Steno is a sophisticated desktop application that transforms audio into accurate text using state-of-the-art AI technology. Built with performance and user experience in mind, it offers both real-time recording and file-based transcription capabilities with support for multiple languages and audio formats.

• Advanced Audio Processing - Support for MP3, WAV, FLAC, OGG, AAC, M4A, WMA with intelligent format conversion
• Real-time Transcription - Live audio capture with simultaneous speech recognition
• Multi-language Support - Optimized for Chinese, English, and Indonesian with automatic language detection
• AI Model Flexibility - Choose from Tiny (39MB), Base (74MB), or Large v3 (1.5GB) Whisper models
• Native macOS Performance - Optimized desktop application for macOS
• Apple Silicon Optimization - Metal GPU acceleration on M1/M2/M3 Macs

# Download and install
curl -L -o Steno.dmg https://github.com/xazaj/Steno/releases/download/v1.0.0/Steno_1.0.0_aarch64.dmg
open Steno.dmg

# Configure permissions (required for unsigned apps)
xattr -rd com.apple.quarantine "/Applications/Steno.app"
codesign --force --deep --sign - "/Applications/Steno.app"

1. Model Selection: Choose your preferred Whisper model based on your needs:

   • Tiny Model (39MB) - Fast processing, basic accuracy
   • Base Model (74MB) - Balanced performance and quality
   • Large v3 Model (1.5GB) - Highest accuracy, slower processing

2. Permissions: Grant microphone access for real-time transcription features

Audio Input → Processing → AI Recognition → Text Output → Export

• Smart Prompts - Context-aware templates for meetings, interviews, medical, and technical content
• Speaker Diarization - Automatic identification and separation of different speakers
• Export Options - Multiple formats including TXT, SRT, JSON, and Markdown
• Search & Organization - Tag-based categorization with powerful filtering capabilities

~/Library/Application Support/com.steno.app/
├── database/
│   └── steno.db                    # SQLite database (transcription records & settings)
├── models/                         # Whisper AI model files
│   ├── ggml-tiny.bin              # Tiny model (~39MB)
│   ├── ggml-base.bin              # Base model (~142MB)
│   └── ggml-large-v3.bin          # Large v3 model (~1.55GB)
├── audio/
│   ├── uploads/                    # User uploaded audio files
│   └── temp/                       # Temporary audio files
└── logs/                          # Application logs
    └── app.log

# View application data
open "~/Library/Application Support/com.steno.app/"

# Backup database
cp "~/Library/Application Support/com.steno.app/database/steno.db" ~/Desktop/steno_backup.db

# Clean temporary files
rm -rf "~/Library/Application Support/com.steno.app/audio/temp/"

• Rust 1.70+
• Node.js 18+
• Git

# Clone repository
git clone https://github.com/xazaj/Steno.git
cd steno

# Install dependencies
npm install

# Start development server
npm run tauri:dev

# Development with hot reload
npm run tauri:dev

# Production build
npm run tauri:build

# Platform-specific builds
npm run build:mac-m1      # Apple Silicon

steno/
├── src/                  # React frontend
│   ├── components/       # UI components
│   ├── hooks/           # Custom React hooks
│   ├── types/           # TypeScript definitions
│   └── utils/           # Utility functions
├── src-tauri/           # Rust backend
│   ├── src/             # Core application logic
│   ├── lib/             # whisper.cpp integration
│   └── capabilities/    # Tauri security permissions
├── docs/                # Documentation
└── models/              # AI model storage

We welcome contributions from the community. Please read our Contributing Guidelines before getting started.

1. Fork the repository
2. Create a feature branch (git checkout -b feature/new-feature)
3. Commit your changes (git commit -m 'Add new feature')
4. Push to the branch (git push origin feature/new-feature)
5. Open a Pull Request

• Follow Conventional Commits for commit messages
• Run tests before submitting: npm test && cargo test
• Ensure code formatting: npm run lint && cargo fmt
• Add documentation for new features

This project is licensed under the MIT License - see the LICENSE file for details.

• OpenAI Whisper - State-of-the-art speech recognition models
• whisper.cpp - High-performance C++ implementation
• Tauri - Modern desktop application framework
• Symphonia - Professional audio decoding library

• Bug Reports: GitHub Issues
• Feature Requests: GitHub Issues
• Documentation: Project Wiki
• Discussions: GitHub Discussions