SCNet: Sparse Compression Network for Music Source Separation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Tong, Weinan, Zhu, Jiaxu, Chen, Jun, Kang, Shiyin, Jiang, Tao, Li, Yang, Wu, Zhiyong, Meng, Helen
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909081503006720
author Tong, Weinan
Zhu, Jiaxu
Chen, Jun
Kang, Shiyin
Jiang, Tao
Li, Yang
Wu, Zhiyong
Meng, Helen
author_facet Tong, Weinan
Zhu, Jiaxu
Chen, Jun
Kang, Shiyin
Jiang, Tao
Li, Yang
Wu, Zhiyong
Meng, Helen
contents Deep learning-based methods have made significant achievements in music source separation. However, obtaining good results while maintaining a low model complexity remains challenging in super wide-band music source separation. Previous works either overlook the differences in subbands or inadequately address the problem of information loss when generating subband features. In this paper, we propose SCNet, a novel frequency-domain network to explicitly split the spectrogram of the mixture into several subbands and introduce a sparsity-based encoder to model different frequency bands. We use a higher compression ratio on subbands with less information to improve the information density and focus on modeling subbands with more information. In this way, the separation performance can be significantly improved using lower computational consumption. Experiment results show that the proposed model achieves a signal to distortion ratio (SDR) of 9.0 dB on the MUSDB18-HQ dataset without using extra data, which outperforms state-of-the-art methods. Specifically, SCNet's CPU inference time is only 48% of HT Demucs, one of the previous state-of-the-art models.
format Preprint
id arxiv_https___arxiv_org_abs_2401_13276
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SCNet: Sparse Compression Network for Music Source Separation
Tong, Weinan
Zhu, Jiaxu
Chen, Jun
Kang, Shiyin
Jiang, Tao
Li, Yang
Wu, Zhiyong
Meng, Helen
Audio and Speech Processing
Deep learning-based methods have made significant achievements in music source separation. However, obtaining good results while maintaining a low model complexity remains challenging in super wide-band music source separation. Previous works either overlook the differences in subbands or inadequately address the problem of information loss when generating subband features. In this paper, we propose SCNet, a novel frequency-domain network to explicitly split the spectrogram of the mixture into several subbands and introduce a sparsity-based encoder to model different frequency bands. We use a higher compression ratio on subbands with less information to improve the information density and focus on modeling subbands with more information. In this way, the separation performance can be significantly improved using lower computational consumption. Experiment results show that the proposed model achieves a signal to distortion ratio (SDR) of 9.0 dB on the MUSDB18-HQ dataset without using extra data, which outperforms state-of-the-art methods. Specifically, SCNet's CPU inference time is only 48% of HT Demucs, one of the previous state-of-the-art models.
title SCNet: Sparse Compression Network for Music Source Separation
topic Audio and Speech Processing
url https://arxiv.org/abs/2401.13276