogue: 0,0:00:27.73,0:00:28.28,Default,,90,90,0,,{\an5}private
-Dialogue: 0,0:00:29.88,0:00:30.05,Default,,100,100,0,,{\an5}and
+Dialogue: 0,0:00:28.28,0:00:28.44,Default,,90,90,0,,{\an5}and
-Dialogue: 0,0:00:30.05,0:00:31.16,Default,,100,100,0,,{\an5}secure.
+Dialogue: 0,0:00:28.44,0:00:29.51,Default,,90,90,0,,{\an5}secure.
-Dialogue: 0,0:00:31.50,0:00:31.63,Default,,100,100,0,,{\an5}The
+Dialogue: 0,0:00:29.90,0:00:30.03,Default,,90,90,0,,{\an5}The
-Dialogue: 0,0:00:31.63,0:00:31.97,Default,,100,100,0,,{\an5}easy
+Dialogue: 0,0:00:30.03,0:00:30.37,Default,,90,90,0,,{\an5}easy
-Dialogue: 0,0:00:31.97,0:00:32.54,Default,,100,100,0,,{\an5}interface
+Dialogue: 0,0:00:30.37,0:00:30.95,Default,,90,90,0,,{\an5}interface
-Dialogue: 0,0:00:32.54,0:00:32.81,Default,,100,100,0,,{\an5}lets
+Dialogue: 0,0:00:30.95,0:00:31.23,Default,,90,90,0,,{\an5}lets
-Dialogue: 0,0:00:32.81,0:00:33.23,Default,,100,100,0,,{\an5}anyone
+Dialogue: 0,0:00:31.23,0:00:31.71,Default,,90,90,0,,{\an5}anyone
-Dialogue: 0,0:00:33.23,0:00:33.66,Default,,100,100,0,,{\an5}create
+Dialogue: 0,0:00:31.71,0:00:32.13,Default,,90,90,0,,{\an5}create
-Dialogue: 0,0:00:33.66,0:00:34.88,Default,,100,100,0,,{\an5}audiobooks,
+Dialogue: 0,0:00:32.13,0:00:33.33,Default,,90,90,0,,{\an5}audiobooks,
-Dialogue: 0,0:00:34.88,0:00:35.54,Default,,100,100,0,,{\an5}voiceovers
+Dialogue: 0,0:00:33.33,0:00:34.07,Default,,90,90,0,,{\an5}voiceovers
-Dialogue: 0,0:00:35.54,0:00:35.70,Default,,100,100,0,,{\an5}for
+Dialogue: 0,0:00:34.07,0:00:34.23,Default,,90,90,0,,{\an5}for
-Dialogue: 0,0:00:35.70,0:00:36.68,Default,,100,100,0,,{\an5}Instagram,
+Dialogue: 0,0:00:34.23,0:00:35.30,Default,,90,90,0,,{\an5}Instagram,
-Dialogue: 0,0:00:36.68,0:00:37.23,Default,,100,100,0,,{\an5}YouTube
+Dialogue: 0,0:00:35.30,0:00:35.93,Default,,90,90,0,,{\an5}YouTube
-Dialogue: 0,0:00:37.23,0:00:37.41,Default,,100,100,0,,{\an5}or
+Dialogue: 0,0:00:35.93,0:00:36.11,Default,,90,90,0,,{\an5}or
-Dialogue: 0,0:00:37.41,0:00:38.52,Default,,100,100,0,,{\an5}TikTok,
+Dialogue: 0,0:00:36.11,0:00:36.92,Default,,90,90,0,,{\an5}TikTok,
-Dialogue: 0,0:00:38.52,0:00:38.64,Default,,100,100,0,,{\an5}or
+Dialogue: 0,0:00:36.92,0:00:37.07,Default,,90,90,0,,{\an5}or
-Dialogue: 0,0:00:38.64,0:00:38.89,Default,,100,100,0,,{\an5}just
+Dialogue: 0,0:00:37.07,0:00:37.30,Default,,90,90,0,,{\an5}just
-Dialogue: 0,0:00:38.89,0:00:39.57,Default,,100,100,0,,{\an5}high-quality
+Dialogue: 0,0:00:37.30,0:00:38.00,Default,,90,90,0,,{\an5}high-quality
-Dialogue: 0,0:00:39.57,0:00:40.39,Default,,100,100,0,,{\an5}text-to-speech
+Dialogue: 0,0:00:38.00,0:00:38.82,Default,,90,90,0,,{\an5}text-to-speech
-Dialogue: 0,0:00:40.39,0:00:40.56,Default,,100,100,0,,{\an5}in
+Dialogue: 0,0:00:38.82,0:00:38.97,Default,,90,90,0,,{\an5}in
-Dialogue: 0,0:00:40.56,0:00:42.03,Default,,100,100,0,,{\an5}seconds.
+Dialogue: 0,0:00:38.97,0:00:40.16,Default,,90,90,0,,{\an5}seconds.
-Dialogue: 0,0:00:42.03,0:00:42.32,Default,,100,100,0,,{\an5}Whether
+Dialogue: 0,0:00:40.16,0:00:40.45,Default,,90,90,0,,{\an5}Whether
-Dialogue: 0,0:00:42.32,0:00:42.47,Default,,100,100,0,,{\an5}you
+Dialogue: 0,0:00:40.45,0:00:40.57,Default,,90,90,0,,{\an5}you
-Dialogue: 0,0:00:42.47,0:00:42.67,Default,,100,100,0,,{\an5}want
+Dialogue: 0,0:00:40.57,0:00:40.77,Default,,90,90,0,,{\an5}want
-Dialogue: 0,0:00:42.67,0:00:42.77,Default,,100,100,0,,{\an5}to
+Dialogue: 0,0:00:40.77,0:00:40.87,Default,,90,90,0,,{\an5}to
-Dialogue: 0,0:00:42.77,0:00:43.43,Default,,100,100,0,,{\an5}listen,
+Dialogue: 0,0:00:40.87,0:00:41.58,Default,,90,90,0,,{\an5}listen,
-Dialogue: 0,0:00:43.43,0:00:43.77,Default,,100,100,0,,{\an5}create
+Dialogue: 0,0:00:41.58,0:00:41.92,Default,,90,90,0,,{\an5}create
-Dialogue: 0,0:00:43.77,0:00:44.70,Default,,100,100,0,,{\an5}content,
+Dialogue: 0,0:00:41.92,0:00:42.65,Default,,90,90,0,,{\an5}content,
-Dialogue: 0,0:00:44.70,0:00:44.85,Default,,100,100,0,,{\an5}or
+Dialogue: 0,0:00:42.65,0:00:42.79,Default,,90,90,0,,{\an5}or
-Dialogue: 0,0:00:44.85,0:00:45.07,Default,,100,100,0,,{\an5}bring
+Dialogue: 0,0:00:42.79,0:00:43.01,Default,,90,90,0,,{\an5}bring
-Dialogue: 0,0:00:45.07,0:00:45.22,Default,,100,100,0,,{\an5}your
+Dialogue: 0,0:00:43.01,0:00:43.12,Default,,90,90,0,,{\an5}your
-Dialogue: 0,0:00:45.22,0:00:45.54,Default,,100,100,0,,{\an5}words
+Dialogue: 0,0:00:43.12,0:00:43.42,Default,,90,90,0,,{\an5}words
-Dialogue: 0,0:00:45.54,0:00:45.67,Default,,100,100,0,,{\an5}to
+Dialogue: 0,0:00:43.42,0:00:43.54,Default,,90,90,0,,{\an5}to
-Dialogue: 0,0:00:45.67,0:00:46.58,Default,,100,100,0,,{\an5}life,
+Dialogue: 0,0:00:43.54,0:00:44.33,Default,,90,90,0,,{\an5}life,
-Dialogue: 0,0:00:46.58,0:00:47.10,Default,,100,100,0,,{\an5}Abogen
+Dialogue: 0,0:00:44.33,0:00:44.90,Default,,90,90,0,,{\an5}Abogen
-Dialogue: 0,0:00:47.10,0:00:47.37,Default,,100,100,0,,{\an5}makes
+Dialogue: 0,0:00:44.90,0:00:45.17,Default,,90,90,0,,{\an5}makes
-Dialogue: 0,0:00:47.37,0:00:47.53,Default,,100,100,0,,{\an5}it
+Dialogue: 0,0:00:45.17,0:00:45.35,Default,,90,90,0,,{\an5}it
-Dialogue: 0,0:00:47.53,0:00:48.31,Default,,100,100,0,,{\an5}fast,
+Dialogue: 0,0:00:45.35,0:00:46.12,Default,,90,90,0,,{\an5}fast,
-Dialogue: 0,0:00:48.31,0:00:48.89,Default,,100,100,0,,{\an5}fun,
+Dialogue: 0,0:00:46.12,0:00:46.65,Default,,90,90,0,,{\an5}fun,
-Dialogue: 0,0:00:48.89,0:00:49.02,Default,,100,100,0,,{\an5}and
+Dialogue: 0,0:00:46.65,0:00:46.77,Default,,90,90,0,,{\an5}and
-Dialogue: 0,0:00:49.02,0:00:50.62,Default,,100,100,0,,{\an5}effortless.
+Dialogue: 0,0:00:46.77,0:00:48.07,Default,,90,90,0,,{\an5}effortless.
-Dialogue: 0,0:00:50.62,0:00:51.35,Default,,100,100,0,,{\an5}Experience
+Dialogue: 0,0:00:48.07,0:00:48.76,Default,,90,90,0,,{\an5}Experience
-Dialogue: 0,0:00:51.35,0:00:51.47,Default,,100,100,0,,{\an5}your
+Dialogue: 0,0:00:48.76,0:00:48.87,Default,,90,90,0,,{\an5}your
-Dialogue: 0,0:00:51.47,0:00:52.23,Default,,100,100,0,,{\an5}stories,
+Dialogue: 0,0:00:48.87,0:00:49.61,Default,,90,90,0,,{\an5}stories,
-Dialogue: 0,0:00:52.23,0:00:52.97,Default,,100,100,0,,{\an5}scripts,
+Dialogue: 0,0:00:49.61,0:00:50.15,Default,,90,90,0,,{\an5}scripts,
-Dialogue: 0,0:00:52.97,0:00:53.11,Default,,100,100,0,,{\an5}and
+Dialogue: 0,0:00:50.15,0:00:50.29,Default,,90,90,0,,{\an5}and
-Dialogue: 0,0:00:53.11,0:00:53.64,Default,,100,100,0,,{\an5}ideas
+Dialogue: 0,0:00:50.29,0:00:50.87,Default,,90,90,0,,{\an5}ideas
-Dialogue: 0,0:00:53.64,0:00:53.73,Default,,100,100,0,,{\an5}in
+Dialogue: 0,0:00:50.87,0:00:50.97,Default,,90,90,0,,{\an5}in
-Dialogue: 0,0:00:53.73,0:00:53.83,Default,,100,100,0,,{\an5}a
+Dialogue: 0,0:00:50.97,0:00:51.10,Default,,90,90,0,,{\an5}a
-Dialogue: 0,0:00:53.83,0:00:54.14,Default,,100,100,0,,{\an5}whole
+Dialogue: 0,0:00:51.10,0:00:51.46,Default,,90,90,0,,{\an5}whole
-Dialogue: 0,0:00:54.14,0:00:54.35,Default,,100,100,0,,{\an5}new
+Dialogue: 0,0:00:51.46,0:00:51.65,Default,,90,90,0,,{\an5}new
-Dialogue: 0,0:00:54.35,0:00:54.64,Default,,100,100,0,,{\an5}way
+Dialogue: 0,0:00:51.65,0:00:51.95,Default,,90,90,0,,{\an5}way
-Dialogue: 0,0:00:54.64,0:00:54.79,Default,,100,100,0,,{\an5}with
+Dialogue: 0,0:00:51.95,0:00:52.10,Default,,90,90,0,,{\an5}with
-Dialogue: 0,0:00:54.79,0:00:55.85,Default,,100,100,0,,{\an5}Abogen.
+Dialogue: 0,0:00:52.10,0:00:53.20,Default,,90,90,0,,{\an5}Abogen.

pyproject.toml

@@ -15,6 +15,7 @@ keywords = ["audiobook", "epub", "pdf", "text-to-speech", "subtitle", "tts", "ko
 dependencies = [
     "PyQt5>=5.15.11",
     "kokoro>=0.9.4",
+    "misaki[zh]>=0.9.4",
     "ebooklib>=0.19",
     "beautifulsoup4>=4.13.4",
     "PyMuPDF>=1.25.5",