YouTube Video Transcription and Chatbot

This project provides a comprehensive tool for processing YouTube videos and engaging in AI-powered discussions about their content. It combines video downloading, audio transcription, and interactive AI chat capabilities using state-of-the-art open-source tools.

Features

Core Functionality

• Download audio from YouTube videos using yt-dlp
• Transcribe audio to text using OpenAI's whisper (or process a local .mp3 file directly)
• Engage in interactive chat sessions about video content using ollama AI models

Advanced Features

• Runtime model selection from available or running Ollama models
• Smart handling of model availability and startup
• Robust error handling and graceful session management
• Support for multiple exit commands and interrupt handling

Installation

1. Clone the repository:

   git clone https://github.com/fdepierre/YouTube2Post.git
   cd YouTube2Post
   
   

2. Install the required packages:

   Make sure you have Python installed. Then run:

   pip install -r requirements.txt

3. Additional Setup:

   • Ensure yt-dlp is installed and accessible in your PATH.
   • Configure your config.json file with necessary parameters like tmp_directory, work_directory, and model.

4. Ollama Setup:

   Ollama must be installed and running on your local machine. To start Ollama, use:

   ollama serve &
   ollama run <model_version>

   The model version is defined in the config.json file under the "model" key.

Usage

The tool provides several command-line options for different workflows:

Basic Operations

# Transcribe a YouTube video
python yt2post.py -t <YouTube_URL>

# Transcribe a local mp3 file (no user prompts, metadata auto-extracted)
python yt2post.py -t /path/to/audio.mp3

# Chat about an existing transcript
python yt2post.py -c <transcript_file>

# Full process: download, transcribe, and chat
python yt2post.py -f <YouTube_URL>

Model Selection

# Use model selection interface
python yt2post.py -m -c <transcript_file>

When using the -m option:

• If models are running, you can select one directly
• If no models are running, available models will be listed
• Press Enter to use a single running model, or type 'no' to see alternatives
• Type 'exit' during chat to end the session

Configuration

Basic Configuration

Create a config.json file in the root directory:

{
    "tmp_directory": "path/to/tmp",
    "work_directory": "path/to/work",
    "model": "llama3.2"  // Default model, can be overridden with -m option
}

Model Configuration

• The model in config.json specifies the default model
• Use the -m option to override this and select a model at runtime
• Available models can be listed using ollama list
• Running models can be viewed using ollama ps

Development

Testing and Quality Assurance

The project maintains high code quality through comprehensive testing:

# Run all tests
pytest

# Run tests with coverage report
pytest --cov=modules tests/

# Generate HTML coverage report
pytest --cov=modules --cov-report=html tests/

View the coverage report at htmlcov/index.html

Project Structure

YouTube2Post/
├── modules/                    # Core functionality modules
│   ├── config_manager.py       # Configuration handling
│   ├── directory_manager.py    # File and directory management
│   ├── ollama_manager.py       # AI model interaction
│   ├── transcriber.py          # Audio transcription
│   └── youtube_downloader.py   # YouTube video processing
├── tests/                      # Test suite
├── config.json                # Configuration file
├── requirements.txt           # Python dependencies
└── yt2post.py                # Main script

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests for new functionality
5. Submit a pull request

Ensure your code follows the project's style and passes all tests.

License

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

Acknowledgments

• yt-dlp for downloading YouTube content.
• Whisper for audio transcription.
• Ollama for AI chatbot interactions.