DisCover: Disentangled Music Representation Learning for Cover Song Identification

TL;DR

DisCover introduces a novel disentangled music representation learning framework for cover song identification, effectively separating version-specific and invariant features to improve identification accuracy amidst high variances.

Contribution

The paper proposes a new framework with knowledge-guided and adversarial modules to disentangle factors, addressing intra- and inter-version biases in cover song identification.

Findings

DisCover outperforms existing methods in cover song identification accuracy.

Disentanglement significantly improves invariant music representation learning.

The framework effectively blocks intra- and inter-version effects.

Abstract

In the field of music information retrieval (MIR), cover song identification (CSI) is a challenging task that aims to identify cover versions of a query song from a massive collection. Existing works still suffer from high intra-song variances and inter-song correlations, due to the entangled nature of version-specific and version-invariant factors in their modeling. In this work, we set the goal of disentangling version-specific and version-invariant factors, which could make it easier for the model to learn invariant music representations for unseen query songs. We analyze the CSI task in a disentanglement view with the causal graph technique, and identify the intra-version and inter-version effects biasing the invariant learning. To block these effects, we propose the disentangled music representation learning framework (DisCover) for CSI. DisCover consists of two critical components: (1) Knowledge-guided Disentanglement Module (KDM) and (2) Gradient-based Adversarial Disentanglement Module (GADM), which block intra-version and inter-version biased effects, respectively. KDM minimizes the mutual information between the learned representations and version-variant factors that are identified with prior domain knowledge. GADM identifies version-variant factors by simulating the representation transitions between intra-song versions, and exploits adversarial distillation for effect blocking. Extensive comparisons with best-performing methods and in-depth analysis demonstrate the effectiveness of DisCover and the and necessity of disentanglement for CSI.