Papers › SCNet: Sparse Compression Network for Music Source Separation

SCNet: Sparse Compression Network for Music Source Separation

24 Jan 2024arXiv:2401.13276archive 2025-07-28

Weinan Tong, Jiaxu Zhu, Jun Chen, Shiyin Kang, Tao Jiang, Yang Li, Zhiyong Wu, Helen Meng

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.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

starrytong/SCNet officialmentioned on GitHubpytorch report
amanteur/SCNet-PyTorch mentioned on GitHubpytorch report
neeraj-j/SCNet mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Music Source Separation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Music Source Separation MUSDB18-HQ SCNet-large SDR (avg) 9.69 #3 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet-large SDR (bass) 9.49 #3 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet-large SDR (drums) 10.98 #3 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet-large SDR (others) 7.44 #3 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet-large SDR (vocals) 10.86 #3 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet SDR (avg) 9.00 #6 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet SDR (bass) 8.82 #6 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet SDR (drums) 10.51 #6 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet SDR (others) 6.76 #6 of 14 Archive leaderboard report
Music Source Separation MUSDB18-HQ SCNet SDR (vocals) 9.89 #6 of 14 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

FocusSCNet

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections