Private, Multi-Source Audio Capture • Real-Time Speech-to-Text • Structured Minutes of Meeting (MOM) • Multi-Model AI Orchestration
📺 Video Demo • 🚀 Quickstart • 🐳 Docker • ✨ Key Capabilities • 🎛️ AI Control Panel • ❤️ Support • 🐛 Issues
Check out the full product walkthrough and see Minomeet AI in action:
👉 Watch on YouTube: https://youtu.be/2RuGooHVJUE
┌─────────────────┐ ┌──────────────────┐ ┌────────────────────────┐
│ 🎙️ Dual Audio │ ──► │ ⚡ Whisper Neural│ ──► │ 🤖 Multi-Model LLM │ ──► 📄 Executive MOM
│ (Mic + System) │ │ Transcription │ │ (Ollama/Claude/GPT/...)│ + Action Matrix
└─────────────────┘ └──────────────────┘ └────────────────────────┘
- 🛡️ 100% Privacy-First Architecture: Audio streams, live transcripts, and embeddings stay strictly local on your machine.
- ⚡ Multi-Source Audio Ingestion: Record microphone audio, browser tabs, Google Meet, Zoom, Slack Huddles, or import
.mp3,.wav,.m4a,.webmfiles. - 🧠 Dynamic Multi-Model AI: Seamlessly connect Ollama (Local LLMs), OpenAI (GPT-4o), Anthropic (Claude 3.7 Sonnet), Google Gemini 2.5, Groq, and Custom OpenAI-Compatible Servers (LM Studio, vLLM, DeepSeek, Together AI).
- 📋 Executive MOM Generation: Automatically generates structured Executive Summaries, Key Decisions, Discussion Highlights, Next Steps, and interactive Action Item matrices.
- 💬 Ask Your Meetings AI: Semantic natural-language Q&A assistant across your historical meeting archives.
- ✉️ One-Click Follow-Up Emails: Generates tailored email drafts in Professional, Concise, or Action-Oriented formats.
- 💻 Global Terminal CLI (
minomeet): Launch casually anytime directly from your terminal by typingminomeetand pressingoto open the app window.
The repository is completely open-source and ready to run locally in under 2 minutes: 👉 GitHub Repository: https://github.com/shareefmx/minomeet
| Requirement | Recommended Version | Purpose |
|---|---|---|
| Node.js | v18.0.0+ (v20+ LTS recommended) |
Web app and backend server runtime |
| Python | 3.10+ (Recommended) |
On-device Whisper STT & neural speech decoding |
| FFmpeg | 4.0+ (Recommended) |
High-fidelity multi-format audio processing |
| AI Key / Server | Gemini, OpenAI, Claude, Groq, or Ollama | MOM summaries, action matrices & Q&A |
# Step 1: Clone the repository
git clone https://github.com/shareefmx/minomeet.git
cd minomeet
# Step 2: Install Node.js workspace dependencies
npm install
# Step 3: Install Python requirements for on-device Whisper STT & neural transcription
pip install -r requirements.txt
# Step 4: Register global CLI command (Run once)
npm link💡 Optional System Dependency (FFmpeg):
- macOS:
brew install ffmpeg- Ubuntu / Debian:
sudo apt install ffmpeg- Windows:
winget install Gyan.FFmpegorchoco install ffmpeg
Once setup is complete, you can launch Minomeet casually at any time from any directory in your terminal:
minomeet __ __ _ _
| \/ (_)_ __ ___ _ __ ___ ___ ___| |_
| |\/| | | '_ \ / _ \| '_ ` _ \ / _ \/ _ \ __|
| | | | | | | | (_) | | | | | | __/ __/ |_
|_| |_|_|_| |_|\___/|_| |_| |_|\___|\___|\__|
Autonomous On-Device & Cloud AI Meeting Intelligence (v1.2.0)
-----------------------------------------------------------------
➜ Web Application: http://localhost:5173
➜ Backend API: http://localhost:5001
➜ Environment: 100% Privacy-First On-Device
-----------------------------------------------------------------
Interactive Shortcuts:
[o] Open Minomeet in browser window
[c] Clear terminal screen
[q] Quit Minomeet (or press Ctrl+C)
-----------------------------------------------------------------
- Press
o: Instantly launches your default browser and opens Minomeet (http://localhost:5173). - Press
c: Clears the console. - Press
q(orCtrl+C): Gracefully stops all Minomeet services.
(Alternatively, you can also start via npm start or npm run dev).
Run Minomeet anywhere with zero manual dependencies using Docker and Docker Compose:
# Clone the repository
git clone https://github.com/shareefmx/minomeet.git
cd minomeet
# Start Minomeet container in the background
docker compose up -dOpen your browser and visit:
👉 http://localhost:5001
# View live container logs
docker compose logs -f
# Stop the container
docker compose down# 1. Build the production Docker image
docker build -t minomeet .
# 2. Run container with persistent data volumes
docker run -d \
--name minomeet \
-p 5001:5001 \
-v minomeet_data:/app/server/data \
-v minomeet_uploads:/app/server/uploads \
minomeet🛡️ Persistent Meeting Data: Meeting notes, audio archives, and settings are preserved across container updates and restarts inside the
minomeet_dataandminomeet_uploadsDocker volumes.
Minomeet connects directly to frontier AI providers or local self-hosted inference servers:
Settings ──► AI Model ──► Provider ──► Model ──► Test Connection ──► Save ──► AI Agent
| Provider | Supported Models | Mode |
|---|---|---|
| Google Gemini | gemini-2.5-flash, gemini-2.5-pro, gemini-1.5-pro |
Cloud API |
| OpenAI | gpt-4o, gpt-4o-mini, o3-mini, gpt-4-turbo |
Cloud API |
| Anthropic Claude | claude-3-7-sonnet, claude-3-5-sonnet, claude-3-5-haiku |
Cloud API |
| Groq (LPUs) | llama-3.3-70b-versatile, llama-3.1-8b-instant, mixtral-8x7b |
Ultra-Fast Cloud |
| Ollama (Self-Hosted) | llama3.3:70b, qwen2.5:7b, deepseek-r1:8b, mistral:7b |
100% Offline Local |
| Custom OpenAI Server | LM Studio, vLLM, DeepSeek, Together AI, LiteLLM | Auto Model Discovery |
💡 Custom OpenAI Server Auto-Discovery: Enter your custom endpoint URL & API key, then click "Fetch Models" — Minomeet automatically retrieves all models running on your server and populates a dynamic selection dropdown.
minomeet/
├── bin/ # Global CLI Executable ('minomeet')
│ └── minomeet.js # Terminal interactive runner with 'o' browser launcher
├── client/ # React 18 + Vite 6 Web Application
│ ├── src/
│ │ ├── components/
│ │ │ ├── layout/ # Titlebar, Sidebar, ToastHost
│ │ │ ├── modals/ # AboutModal, ImportAudio, AskMeetings, FollowUpEmail
│ │ │ └── screens/ # HomeScreen, RecordingScreen, NotesScreen, SettingsScreen
│ │ ├── context/ # MeetingContext (State management & Audio streaming)
│ │ ├── services/ # API Client & Export Handlers
│ │ ├── types/ # TypeScript Type Definitions
│ │ └── utils/ # AI Model Config & Robust JSON Formatters
│ ├── index.html
│ ├── package.json
│ └── vite.config.ts
├── server/ # Node.js + Express + TypeScript Backend
│ ├── src/
│ │ ├── routes/ # Meetings, AI, Audio, Settings, Templates
│ │ ├── services/ # AIService, AudioService, StorageService
│ │ ├── types/ # Server Type Definitions
│ │ ├── utils/ # Model Resolution & Robust AI JSON Parser
│ │ └── index.ts # Server Entrypoint (Port 5001)
│ ├── package.json
│ └── tsconfig.json
├── package.json # Root Workspace Configuration & CLI 'bin'
└── README.md # Project Documentation
| Action / Shortcut | Function |
|---|---|
minomeet |
Global terminal command to start services |
Press 'o' |
Open web app in default browser |
Start Live Recording |
Capture dual-channel audio (Mic + Meeting participants) |
Import Audio |
Upload .mp3, .wav, .m4a, or .webm for local transcription |
Ask Your Meetings |
Launch semantic multi-meeting Q&A search assistant |
Export MOM |
Export notes to Markdown, Plain Text, or formatted PDF |
Email Recap |
Generate 1-click tailored email drafts with action items |
- Zero Third-Party Telemetry: Conversations, audio streams, and generated notes are processed and stored locally on your machine.
- Local Storage: All recordings and meeting history are archived in local JSON databases.
- Local LLM Compatibility: Full support for Ollama and local OpenAI-compatible endpoints so zero meeting data ever leaves your computer.
If you run into any bugs, have feature requests, or want to suggest improvements: 👉 Open an Issue on GitHub: https://github.com/shareefmx/minomeet/issues
If you find Minomeet useful for your workflow and meetings, here are a few great ways to support the project:
- 🚀 Give it a try: Run it for your next meeting or conference call and let me know your thoughts, feature requests, or feedback!
- ⭐ Star the repository: Dropping a ⭐ on
shareefmx/minomeethelps boost visibility and reach more open-source builders. - 👀 Follow on GitHub: Follow @shareefmx to stay updated on new releases, tools, and local AI projects.
- 💖 Sponsor on GitHub: If you'd like to support ongoing development and local AI research, consider sponsoring the project directly via GitHub Sponsors.
Distributed under the MIT License. See LICENSE for details.
Developed by @shareefmx.