PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Predictive models for music
Jean-Francois Paiement, Yves Grandvalet and Samy Bengio
Connection Science Volume 21, Number 2, pp. 253-272, 2009.

Abstract

Modeling long-term dependencies in time series has proved very difficult to achieve with traditional machine learning methods. This problem occurs when considering music data. In this paper, we introduce predictive models for melodies. We decompose melodic modeling into two subtasks. We first propose a rhythm model based on the distributions of distances between subsequences. Then, we define a generative model for melodies given chords and rhythms based on modeling sequences of Narmour features. The rhythm model consistently outperforms a standard Hidden Markov Model in terms of conditional prediction accuracy on two different music databases. Using a similar evaluation procedure, the proposed melodic model consistently outperforms an Input/Output Hidden Markov Model. Furthermore, these models are able to generate realistic melodies given appropriate musical contexts.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:6822
Deposited By:Yves Grandvalet
Deposited On:08 March 2010