ASR Pro

A professional desktop application for AI-powered speech recognition and transcription, built with Tauri + React + Python Sidecar architecture.

Features

• Global hotkey transcription with customizable key combinations
• Real-time audio processing and transcription
• Multi-model AI support (Whisper, Parakeet)
• SRT subtitle file generation
• Drag-and-drop file transcription
• Native macOS-style UI components
• Dark theme interface
• System tray integration
• Cross-platform desktop application

Architecture

• Frontend: React 18 + TypeScript + Vite
• Desktop: Tauri v2 (Rust-based)
• Backend: Python FastAPI sidecar
• AI Models: ONNX Runtime with Whisper/Parakeet models
• UI Components: Custom macOS-native component library

Prerequisites

Required Software

• Node.js: Version 20.19+ or 22.12+ (current: 20.17.0 with warnings)
• Python: Version 3.8+ with pip
• Rust: Latest stable toolchain via rustup
• Git: For cloning and version control

System Requirements

• Windows: Windows 10/11 (x64)
• macOS: macOS 10.15+ (Intel/Apple Silicon)
• Linux: Ubuntu 18.04+ or equivalent
• Memory: 4GB RAM minimum, 8GB recommended
• Storage: 2GB free space for models and dependencies

Installation

1. Clone Repository

git clone <repository-url>
cd asrpro

2. Install Node.js Dependencies

npm install

3. Install Python Dependencies

cd sidecar
pip install -r requirements.txt
cd ..

4. Install Rust Toolchain

# Windows/macOS/Linux
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
source ~/.cargo/env

# Or use the installer directly
# Windows: Download from https://win.rustup.rs/

5. Download AI Models

The application will automatically download required ONNX models on first run:

• Whisper-tiny (39MB)
• Whisper-base (74MB)
• Additional models available through UI

Development

Start Development Server

# Terminal 1: Start Python sidecar
cd sidecar
python main.py

# Terminal 2: Start Tauri development
npm run tauri dev

Available Scripts

# Frontend development
npm run dev          # Start Vite dev server
npm run build        # Build frontend for production
npm run preview      # Preview production build

# Tauri desktop
npm run tauri dev    # Start Tauri development mode
npm run tauri build  # Build desktop application

Project Structure

asrpro/
├── src/                    # React frontend source
│   ├── components/         # React components
│   │   ├── macos/         # macOS-native UI components
│   │   └── ...
│   ├── pages/             # Application pages
│   └── ...
├── src-tauri/             # Tauri configuration and Rust code
├── sidecar/               # Python backend
│   ├── api/              # FastAPI server
│   ├── models/           # AI model management
│   ├── utils/            # Utilities
│   └── tests/            # Backend tests
└── dist/                 # Built frontend assets

Testing

Backend Tests

cd sidecar
python -m pytest
python -m pytest -v                    # Verbose output
python -m pytest tests/test_api.py     # Specific test file

Frontend Tests

npm run test            # Run frontend tests (if configured)
npm run type-check      # TypeScript type checking

Clean Python Cache

# Remove all __pycache__ directories and .pyc files
python scripts/clean_python_cache.py

# Or manually with PowerShell (Windows)
Get-ChildItem -Path . -Name "__pycache__" -Recurse -Directory | Remove-Item -Recurse -Force

Manual Testing

1. Start development servers
2. Test global hotkey functionality
3. Upload audio files for transcription
4. Verify model switching
5. Test system tray integration

Building for Production

Desktop Application

# Build complete desktop application
npm run tauri build

Output files:

• Windows: src-tauri/target/release/bundle/nsis/asrpro_0.1.0_x64-setup.exe
• Windows: src-tauri/target/release/bundle/msi/asrpro_0.1.0_x64_en-US.msi
• macOS: src-tauri/target/release/bundle/macos/asrpro.app
• Linux: src-tauri/target/release/bundle/appimage/asrpro_0.1.0_amd64.AppImage

Frontend Only

npm run build
# Output: dist/ directory

Configuration

Environment Variables

Create .env files as needed:

# .env.local (frontend)
VITE_API_URL=http://localhost:8000

# sidecar/.env (backend)
MODEL_CACHE_DIR=./models
LOG_LEVEL=INFO

Model Configuration

Models are automatically managed through the UI. Manual configuration in sidecar/models/registry.py.

Troubleshooting

Common Issues

Node.js Version Warning

• Current version 20.17.0 works but upgrade to 20.19+ recommended
• Use nvm/volta for Node.js version management

Rust Installation

• Ensure ~/.cargo/bin is in PATH
• Restart terminal after installation
• Run rustc --version to verify

Python Dependencies

• Use virtual environment: python -m venv venv && source venv/bin/activate
• Install Microsoft Visual C++ Build Tools (Windows)
• Install system dependencies for audio processing

Build Failures

• Clear node_modules: rm -rf node_modules && npm install
• Clear Rust cache: cargo clean in src-tauri/
• Verify all prerequisites are installed

Performance

• Models download automatically but can be pre-cached
• First transcription may be slower due to model loading
• GPU acceleration available with compatible hardware

Contributing

1. Fork the repository
2. Create feature branch: git checkout -b feature-name
3. Clean Python cache: python scripts/clean_python_cache.py
4. Run tests: npm run test && cd sidecar && python -m pytest
5. Commit changes: git commit -m "Description"
6. Push and create pull request

Repository Cleanliness

• Python cache files (__pycache__/, *.pyc) are automatically ignored
• Run cleanup script before committing: python scripts/clean_python_cache.py
• Virtual environments (.venv/, venv/) are ignored
• Node modules and build artifacts are ignored

License

See LICENSE file for details.

Author

Made by Suraj Mandal