Moises-Light: Resource-efficient Band-split U-Net For Music Source Separation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yun-Ning, Hung, Pereira, Igor, Korzeniowski, Filip
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916996281532416
author Yun-Ning
Hung
Pereira, Igor
Korzeniowski, Filip
author_facet Yun-Ning
Hung
Pereira, Igor
Korzeniowski, Filip
contents In recent years, significant advances have been made in music source separation, with model architectures such as dual-path modeling, band-split modules, or transformer layers achieving comparably good results. However, these models often contain a significant number of parameters, posing challenges to devices with limited computational resources in terms of training and practical application. While some lightweight models have been introduced, they generally perform worse compared to their larger counterparts. In this paper, we take inspiration from these recent advances to improve a lightweight model. We demonstrate that with careful design, a lightweight model can achieve comparable SDRs to models with up to 13 times more parameters. Our proposed model, Moises-Light, achieves competitive results in separating four musical stems on the MUSDB-HQ benchmark dataset. The proposed model also demonstrates competitive scalability when using MoisesDB as additional training data.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06785
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Moises-Light: Resource-efficient Band-split U-Net For Music Source Separation
Yun-Ning
Hung
Pereira, Igor
Korzeniowski, Filip
Audio and Speech Processing
Sound
In recent years, significant advances have been made in music source separation, with model architectures such as dual-path modeling, band-split modules, or transformer layers achieving comparably good results. However, these models often contain a significant number of parameters, posing challenges to devices with limited computational resources in terms of training and practical application. While some lightweight models have been introduced, they generally perform worse compared to their larger counterparts. In this paper, we take inspiration from these recent advances to improve a lightweight model. We demonstrate that with careful design, a lightweight model can achieve comparable SDRs to models with up to 13 times more parameters. Our proposed model, Moises-Light, achieves competitive results in separating four musical stems on the MUSDB-HQ benchmark dataset. The proposed model also demonstrates competitive scalability when using MoisesDB as additional training data.
title Moises-Light: Resource-efficient Band-split U-Net For Music Source Separation
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2510.06785