Skip to content

README.md

Latest commit

 

History

History
106 lines (70 loc) · 2.75 KB

README.md

File metadata and controls

106 lines (70 loc) · 2.75 KB
Raw

🎥 TubeGPT - AI Video Assistant

Transform any YouTube video into an interactive knowledge base with the power of Google Gemini AI.

image

✨ Features

  • 🎥 YouTube Integration - Load any video with captions instantly
  • 🧠 AI-Powered Q&A - Chat with videos using Google Gemini 2.0 Flash
  • ⚡ Quick Actions - Pre-built queries for summaries and insights
  • 📊 Smart Analytics - Detailed video content metrics
  • 🔍 Intelligent Search - Find relevant content through semantic search

🚀 Quick Start

  1. Install Dependencies
pip install -r requirements.txt
  1. Get API Key
image
  1. Run Application
streamlit run app.py
  1. Start Chatting

  • Enter your API key

  • Load any YouTube video image

  • Ask questions about the content! image image

📋 Requirements

  • Python 3.8+
  • Streamlit
  • Google Gemini API key (free)
  • Internet connection

🎯 How It Works

  1. Load Video → Extracts transcript from YouTube
  2. Process Content → Breaks transcript into smart chunks
  3. Ask Questions → Uses AI to find relevant information
  4. Get Answers → Receives contextual responses from video content

📁 Project Structure

tubegpt/
├── README.md
├── app.py
├── requirements.txt
├── components/
│ ├── init.py
│ ├── ui_components.py
│ ├── auth_component.py
│ ├── video_component.py
│ ├── chat_component.py
│ └── analytics_component.py
├── core/
│ ├── init.py
│ └── gemini_service.py
├── utils/
│ ├── init.py
│ └── constants.py
└── .streamlit/
└── config.toml

🔗 Links

👨‍💻 Developer

Built with ❤️ by Adhiraj Singh

Powered by Google Gemini 2.0 Flash & Streamlit

⭐ Star this repo if you find TubeGPT helpful!