🎬 YouTube AI Transcript Assistant 🔍🤖 A Chrome Extension + AI Backend that allows users to ask questions about any YouTube video. It fetches the video's transcript, chunks it, embeds it with SentenceTransformers, stores it in FAISS, and uses Gemini AI to answer questions contextually.

🚀 Ask anything about the video — no need to scrub through the timeline!

📹 Demo

screen-recording-2025-06-15-150244_QvE2gRoQ.mp4

🧠 Tech Stack

🧩 Frontend: Chrome Extension (HTML, JS, CSS) ⚙️ Backend: FastAPI 📼 Transcript Fetching: yt-dlp, .vtt conversion 🧠 Embeddings: sentence-transformers 🔎 Vector Store: FAISS 🧠 LLM: langchain-google-genai using Gemini

🛠️ Features

🌐 Chrome Extension to interact with any YouTube video 📄 Automatic transcript download + parsing 🔍 Vector-based semantic search using FAISS 🧠 Context-aware AI answers via Gemini API 🪄 Seamless integration between frontend & backend ⚡ Real-time processing with loading states 🎨 Modern glassmorphism UI design 📱 Responsive extension popup interface

🚀 Running the Project 🔧 1. Backend Setup bashcd backend python -m venv venv source venv/bin/activate # or venv\Scripts\activate on Windows pip install -r requirements.txt uvicorn main:app --reload Make sure .env contains: envGOOGLE_API_KEY=your_google_gemini_api_key 🌐 2. Chrome Extension Setup

Open chrome://extensions/ Enable Developer mode Click Load unpacked Select the extension/ folder Click the extension icon on YouTube, ask a question!

🔑 API Setup Get Google Gemini API Key

Go to Google AI Studio Create a new API key Add to .env file in backend folder

🎯 Usage

Navigate to any YouTube video Click the extension icon in your browser Type your question about the video content Get AI-powered answers based on the transcript!

Example Questions: "What are the main points discussed?" "Summarize the video in 3 bullet points" "What does the speaker say about [specific topic]?" "At what timestamp is [specific topic] mentioned?"