🎙 MeetingMind Intelligence

MeetingMind is an enterprise-grade AI platform for recording, transcribing, and analyzing meetings. It natively captures browser audio (tab audio) and provides deep meeting intelligence including executive summaries, action item tracking, and grounded Q&A.

🚀 Key Features

• Native Tab Capture: Record high-quality meeting audio directly from your browser tab (Google Meet, Zoom, Teams) without bots.
• Whisper Transcription: High-accuracy multi-speaker transcription powered by faster-whisper.
• Speaker Diarization: Automatically identifies who said what using pyannote.audio.
• Intelligent Summaries: Generates industry-standard MOM reports (JSON-structured) including summaries, decisions, risks, and action items.
• Grounded Q&A (RAG): Chat with your meeting history. Answers are strictly grounded in meeting transcripts to prevent hallucinations.
• Enterprise Ready: Full FastAPI backend with SQLite/PostgreSQL support, Next.js frontend with Tailwind CSS, and secure JWT authentication.

🛠 Tech Stack

• Frontend: Next.js (TypeScript), Tailwind CSS, Lucide Icons, Axios.
• Backend: FastAPI (Python), SQLAlchemy, Alembic.
• AI/ML: faster-whisper, pyannote.audio, chromadb (Vector DB), sentence-transformers.
• LLM Support: Groq (Llama 3), Google Gemini, Ollama (Local).
• Deployment: Vercel (Frontend), AWS EC2 + CloudFront (Backend).

📂 Project Structure

meetapp/
├── backend/            # FastAPI Application
│   ├── app/
│   │   ├── api/       # REST Endpoints
│   │   ├── core/      # Pipeline & Configuration
│   │   ├── db/        # Database setup
│   │   ├── models/    # DB Schemas
│   │   ├── services/  # AI, Transcription, Export logic
│   ├── alembic/       # DB Migrations
│   └── deploy/        # Nginx & EC2 setup
├── frontend/           # Next.js Application
│   ├── src/
│   │   ├── app/       # Pages & UI
│   │   ├── hooks/     # Custom React Hooks (Recorder, etc.)
│   │   ├── lib/       # API Clients
│   │   └── store/     # State Management (Zustand)
└── README.md           # This guide

⚡ Quick Start

1. Backend Setup

cd backend
python -m venv venv
./venv/Scripts/activate
pip install -r requirements.txt
# Create .env file based on .env.example
uvicorn app.main:app --reload

2. Frontend Setup

cd frontend
npm install
# Create .env file based on .env.example
npm run dev

🛡 Security & Edge Cases

• FFmpeg Handling: Robust audio chunking logic handles extremely long meetings without memory issues or timeouts.
• Cumulative Offsets: Fixed timestamp logic ensures continuous transcription across chunks.
• Graceful Diarization: Automatic CPU/GPU detection and fallback if speaker identification fails.
• JWT Auth: Secure user-isolated meeting storage and search.

📄 License

© 2026 MeetingMind. All rights reserved.