Turn meeting recordings into actionable summaries in minutes, not hours.
Stop manually transcribing hours of meetings. Summeet uses AI to automatically transcribe audio files and generate structured summaries with speaker identification and key action items.
🎁 Free to start: Use AssemblyAI's free tier (up to 5 hours/month) and any OpenAI-compatible model including free options like Groq, DeepSeek, or local models.
📁 Upload audio → 🤖 AI transcribes → ✏️ Edit & label speakers → 📋 Get summary → 📄 Export markdown
- Drag & drop audio files (MP3, WAV, M4A, etc.)
- Automatic transcription with speaker detection
- Edit transcripts and add speaker names/roles
- AI-powered summaries with action items and key points
- Export clean markdown files for sharing
# Clone and start
git clone https://github.com/lyzgeorge/summeet.git
cd summeet
cp env.example .env
# Add your API keys to .env
# OPENAI_API_KEY=your_key_here
# ASSEMBLYAI_API_KEY=your_key_here
# Run with Docker
docker compose -f docker-compose.simple.yml up -d
# Open http://localhost:3000# Backend
cd backend
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
python main.py
# Frontend (new terminal)
cd frontend
npm install && npm run dev- AssemblyAI: 5 hours free transcription/month
- OpenAI alternatives: Use Groq, DeepSeek, or local models
- Hosting: Deploy on your own server or free tiers
- AssemblyAI: $0.37/hour for additional transcription
- OpenAI: ~$0.01 per summary with GPT-4o-mini
Create .env file:
# Required
OPENAI_API_KEY=your_openai_key
ASSEMBLYAI_API_KEY=your_assemblyai_key
# Optional: Use alternative providers
OPENAI_BASE_URL=https://api.groq.com/openai/v1 # Groq (free tier)
# OPENAI_BASE_URL=https://api.deepseek.com # DeepSeek (cheap)
# OPENAI_BASE_URL=http://localhost:1234/v1 # Local model
TEXT_MODEL_NAME=llama-3.1-8b-instant # For Groq
# TEXT_MODEL_NAME=deepseek-chat # For DeepSeek
# TEXT_MODEL_NAME=gpt-4o-mini # For OpenAI| Main Interface | Summary Export |
|---|---|
|
 |
Stack: FastAPI + Vue.js + SQLite + Docker
AI Services: AssemblyAI (transcription) + OpenAI-compatible APIs (summarization)
Deployment: Single container, runs anywhere
POST /upload- Upload and transcribe audioGET /transcription/{id}- Get transcriptionPOST /summarize/{id}- Generate summaryGET /export/{id}- Download markdown
MP3, WAV, M4A, FLAC, OGG (auto-converted with FFmpeg)
- Fork the repo
- Create feature branch:
git checkout -b feature/amazing-feature - Commit changes:
git commit -m 'Add amazing feature' - Push branch:
git push origin feature/amazing-feature - Open Pull Request
MIT License - feel free to use this commercially or personally.
- Issues: Report bugs
- Discussions: Ask questions
⭐ Star this repo if it saves you time transcribing meetings!