Monte Carlo Methods for Tempo Tracking and Rhythm Quantization

TL;DR

This paper introduces a probabilistic model for expressive music timing deviations, utilizing Monte Carlo methods to improve tempo tracking and rhythm transcription, with sequential Monte Carlo methods showing superior performance.

Contribution

It presents a novel probabilistic generative model for music timing, applying Monte Carlo techniques for inference in tempo tracking and rhythm quantization tasks.

Findings

Sequential Monte Carlo methods outperform MCMC in simulations.

The methods are effective for online and batch music analysis.

Applications include automatic accompaniment and music retrieval.

Abstract

We present a probabilistic generative model for timing deviations in expressive music performance. The structure of the proposed model is equivalent to a switching state space model. The switch variables correspond to discrete note locations as in a musical score. The continuous hidden variables denote the tempo. We formulate two well known music recognition problems, namely tempo tracking and automatic transcription (rhythm quantization) as filtering and maximum a posteriori (MAP) state estimation tasks. Exact computation of posterior features such as the MAP state is intractable in this model class, so we introduce Monte Carlo methods for integration and optimization. We compare Markov Chain Monte Carlo (MCMC) methods (such as Gibbs sampling, simulated annealing and iterative improvement) and sequential Monte Carlo methods (particle filters). Our simulation results suggest better results with sequential methods. The methods can be applied in both online and batch scenarios such as tempo tracking and transcription and are thus potentially useful in a number of music applications such as adaptive automatic accompaniment, score typesetting and music information retrieval.