English

Music Source Separation with Deep Equilibrium Models

Sound 2022-04-29 v2 Audio and Speech Processing

Abstract

While deep neural network-based music source separation (MSS) is very effective and achieves high performance, its model size is often a problem for practical deployment. Deep implicit architectures such as deep equilibrium models (DEQ) were recently proposed, which can achieve higher performance than their explicit counterparts with limited depth while keeping the number of parameters small. This makes DEQ also attractive for MSS, especially as it was originally applied to sequential modeling tasks in natural language processing and thus should in principle be also suited for MSS. However, an investigation of a good architecture and training scheme for MSS with DEQ is needed as the characteristics of acoustic signals are different from those of natural language data. Hence, in this paper we propose an architecture and training scheme for MSS with DEQ. Starting with the architecture of Open-Unmix (UMX), we replace its sequence model with DEQ. We refer to our proposed method as DEQ-based UMX (DEQ-UMX). Experimental results show that DEQ-UMX performs better than the original UMX while reducing its number of parameters by 30%.

Keywords

Cite

@article{arxiv.2110.06494,
  title  = {Music Source Separation with Deep Equilibrium Models},
  author = {Yuichiro Koyama and Naoki Murata and Stefan Uhlich and Giorgio Fabbro and Shusuke Takahashi and Yuki Mitsufuji},
  journal= {arXiv preprint arXiv:2110.06494},
  year   = {2022}
}

Comments

5 pages, 4 figures, accepted for publication in IEEE ICASSP 2022

R2 v1 2026-06-24T06:50:57.888Z