StenoAI

Your very own stenographer for every meeting

AI-powered meeting intelligence that runs entirely on your device, your private data never leaves anywhere. Record, transcribe, summarize, and query your meetings using local AI models. Perfect for healthcare, legal and finance professionals with confidential data needs.

_{Trusted by users at AWS, Deliveroo & Tesco.}

_{Disclaimer: This is an independent open-source project for meeting-notes productivity and is not affiliated with, endorsed by, or associated with any similarly named company.}

• 2026-03-11 🌍 Automatic language detection — Automatically detect your language
• 2026-03-04 🏷️ Auto-generated meeting titles — AI creates short titles from your transcripts
• 2026-03-04 ⌨️ macOS Shortcuts support — Start/stop recordings via Apple Shortcuts deep links
• 2026-03-04 🤖 Updated model lineup — Qwen 3.5 9B, DeepSeek-R1 14B, GPT-OSS 20B; Ollama v0.17.5
• 2026-03-04 🙈 Hide dock icon — Run StenoAI in menu-bar-only mode
• 2026-02-22 🌍 Multi-language support — Transcribe and summarize in 10 languages
• 2026-02-22 🖥️ Remote Ollama server — Run AI models on another machine on your network
• 2026-02-19 🎧 System audio capture — Record both sides of virtual meetings, even with headphones on
• 2026-02-19 📅 Google & Outlook Calendar integration — Auto-name recordings from your meetings

• Local transcription using whisper.cpp
• AI summarization with Ollama models
• Privacy-first - 100% local processing, your data never leaves your device
• Multiple AI models - Choose from 5 models optimized for different use cases
• Ask Steno - Query your meetings with natural language questions
• Multi-language support - Transcribe and summarize in 11 languages
• Remote Ollama server - Run AI models on another machine on your network
• System audio capture - Record mic + system audio simultaneously for virtual meetings with headphones
• macOS desktop app with intuitive interface

Transcription Models (Whisper):

• small: Default model - good accuracy and speed on Apple Silicon (default)
• base: Faster but lower accuracy for basic meetings
• medium: High accuracy for important meetings (slower)

Summarization Models (Ollama):

• llama3.2:3b (2GB): Fast and lightweight for quick meetings (default)
• gemma3:4b (2.5GB): Lightweight and efficient
• qwen3.5:9b (6.6GB): Excellent at structured output and action items
• deepseek-r1:14b (9.0GB): Strong reasoning and analysis capabilities
• gpt-oss:20b (14GB): OpenAI open-weight model with reasoning capabilities

_{Scored on: overall quality, factual accuracy, completeness, and hallucination (each out of 10).}

• Custom summarization templates
• Speaker Diarisation

• HIPAA compliance for healthcare workflows
• EHR integration for medical notes

Download the latest release for your Mac (requires macOS 14 Sonoma or later):

• Apple Silicon (M1-M5)
• Intel Macs Performance on Intel Macs is limited due to lack of dedicated AI inference capabilities on these older chips.

1. Download and open the DMG file

2. Drag the app to Applications

3. When you first launch the app, macOS may show a security warning

4. To fix this warning:

   • Go to System Settings > Privacy & Security and click "Open Anyway"

   Alternatively:

   • Right-click StenoAI in Applications and select "Open"
   • Or run in Terminal: xattr -cr /Applications/StenoAI.app

5. The app will work normally on subsequent launches

You can run it locally as well (see below) if you don't want to install a DMG.

• Python 3.9+
• Node.js 18+

git clone https://github.com/ruzin/stenoai.git
cd stenoai

# Backend setup
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

# Download bundled binaries (Ollama, ffmpeg)
./scripts/download-ollama.sh

# Build the Python backend
pip install pyinstaller
pyinstaller stenoai.spec --noconfirm

# Frontend
cd app
npm install
npm start

Note: Ollama and ffmpeg are bundled - no system installation needed. The setup wizard in the app will download the required AI models automatically.

cd app
npm run build

stenoai/
├── app/                  # Electron desktop app
├── src/                  # Python backend
├── website/              # Marketing site
├── recordings/           # Audio files
├── transcripts/          # Text output
└── output/              # Summaries

StenoAI includes a built-in debug panel for troubleshooting issues:

In-App Debug Panel:

1. Launch StenoAI
2. Click the 🔨 hammer icon (next to settings)
3. The debug panel shows real-time logs of all operations

Terminal Logging (Advanced): For detailed system-level logs, run the app from Terminal:

# Launch StenoAI with full logging
/Applications/StenoAI.app/Contents/MacOS/StenoAI

This displays comprehensive logs including:

• Python subprocess output
• Whisper transcription details
• Ollama API communication
• HTTP requests and responses
• Error stack traces
• Performance timing

System Console Logs: For system-level debugging:

# View recent StenoAI-related logs
log show --last 10m --predicate 'process CONTAINS "StenoAI" OR eventMessage CONTAINS "ollama"' --info

# Monitor live logs
log stream --predicate 'eventMessage CONTAINS "ollama" OR process CONTAINS "StenoAI"' --level info

Common Issues:

• Recording stops early: Check microphone permissions and available disk space
• "Processing failed": Usually Ollama service or model issues - check terminal logs
• Empty transcripts: Whisper couldn't detect speech - verify audio input levels
• Slow processing: Normal for longer recordings - Ollama processing is CPU-intensive especially on older intel Macs

• User Data: ~/Library/Application Support/stenoai/
• Recordings: ~/Library/Application Support/stenoai/recordings/
• Transcripts: ~/Library/Application Support/stenoai/transcripts/
• Summaries: ~/Library/Application Support/stenoai/output/

This project is licensed under the MIT License.