Skip to content

Relase 1.0 Beta.

Latest
Latest

Choose a tag to compare

View all tags
@0GithubDoN 0GithubDoN released this 03 Mar 10:55
· 1 commit to main since this release
v1.0
ae19767
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

SubGen - AI Subtitle Generator

A powerful, easy-to-use AI-powered subtitle generator for personal use. This tool allows you to automatically generate subtitles for video and audio files using OpenAI's Whisper model, with a focus on user experience and design.

Features

  • ✅ Modern & Enhanced UI

    • Beautiful, modern interface with Material Design icons
    • NEW: Light and dark theme support with automatic system detection
    • NEW: Responsive design that adapts to any window size
    • Intuitive drag & drop functionality with visual feedback
    • NEW: Improved progress tracking with time estimation
    • NEW: Segment-based subtitle editor for precise editing

  • ✅ AI-Powered Transcription & Translation

    • Uses OpenAI Whisper for high-accuracy transcription
    • Multiple language support (including Romanian, English, Spanish, and more)
    • NEW: Enhanced translation with parallel processing and better server fallback
    • NEW: Improved error handling for translation services
    • Built-in translation using LibreTranslate
    • SRT and VTT export formats

  • ✅ Versatile File Support

    • Video: MKV, MP4, AVI, MOV, WEBM, FLV
    • Audio: MP3, WAV, AAC, FLAC, OGG, EAC3
    • Automatic format detection and processing

  • ✅ Advanced Features

    • Automatic audio extraction from video
    • One-click subtitle generation with cancellation support
    • NEW: Intelligent dependency management - installs packages only when needed
    • NEW: Update option to refresh dependencies when needed
    • Options for soft subtitles (external files) or hardcoded subtitles
    • Enhanced error handling and user feedback
    • NEW: Automatic resource management

Installation

Prerequisites

  • Python 3.8+ (Python 3.10 or 3.11 recommended)
  • FFmpeg (for audio extraction and subtitle embedding)

Windows Installation

  1. Make sure Python and FFmpeg are installed and added to your PATH
  2. Run the install.bat script
  3. Wait for the installation to complete
  4. Navigate to the created SubGen folder
  5. Run SubGen.bat to start the application

Mac/Linux Installation

  1. Make sure Python and FFmpeg are installed
  2. Make the install script executable: chmod +x install.sh
  3. Run the script: ./install.sh
  4. Wait for the installation to complete
  5. Launch the application using ./SubGen.sh or the desktop shortcut

Updating Dependencies

To check for and install updates to dependencies:

  • Windows: Run SubGen.bat --update
  • Mac/Linux: Run ./SubGen.sh --update

Troubleshooting Whisper Installation

If you encounter an error installing Whisper during setup, manually install it with:

cd SubGen
venv\Scripts\activate    # On Windows
source venv/bin/activate # On Mac/Linux

# Try this first:
pip install git+https://github.com/openai/whisper.git

# If that doesn't work, try:
pip install --upgrade setuptools wheel
pip install openai-whisper

How to Use

  1. Open the application by running SubGen.bat (Windows) or SubGen.sh (Mac/Linux)
  2. Select a file by clicking "Browse Files" or dragging and dropping a video/audio file
  3. Choose language settings:
    • Select source language for transcription
    • Optionally enable translation and select target language
  4. Click "Generate Subtitles" and wait for the transcription to complete
    • You can monitor progress and estimated completion time
    • Cancel the process at any time if needed
  5. Preview and edit the subtitles in the segment-based editor:
    • Each subtitle segment can be individually edited
    • Duration and timing information is displayed for each segment
  6. Save or embed the subtitles:
    • Save as .SRT or .VTT file
    • Embed directly into video (soft or hard subtitles)
  7. Toggle Dark Mode using the theme button in the top-right corner

Advanced Options

  • Translation Service: Uses LibreTranslate for free, open-source translation with robust server fallback
  • Subtitle Embedding: Choose between soft subtitles (can be turned on/off) or hardcoded subtitles (permanently visible)
  • GPU Acceleration: The application automatically detects and uses GPU acceleration for faster transcription if available
  • Dependency Management: Smart dependency checking that only installs what's needed and offers updates when available
  • Theme Settings: Toggle between light and dark modes to match your preference

System Requirements

  • OS: Windows 10/11, macOS 10.13+, or Linux
  • RAM: 4GB minimum, 8GB+ recommended
  • Storage: 5GB free space for the application and models
  • GPU: Optional, but recommended for faster processing (NVIDIA GPU with CUDA support)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Credits

  • OpenAI Whisper: For the AI transcription model
  • LibreTranslate: For open-source translation capabilities
  • FFmpeg: For media processing capabilities
  • PyQt5: For the user interface components
  • Material Design Icons: For the modern UI elements

Support

For issues and feature requests, please create an issue on the GitHub repository or contact the developer directly.

Recent Updates

  • NEW (v1.1.0): Added light/dark theme toggle
  • NEW (v1.1.0): Enhanced subtitle editor with segmented interface
  • NEW (v1.1.0): Improved translation with parallel processing
  • NEW (v1.1.0): Added Romanian language support
  • NEW (v1.1.0): Intelligent dependency management
  • NEW (v1.1.0): Responsive design for all window sizes
  • NEW (v1.1.0): Progress time estimation and cancellation
  • NEW (v1.1.0): Reorganized code structure for better maintainability
  • Added modern UI with Material Design icons
  • Integrated LibreTranslate for translation support
  • Improved file handling with direct paths
  • Enhanced drag & drop interface with visual feedback
Assets 3
Loading