CubbyScore Converter

Convert PDF music scores to MusicXML format using optical music recognition (OMR).

Download

Platform	Architecture	Download
macOS	Apple Silicon	CubbyScore Converter-1.0.0-arm64.dmg
macOS	Intel	CubbyScore Converter-1.0.0.dmg
Windows	x64	CubbyScore Converter Setup 1.0.0.exe

Features

• PDF to MusicXML conversion - Convert any PDF music score to editable MusicXML
• Intelligent validation - Prioritized accuracy reporting for:
  1. Metadata (Title, Composer, Instruments)
  2. Clefs
  3. Time Signatures
  4. Tempo Markings
  5. Notes
• Confidence scoring - See how accurately each element was detected
• No installation hassles - Everything bundled (Python backend, Audiveris OMR, Java runtime)
• Runs locally - Your scores never leave your computer

Screenshots

Coming soon

System Requirements

• macOS: 10.15 (Catalina) or later
• Windows: Windows 10 or later (64-bit)
• RAM: 4GB minimum, 8GB recommended
• Storage: 500MB for installation

How It Works

CubbyScore Converter uses Audiveris, a powerful open-source OMR engine, to analyze PDF scores and extract musical notation. The results are validated using music21 and presented with confidence scores.

Conversion Process

1. Upload - Select a PDF music score
2. Process - Audiveris analyzes the score (typically 30-60 seconds per page)
3. Validate - music21 extracts and validates musical elements
4. Download - Get your MusicXML file ready for use in notation software

Tips for Best Results

• Clean PDFs work best - Digital/vector PDFs give better results than scanned images
• Split large scores - For scores over 50 pages, consider splitting into sections
• Check the confidence scores - Lower scores indicate areas that may need manual correction

Technology

• Electron - Cross-platform desktop app framework
• Audiveris - Optical Music Recognition engine
• FastAPI - Python backend for processing
• music21 - Music notation analysis
• Next.js - Frontend UI

Development

See PROGRESS.md for development status and build instructions.

Building from Source

# Clone the repository
git clone https://github.com/yourusername/cubby-score-conversion.git
cd cubby-score-conversion

# Build Python backend
cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
pyinstaller cubbyscore-backend.spec

# Build Electron app
cd ../electron-app
npm install
npm run electron:build:mac  # or electron:build:win

License

MIT License - See LICENSE for details.

Author

Willard Jansen - Cubby

Troubleshooting

Browser shows ERR_SSL_PROTOCOL_ERROR

If you see SSL errors:

1. Make sure you're using http:// (not https://) in the URL
2. Try using an incognito/private browsing window
3. Clear the browser cache if you previously tried with HTTPS

Acknowledgments

• Audiveris - Open source OMR engine
• music21 - Music analysis toolkit
• Electron - Desktop app framework