🎥 YouTube RAG Question Answering App

A Streamlit web application that enables users to ask questions about YouTube videos using Retrieval-Augmented Generation (RAG). The app fetches video transcripts, creates embeddings, and uses AI to answer questions based on the video content.

Main Purpose: This project was created to understand and utilize different LangChain components in a practical, real-world application. It demonstrates the integration of various LangChain modules including text splitters, vector stores, embeddings, retrievers, prompts, and chains to build a complete RAG system.

✨ Features

• 🎯 RAG-powered Q&A: Ask natural language questions about any YouTube video
• 🔍 Semantic Search: Advanced retrieval using vector embeddings
• 🧠 Context-aware Answers: AI generates answers based on actual video content
• 🌐 Multi-language Support: Supports videos with English and Hindi transcripts
• 🎨 Clean UI: Beautiful Streamlit interface with intuitive design
• ⚡ Fast Processing: Efficient vector storage and retrieval
• 🔗 LangChain Integration: Demonstrates practical usage of multiple LangChain components
• 📚 Educational: Perfect for learning how to build RAG systems with LangChain

🚀 Demo

1. Enter a YouTube URL
2. Click "Process Video" to fetch transcript and create embeddings
3. Ask any question about the video content
4. Get AI-powered answers based on the actual transcript

🛠️ Installation

Prerequisites

• Python 3.8 or higher
• Perplexity API key (get it from Perplexity AI)

Setup

1. Clone the repository

   git clone https://github.com/yourusername/youtube-rag-app.git
   cd youtube-rag-app

2. Create a virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Set up environment variables

   cp .env.template .env
   # Edit .env and add your PPLX_API_KEY

5. Run the application

   streamlit run app.py

📝 Environment Variables

Create a .env file with the following variables:

# Required
PPLX_API_KEY=your_perplexity_api_key_here

# Optional
HUGGINGFACE_API_TOKEN=your_huggingface_token_here

🏗️ Project Structure

youtube-rag-app/
├── app.py                      # Main Streamlit application
├── src/
│   ├── __init__.py
│   └── utils/
│       ├── __init__.py
│       ├── transcript_fetcher.py   # YouTube transcript handling
│       ├── vector_store.py         # Text chunking and embeddings
│       └── rag_chain.py           # RAG chain implementation
├── requirements.txt            # Python dependencies
├── .env.template              # Environment variables template
├── .gitignore                 # Git ignore rules
└── README.md                  # This file

🔧 How It Works

This project demonstrates the practical implementation of various LangChain components:

1. Transcript Fetching: Uses youtube-transcript-api to extract video transcripts
2. Text Processing: Implements LangChain's RecursiveCharacterTextSplitter for optimal text chunking
3. Embeddings: Utilizes LangChain's HuggingFaceEmbeddings with sentence-transformers/all-mpnet-base-v2 model
4. Vector Storage: Leverages LangChain's FAISS integration for efficient similarity search
5. Retrieval: Creates a retriever using LangChain's vector store interface
6. Prompt Engineering: Uses LangChain's PromptTemplate for structured prompt creation
7. Chain Construction: Implements LangChain's RunnableParallel and RunnableLambda for complex workflows
8. Question Answering: Integrates everything into a complete RAG pipeline using LangChain's chain paradigm

🎯 Usage Examples

Example Questions:

• "What is the main topic of this video?"
• "Can you summarize the key points discussed?"
• "Who are the people mentioned in the video?"
• "What are the important concepts explained?"

Supported Video Types:

• Educational content
• Tutorials and how-to videos
• Lectures and presentations
• Interviews and discussions
• Any video with available transcripts

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

Development Setup

1. Fork the repository
2. Create a feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📄 License

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

🙏 Acknowledgments

• LangChain for the RAG framework
• Streamlit for the web app framework
• Perplexity AI for the language model
• HuggingFace for embeddings models
• youtube-transcript-api for transcript extraction

⚠️ Limitations

• Requires videos to have available transcripts
• Transcript availability depends on YouTube's auto-generation or manual upload
• API rate limits may apply based on your Perplexity AI plan
• Processing time depends on video length and transcript size

🔮 Future Enhancements

• Support for multiple video URLs
• Transcript translation capabilities
• Chat history and conversation context
• Video timestamp references in answers
• Export Q&A sessions
• Support for additional language models

📧 Contact

For questions or suggestions, please open an issue on GitHub or contact harshvirani.91@gmail.com.

---

Made with ❤️ and AI