Tempo-Invariant Processing of Rhythm with Convolutional Neural Networks

TL;DR

This paper proposes a log-frequency representation of rhythm activations processed with CNNs to achieve tempo invariance in rhythm analysis, enhancing robustness across different tempi in music information retrieval tasks.

Contribution

It introduces a novel log-frequency rhythm representation that inherently promotes tempo invariance when used with convolutional neural networks, advancing rhythm processing techniques.

Findings

Log-frequency representation improves tempo invariance

Different methods based on magnitude and phase relationships are explored

Variations provide insights for future rhythm processing implementations

Abstract

Rhythm patterns can be performed with a wide variation of tempi. This presents a challenge for many music information retrieval (MIR) systems; ideally, perceptually similar rhythms should be represented and processed similarly, regardless of the specific tempo at which they were performed. Several recent systems for tempo estimation, beat tracking, and downbeat tracking have therefore sought to process rhythm in a tempo-invariant way, often by sampling input vectors according to a precomputed pulse level. This paper describes how a log-frequency representation of rhythm-related activations instead can promote tempo invariance when processed with convolutional neural networks. The strategy incorporates invariance at a fundamental level and can be useful for most tasks related to rhythm processing. Different methods are described, relying on magnitude, phase relationships of different rhythm channels, as well as raw phase information. Several variations are explored to provide direction for future implementations.