YouTube Channel Transcriber

A Python script that downloads and transcribes videos from YouTube channels using OpenAI's Whisper. The script first attempts to use YouTube's built-in transcription API, and falls back to Whisper if no transcript is available.

Features

• Fetches all playlists from a YouTube channel matching specified keywords
• Downloads and processes videos in batches
• Uses YouTube's transcript API when available
• Falls back to OpenAI's Whisper for videos without transcripts
• Stores transcripts in SQLite database
• Shows progress with tqdm progress bars

Prerequisites

• Python 3.7+
• ffmpeg (required for Whisper)
• Chrome/Chromium browser (for Selenium)
• YouTube Data API key

Installation

1. Clone this repository
2. Install required packages:

pip install -r requirements.txt

3. Install ffmpeg (if not already installed):

   • Ubuntu: sudo apt install ffmpeg
   • macOS: brew install ffmpeg
   • Windows: Download from ffmpeg website

4. Copy config.example.json to config.json and update with your settings:

   • Get a YouTube API key from Google Cloud Console
   • Set your target channel URL
   • Define keywords to match playlists

Usage

1. Configure your settings in config.json
2. Run the script:

python main.py

Configuration

Edit config.json with your settings:

• youtube_api_key: Your YouTube Data API key
• channel_url: URL of the YouTube channel to process
• playlist_keywords: List of keywords to match playlists
• whisper_model: Whisper model to use (tiny, base, small, medium, large)

License

MIT License