Drum-Aware Ensemble Architecture for Improved Joint Musical Beat and Downbeat Tracking

TL;DR

This paper introduces a novel ensemble system that combines source separation with joint beat and downbeat tracking, improving accuracy by adaptively weighting percussive and non-percussive components in musical signals.

Contribution

The paper proposes a new architecture integrating blind source separation with beat and downbeat tracking, enhancing performance over traditional methods without source separation.

Findings

Outperforms baseline architecture across multiple datasets

Effectively adapts to varying levels of drum sounds

Demonstrates robustness in diverse musical contexts

Abstract

This paper presents a novel system architecture that integrates blind source separation with joint beat and downbeat tracking in musical audio signals. The source separation module segregates the percussive and non-percussive components of the input signal, over which beat and downbeat tracking are performed separately and then the results are aggregated with a learnable fusion mechanism. This way, the system can adaptively determine how much the tracking result for an input signal should depend on the input's percussive or non-percussive components. Evaluation on four testing sets that feature different levels of presence of drum sounds shows that the new architecture consistently outperforms the widely-adopted baseline architecture that does not employ source separation.