Open Meet Sum 👋

Open Meet Sum is an AI meeting assistant that automatically transcribes, summarizes, and analyses audio with human-in-the-loop collaboration AI-powered insights. It's a self-hosted solution, supporting both local, model API endpoints and built-in transcription engines.

✨ Key Features

• 🎙️ Real-time Transcription - Automatic speech-to-text with built-in Whisper (small, mid, large)
• 🎯 AI Summarization - Generate smart summaries with bullet points and key takeaways
• 👤 Speaker Recognition - Automatically identify and track speakers in meetings
• 📝 Live Meeting Support - Join meetings with real-time transcription and processing
• 🔐 Secure & Private - JWT authentication, PostgreSQL storage, fully self-hosted
• 👥 Multi-user & Teams - Team-based meeting organization and user management
• 💾 Persistent Storage - Database-backed persistent storage for all your meetings
• 📊 Admin Dashboard - Full administration interface for users and settings
• 🚀 REST API - Complete API for integration with other tools
• 🐳 Docker Ready - Single-image deployment (local test -> registry -> container app)

🚀 Install

Local Development

Terminal 1 - Backend:

# Create a virtual environment 
conda create -n meetSum python=3.11
conda activate meetSum
pip install -r requirements.txt

# Create a file for environment variables
cd backend
cp .env.example .env

#  Run
sh run.sh

Terminal 2 - Frontend (new terminal):

conda activate meetSum
cd frontend
npm install
npm run dev

Open http://localhost:5173

Note

Run in two separate terminals - one for backend, one for frontend. Keep both running.

Docker (Single Image)

git clone https://github.com/kraack-tech/open-meet-sum.git
cd open-meet-sum
docker build -t open-meet-sum:local .

Run container with your external PostgreSQL:

docker run --rm -p 8001:8001 \
	-e DATABASE_URL='postgresql://postgres:password@host.docker.internal:8000/meetsum' \
	-e JWT_SECRET_KEY='super-secret-key' \
	open-meet-sum:local

Optional: persist model cache to avoid large re-downloads:

docker volume create meetsum-model-cache
docker run --rm -p 8001:8001 \
	-e DATABASE_URL='postgresql://postgres:password@host.docker.internal:8000/meetsum' \
	-e JWT_SECRET_KEY='super-secret-key' \
	-e XDG_CACHE_HOME='/home/appuser/.cache' \
	-e HF_HOME='/home/appuser/.cache/huggingface' \
	-v meetsum-model-cache:/home/appuser/.cache \
	open-meet-sum:local

Access at http://localhost:8001 API docs at http://localhost:8001/docs

Note

NOTE: This can take up to about 3 minutes on first run due to downloading the local models.

⚙️ Configuration

Environment Variables

Backend (.env):

DATABASE_URL=postgresql://postgres:password@localhost:5432/meetsum
JWT_SECRET_KEY=your-secret-key
CORS_ORIGINS=http://localhost:5173,http://localhost:8001

Frontend (.env):

VITE_API_URL=http://localhost:8001

Made with ❤️ for better meetings

⬆ Back to top