add_dub — Subtitle Vocalization & Automated Voice-Over for Windows

add_dub is a powerful tool that automatically transforms video subtitles into spoken voice (TTS) and creates a new video with audio voice-over. Its primary goal is to make content accessible to those who have difficulty reading subtitles—whether due to visual fatigue, visual impairment, or simply preference—by allowing them to listen instead.

With the new Translation Feature, you can now vocalize videos in your native language, even if the original subtitles are in a foreign language!

🚀 Features

• Subtitle Vocalization: Converts SRT subtitles into synchronized audio voice-over.
• AI Translation: Automatically translates subtitles before vocalization (supports 100+ languages via EasyNMT).
• Multiple TTS Engines:
  • OneCore (Windows): Fast, offline, uses installed system voices.
  • Edge TTS: High-quality, natural-sounding voices (requires Internet).
  • gTTS: Simple, lightweight (requires Internet).
• Smart Mixing: Automatically "ducks" (lowers) the original audio volume when the TTS speaks.
• Batch Processing: Process entire folders of videos recursively.
• Portable: No complex installation required.

📥 Input & Output

• Input: A video file (MKV, MP4, AVI, etc.) with embedded subtitles or an external .srt file.
• Output: An MKV video file containing:
  • Track 0: Original Video
  • Track 1: Mixed Audio (TTS + Original Background)
  • Track 2: Original Audio (Isolated)
  • Track 3: Subtitles

---

📦 Installation

Option 1: Portable Version (Recommended)

Download the latest release, unzip it, and you're ready to go! 📥 Download add_dub_win64.zip

Option 2: From Source

If you prefer to run from source:

git clone https://github.com/Jobijoba2000/add_dub.git
cd add_dub
start_add_dub.bat

---

🎮 Usage

First Launch

Simply run:

start_add_dub.bat

On the first run, the script will:

1. Download/Install the Toolbox (Portable Python, FFmpeg, MKVToolNix, etc.).
2. Create necessary directories: input, output, tmp, srt.
3. Generate a default options.conf.

Interactive Mode

Just run start_add_dub.bat without arguments. The tool will guide you through:

1. Selecting Videos: Choose from the input/ folder.
2. Audio Track: Select the source audio track.
3. Subtitles: Choose embedded or external subtitles.
4. TTS Engine: Select OneCore, Edge TTS, or gTTS.
5. Translation: Optionally translate subtitles to your preferred language (e.g., fr, en, es, ar, zh...).
6. Configuration: Adjust volume, speed, and offsets.

Batch Mode (CLI)

For automated processing, use the --batch flag.

Examples:

• Process a single file:

  start_add_dub.bat --batch -i "C:\Videos\movie.mkv" --tts-engine edge --voice "en-US-AriaNeural"

• Process a folder recursively:

  start_add_dub.bat --batch -i "C:\Videos\Series" --recursive

• Translate and Dub (e.g., English to French):

  start_add_dub.bat --batch -i "C:\Videos\movie.mkv" --translate --translate-to fr --voice "fr-FR-DeniseNeural"

• Custom Mix Settings:

  start_add_dub.bat --batch -i "C:\Videos\movie.mkv" --bg-mix 0.8 --tts-mix 1.2 --ducking-db -5.0

---

⚙️ Configuration (options.conf)

The options.conf file allows you to set default values. Format: key = value or key = value d (add d to ask the user at runtime).

Key Options:

• tts_engine: onecore, edge, gtts.
• voice_id: Specific voice ID (use --list-voices to find them).
• translate: true or false (enable translation by default).
• translate_to: Target language code (e.g., fr, en).
• min_rate_tts / max_rate_tts: Speed limits for TTS.
• ducking: Volume reduction of background audio in dB (e.g., -5.0).

---

🌍 Supported Languages

UI Languages: The interface automatically detects your system language. Supported languages include:

• English (en)
• French (fr)
• Spanish (es)
• Italian (it)
• German (de)
• Portuguese (pt)
• Russian (ru)
• Greek (el)
• Chinese (zh)
• Arabic (ar)
• Korean (ko)
• Japanese (ja)

Translation Support: You can translate subtitles from and to almost any language supported by EasyNMT/HuggingFace models.

---

🛠️ Troubleshooting

• "ffmpeg not found": Ensure the initial setup completed successfully. The tools folder should contain ffmpeg.
• TTS fails: Check your internet connection if using Edge TTS or gTTS. For OneCore, ensure the voice is installed in Windows settings.
• Translation errors: The first translation might be slow as it downloads models. Ensure you have a stable internet connection.

---

📄 License

This project is licensed under the MIT License. Dependencies included in the Toolbox are subject to their respective licenses.