An AI-powered platform that transforms lecture videos into interactive study tools — including auto-generated notes, intelligent Q&A with timestamps, and customizable quizzes. It internally uses RAG(Retrieval-Augmented Generation) with a FAISS vector database for fast, accurate responses. Built with Flask, Flutter, langChain, Gemini API, and Whisper and deployed on AWS with a live demo.
- 🎥 Video Input: Upload your own video or provide a YouTube link.
- 📄 Smart Note Generation: Summarized, structured notes in a PDF-style format using advanced LLM summarization.
- 🤖 AI-Powered Q&A with RAG: Ask questions and get accurate answers based on FAISS vector similarity search and Gemini 2.0 LLM with timestamped references.
- 🧩 Quiz Generator: Test your knowledge with multiple-choice questions with adjustable difficulty.
- 🔗 Timestamps: Jump directly to the moment in the video where the answer was discussed.
- ☁️ Deployed on AWS: Live site running on EC2 with Docker.
- Backend: Python, Flask
- Frontend: Flutter
- AI & NLP: Gemini API (for Q&A), Whisper (for transcription), LangChain (for RAG), FAISS (vector store)
- Deployment: AWS EC2, Docker, S3 (for config)
This project loads environment variables from a .env file stored in AWS S3.
For local development, you have two options:
- Manual: Place your .env file in the backend root folder (same level as run.py).
- Automatic (if using AWS credentials):The backend will attempt to download
.envfrom S3 — for this, you must:- Set up AWS CLI credentials (
~/.aws/credentials) - Ensure your IAM user has
s3:GetObjectpermission for the S3 bucket used in the project.
- Set up AWS CLI credentials (
Ensure you have the following installed on your system:
- Docker
- Python 3
- Flutter
- Gemini API key in .env file
- VIDCAP_API_KEY in .env file
docker-compose up --buildThat’s it — the backend and frontend will start up, and you can access the app at: http://localhost:8080/
-
Requirements: The backend/requirements.txt file contains all necessary Python dependencies.
-
Set up backend:
cd backend python -m venv venv source venv/bin/activate pip install -r requirements.txt python app.py
-
Set up frontend:
cd frontend/flutter_app flutter pub get flutter run -d <browser>
- backend
app │ app.py │ ├───api │ │ cache.py │ │ user.py │ │ │ └───endpoints │ notes.py │ questions.py │ quiz.py │ upload.py │ └───services llm_rag.py relevant_time_stamps.py s3_service.py transcript_extraction.py youtube_transcript.py - frontend
│ main.dart │ ├───components │ button.dart │ common_background.dart │ square_tile.dart │ ├───pages │ chat_page.dart │ └───transcripts transcript_item.dart
Check out the live demo of LecLens at Live LecLens Demo.