Self-Supervised Hierarchical Metrical Structure Modeling

TL;DR

This paper introduces a self-supervised approach to model hierarchical metrical structures in music, capable of analyzing both symbolic and audio signals without extensive labeled data, achieving performance comparable to supervised methods.

Contribution

The novel method models hierarchical metrical structures using minimal domain knowledge and self-supervision, applicable to both symbolic and audio music signals.

Findings

Achieves comparable performance with supervised methods

Works on both symbolic music and audio signals

Requires no hierarchical labels beyond beat annotations

Abstract

We propose a novel method to model hierarchical metrical structures for both symbolic music and audio signals in a self-supervised manner with minimal domain knowledge. The model trains and inferences on beat-aligned music signals and predicts an 8-layer hierarchical metrical tree from beat, measure to the section level. The training procedure does not require any hierarchical metrical labeling except for beats, purely relying on the nature of metrical regularity and inter-voice consistency as inductive biases. We show in experiments that the method achieves comparable performance with supervised baselines on multiple metrical structure analysis tasks on both symbolic music and audio signals. All demos, source code and pre-trained models are publicly available on GitHub.