HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Conference papers

Autoregressive hidden semi-Markov model of symbolic music performance for score following

Eita Nakamura 1, * Philippe Cuvillier 2, 3 Arshia Cont 3, 2 Nobutaka Ono 1 Shigeki Sagayama 4
* Corresponding author
2 MuTant - Synchronous Realtime Processing and Programming of Music Signals
Inria Paris-Rocquencourt, UPMC - Université Pierre et Marie Curie - Paris 6, IRCAM - Institut de Recherche et Coordination Acoustique/Musique, CNRS - Centre National de la Recherche Scientifique
3 Repmus - Représentations musicales
STMS - Sciences et Technologies de la Musique et du Son
Résumé : A stochastic model of symbolic (MIDI) performance of polyphonic scores is presented and applied to score following. Stochastic modelling has been one of the most successful strategies in this field. We describe the performance as a hierarchical process of performer's progression in the score and the production of performed notes, and represent the process as an extension of the hidden semi-Markov model. The model is compared with a previously studied model based on hidden Markov model (HMM), and reasons are given that the present model is advantageous for score following especially for scores with trills, tremolos, and arpeggios. This is also confirmed empirically by comparing the accuracy of score following and analysing the errors. We also provide a hybrid of this model and the HMM-based model which is computationally more efficient and retains the advantages of the former model. The present model yields one of the state-of-the-art score following algorithms for symbolic performance and can possibly be applicable for other music recognition problems.
Document type :
Conference papers
Complete list of metadata

Cited literature [16 references]  Display  Hide  Download

Contributor : Arshia Cont Connect in order to contact the contributor
Submitted on : Tuesday, August 11, 2015 - 1:12:29 PM
Last modification on : Tuesday, March 15, 2022 - 3:21:20 AM
Long-term archiving on: : Thursday, November 12, 2015 - 10:34:46 AM


Files produced by the author(s)


  • HAL Id : hal-01183820, version 1


Eita Nakamura, Philippe Cuvillier, Arshia Cont, Nobutaka Ono, Shigeki Sagayama. Autoregressive hidden semi-Markov model of symbolic music performance for score following. 16th International Society for Music Information Retrieval Conference (ISMIR), Oct 2015, Malaga, Spain. ⟨hal-01183820⟩



Record views


Files downloads