RAGTube 🎬🤖

AI-powered YouTube Transcript Q&A System using Retrieval-Augmented Generation (RAG).

🚀 Overview

RAGTube extracts YouTube video transcripts, stores them in a vector database (pgvector), and enables AI-powered Q&A using DeepSeek R1 (Ollama) and PostgreSQL.

🔹 Extracts transcripts from YouTube videos
🔹 Generates embeddings for semantic search
🔹 Uses RAG (Retrieval-Augmented Generation) to answer questions
🔹 Stores data in PostgreSQL (pgvector) for fast AI retrieval

🏗️ Features

✅ YouTube Transcript Extraction
✅ Vector Search for Similar Content
✅ AI-Powered Answers via DeepSeek R1
✅ PostgreSQL pgvector for Semantic Search
✅ FastAPI Backend for API Integration (optional)

---

📌 How It Works

1️⃣ Extract transcript from a YouTube video
2️⃣ Generate AI embeddings using DeepSeek R1
3️⃣ Store transcript & embeddings in a PostgreSQL database (pgvector)
4️⃣ Retrieve relevant content based on user queries
5️⃣ Use RAG to generate AI-powered answers