Eita Nakamura: Bayesian Learning for Modelling Repeated and Modified Musical Note Patterns

When: Wednesday 21st June, 2017 @ 5:30 PM

Where: 4.31/4.33 Informatics Forum, University of Edinburgh

Seminar Title

Bayesian Learning for Modelling Repeated and Modified Musical Note Patterns

Seminar Speaker

Eita Nakamura (Kyoto University, Japan)

Seminar Abstract

Repetition is a characteristic structure in music data. In this talk, a statistical generative model incorporating the repetitive structure is discussed. From the viewpoint of statistical modelling, there are two challenges: (1) a generic model cannot describe the structure since the repeated patterns are different for individual pieces; (2) repetitions are often accompanied by modifications. A hierarchical Bayesian Markov model is proposed to solve these problems and semi-supervised Bayesian learning technique is proposed to induce the individual score model for each piece. Application for symbolic music transcription is also discussed.

Speaker Bio

Eita Nakamura is a JSPS Postdoctoral Research Fellow in the Speech and Audio Processing Group at Kyoto University and he is currently a visitor at the Centre for Digital Music at Queen Mary University of London. He received a PhD in Physics at the University of Tokyo in 2012 and have published papers on various topics on symbolic music processing including automatic music accompaniment, music transcription, and automatic music arrangement. His research interests include music modelling and analysis, music information processing, statistical machine learning, and application of complex systems for music phenomena.