YT-Summarizer

Most of the code in this project was generated by Claude AI.

YT-Summarizer is a Python-based application built with Flask that automates the process of downloading YouTube videos, transcribing their content, and summarizing the transcripts using ChatGPT. This tool is designed for users who want quick insights from YouTube videos without watching them in full.

Installation

1. Install FFmpeg and add its executable path to the system environment variables:

   • Download FFmpeg from here.

2. Install CUDA (recommended: check the compatible CUDA version for PyTorch):

   • Download CUDA from NVIDIA's official site.

   • Verify compatibility with PyTorch here.

3. Clone the repository:

   git clone https://github.com/Bu4275/YT-Summarizer
   cd YT-Summarizer
   

4. Install Python dependencies:

   pip install -r requirements.txt
   

5. Configure the .env file and set OPENAI_API_KEY accordingly.

Usage

• Run the Flask application:

  python app.py
  

• Access the web interface at http://localhost:5000

• You can modify the contents of the prompts.yaml file to include your own prompts.

Technologies Used

• Python

• Flask

• YouTube API / yt-dlp (for downloading videos)

• Whisper AI (for transcription)

• ChatGPT API (for summarization)

Screenshot