MML 2021

14th International Workshop on Machine Learning and Music

Barcelona & online, December 15, 2021


Machine learning and artificial intelligence have permeated nearly every area of music informatics, driven by a profusion of recordings available in digital audio formats, steady improvements to the accessibility and quality of symbolic corpora, availability of powerful algorithms in standard machine learning toolboxes, and theoretical advances in machine learning and data mining. As the complexity of the problems investigated by researchers on machine learning and music increases, there is a need to develop new algorithms and methods. As a consequence, research on machine learning and music is an active and growing field reflected in international meetings. MML 2020 is the 13th edition of the International Workshops on Machine Learning and Music (MML): 2008 (ICML, Helsinki, Finland), 2009 (ECML, Bled, Slovenia), 2010 (ACM-MM, Florence, Italy), 2011 (NIPS, Sierra Nevada, Spain), 2012 (ICML, Edinburgh, Scotland), 2013 (ECML/PKDD, Prague, Czech Republic), 2014 (Barcelona, Spain), 2015 (ISEA, Vancouver, Canada), 2016 (ECML/PKDD, Riva del Garda, Italy), 2017 (Barcelona, Spain), 2018 (ICML, Stockholm, Sweden), 2019 (ECML, Würzburg, Germany), and 2020 (ECML, Online).


Papers in all applications on music and machine learning are welcome, including but not limited to

  • Automatic classification of music (audio and MIDI)

  • Style-based interpreter recognition

  • Music recommender systems, genre and tag prediction

  • Automatic score alignment

  • Polyphonic pitch detection

  • Chord extraction

  • Pattern discovery

  • Deep learning in music applications

  • Beat tracking

  • Expressive performance modeling

  • Statistical models for music prediction and generation


Rafael Ramirez, Universitat Pompeu Fabra, Spain

Darrell Conklin, University of the Basque Country, Spain

José Manuel Iñesta, University of Alicante, Spain


15:00 Welcome

15:10 MultiScore Project: Multimodal Transcription of Music Scores

Jorge Calvo-Zaragoza, A. Pertusa, A.J. Gallego, José M. Iñesta, L. Micó, J. Oncina, C. Pérez-Sancho, P-J. Ponce de León, D. Rizo

15:20 3-D motion generation for double bass performance from musical score

Shinji Sako, Takeru Shirai

15:30 Playable Audio Texture Models

Lonce Wyse, Chitralekha Gupta, Purnima Kamath

15:40 Music demixing with the sliCQ transform

Sevag Hanssian

15:50 Generating expressive features of music performances with deep sequence models

Fabio Muneratti Ortega, Rafael Ramirez

16:00 An Unsupervised Domain Adaptation framework for LayoutAnalysis of Music Score Images

Francisco J. Castellanos, Antonio Javier Gallego, Jorge Calvo-Zaragoza

16:10 AugmentedNet: A Convolutional Recurrent Neural Network forAutomatic Roman Numeral Analysis with Improved Data Augmentation

Néstor Népoles López, Mark Gotham, and Ichiro Fujinaga

16:20 Automatic Assessment of Tone Quality in Violin Music Performance

Sergio Giraldo, George Waddell, Rafael Ramirez

16:30 Error modeling and correction in Automatic Music Transcriptionvia note-level Music Language Models

Jose J. Valero-Mas, Andrew McLeod

16:40 CTC-based end-to-end approach for full page Optical Music Recognition

Antonio Ríos-Vila, Jorge Calvo-Zaragoza, José M. Iñesta

16:50 Error Detection in Symbolic Music for OMR Post-Processing

Timothy de Reuse, Ichiro Fujinaga

17:00 Closing

*Central European time


You can find the proceedings here