Determined Multichannel Blind Source Separation with Clustered Source Model

TL;DR

This paper introduces a novel multichannel blind source separation method using a clustered source model based on nonnegative block-term decomposition, improving source independence modeling and outperforming existing ILRMA techniques.

Contribution

It proposes a new clustered source model with NBTD that captures inter-channel dependencies and allows for orthogonality constraints, enhancing separation performance.

Findings

Outperforms ILRMA in anechoic conditions

Surpasses ILRMA in reverberant environments

Provides interpretable latent source representations

Abstract

The independent low-rank matrix analysis (ILRMA) method stands out as a prominent technique for multichannel blind audio source separation. It leverages nonnegative matrix factorization (NMF) and nonnegative canonical polyadic decomposition (NCPD) to model source parameters. While it effectively captures the low-rank structure of sources, the NMF model overlooks inter-channel dependencies. On the other hand, NCPD preserves intrinsic structure but lacks interpretable latent factors, making it challenging to incorporate prior information as constraints. To address these limitations, we introduce a clustered source model based on nonnegative block-term decomposition (NBTD). This model defines blocks as outer products of vectors (clusters) and matrices (for spectral structure modeling), offering interpretable latent vectors. Moreover, it enables straightforward integration of orthogonality constraints to ensure independence among source images. Experimental results demonstrate that our proposed method outperforms ILRMA and its extensions in anechoic conditions and surpasses the original ILRMA in simulated reverberant environments.