Repertorium’s OMR tools
Pioneering Music Digitization with Advanced Deep Learning: prAig's Role in REPERTORIUM
The Pattern Recognition and Artificial Intelligence Group (prAig) from the University of Alicante in Spain is considered one of the most relevant research groups in the world when working with musical information. In REPERTORIUM, the digitisation process of musical manuscripts using Optical Music Recognition will be done by them.
prAig employs state-of-the-art deep learning technology, integrating Convolutional Recurrent Neural Networks and the Transformer architecture to achieve a comprehensive transcription of music notation from images. This approach has demonstrated remarkable success across diverse scenarios, including monophonic music [1], piano scores [2], mensural notation [3], and even simultaneous transcription of notes and lyrics [4]. This advanced technology will also be leveraged to train new models for processing medieval and classical music within REPERTORIUM.
References
[1] Calvo-Zaragoza, J., & Rizo, D. (2018). Camera-PrIMuS: Neural End-to-End Optical Music Recognition on Realistic Monophonic Scores. In Proceedings of the 19th International Society for Music Information Retrieval Conference, ISMIR 2018 (pp. 248-255).
[2] Ríos-Vila, A., Rizo, D., Iñesta, J. M., & Calvo-Zaragoza, J. (2023). End-to-end optical music recognition for pianoform sheet music. International Journal on Document Analysis and Recognition (IJDAR), 1-16.
[3] Calvo-Zaragoza, J., Toselli, A. H., & Vidal, E. (2019). Handwritten music recognition for mensural notation with convolutional recurrent neural networks. Pattern Recognition Letters, 128, 115-121.
[4] Martinez-Sevilla, J. C., Rios-Vila, A., Castellanos, F. J., & Calvo-Zaragoza, J. (2023). A Holistic Approach for Aligned Music and Lyrics Transcription. In International Conference on Document Analysis and Recognition (pp. 185-201). Cham: Springer Nature Switzerland.