Acknowledgements

Development Team

  • Facundo Franchino - Lead Developer, School of Physics, Engineering and Technology, University of York, UK

Academic Supervision

Special thanks to Andy Hunt for supervision throughout the development of Amadeus.

External Consultants

The following experts provided invaluable consulting regarding automatic chord recognition research and early implementation approaches:

  • Michael McLoughlin (University of York)

  • Filip Korzeniowski (Moises.ai)

  • Christian Dittmar (AudioLabs Erlangen)

  • Eloi Moliner (Aalto/Meta)

Their guidance was particularly helpful during the exploration of non-negative matrix factorisation and other DSP-based approaches, which ultimately informed the decision to adopt a more pragmatic architecture for the current implementation.

Research Support

N8 CiR Partnership

Thanks to N8 CiR (N8 Centre of Excellence in Computationally Intensive Research) for funding upcoming research on automatic chord recognition, which will be used to improve the algorithm and the Swift Basic Pitch implementation of Amadeus.

This partnership will enable continued development throughout 2026 and beyond, ensuring Amadeus evolves from its current formative state into a production-ready system.

Open Source Technologies

Amadeus builds upon several open source projects:

  • Basic Pitch by Spotify - Temporary transcription backend

  • FastAPI - High-performance Python web framework

  • SwiftUI - Apple’s declarative UI framework

  • librosa - Audio analysis library (used in experiments)

References

Key research papers that informed the development:

      1. Bittner, J. J. Bosch, D. Rubinstein, G. Meseguer-Brocal, and S. Ewert, “A Lightweight Instrument-Agnostic Model for Polyphonic Note Transcription and Multipitch Estimation,” arXiv:2203.09893, 2022.

    1. Pauwels, K. O’Hanlon, E. Gómez, and M. B. Sandler, “20 Years of Automatic Chord Recognition from Audio,” 2019.

    1. López-Serrano and C. Dittmar, “NMF Toolbox: Music Processing Applications of Nonnegative Matrix Factorization,” 2019.

Community

This project is part of ongoing research in music information retrieval and aims to make chord recognition technology accessible to musicians worldwide.

Future contributions and collaborations are welcome as the project evolves through 2026 and beyond.