MSRBench: A Benchmarking Dataset for Music Source Restoration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zang, Yongyi, Hai, Jiarui, Ge, Wanying, Kong, Qiuqiang, Dai, Zheqi, Wang, Helin, Mitsufuji, Yuki, Plumbley, Mark D.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918159322185728
author Zang, Yongyi
Hai, Jiarui
Ge, Wanying
Kong, Qiuqiang
Dai, Zheqi
Wang, Helin
Mitsufuji, Yuki
Plumbley, Mark D.
author_facet Zang, Yongyi
Hai, Jiarui
Ge, Wanying
Kong, Qiuqiang
Dai, Zheqi
Wang, Helin
Mitsufuji, Yuki
Plumbley, Mark D.
contents Music Source Restoration (MSR) extends source separation to realistic settings where signals undergo production effects (equalization, compression, reverb) and real-world degradations, with the goal of recovering the original unprocessed sources. Existing benchmarks cannot measure restoration fidelity: synthetic datasets use unprocessed stems but unrealistic mixtures, while real production datasets provide only already-processed stems without clean references. We present MSRBench, the first benchmark explicitly designed for MSR evaluation. MSRBench contains raw stem-mixture pairs across eight instrument classes, where mixtures are produced by professional mixing engineers. These raw-processed pairs enable direct evaluation of both separation accuracy and restoration fidelity. Beyond controlled studio conditions, the mixtures are augmented with twelve real-world degradations spanning analog artifacts, acoustic environments, and lossy codecs. Baseline experiments with U-Net and BSRNN achieve SI-SNR of -37.8 dB and -23.4 dB respectively, with perceptual quality (FAD CLAP) around 0.7-0.8, demonstrating substantial room for improvement and the need for restoration-specific architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10995
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MSRBench: A Benchmarking Dataset for Music Source Restoration
Zang, Yongyi
Hai, Jiarui
Ge, Wanying
Kong, Qiuqiang
Dai, Zheqi
Wang, Helin
Mitsufuji, Yuki
Plumbley, Mark D.
Sound
Music Source Restoration (MSR) extends source separation to realistic settings where signals undergo production effects (equalization, compression, reverb) and real-world degradations, with the goal of recovering the original unprocessed sources. Existing benchmarks cannot measure restoration fidelity: synthetic datasets use unprocessed stems but unrealistic mixtures, while real production datasets provide only already-processed stems without clean references. We present MSRBench, the first benchmark explicitly designed for MSR evaluation. MSRBench contains raw stem-mixture pairs across eight instrument classes, where mixtures are produced by professional mixing engineers. These raw-processed pairs enable direct evaluation of both separation accuracy and restoration fidelity. Beyond controlled studio conditions, the mixtures are augmented with twelve real-world degradations spanning analog artifacts, acoustic environments, and lossy codecs. Baseline experiments with U-Net and BSRNN achieve SI-SNR of -37.8 dB and -23.4 dB respectively, with perceptual quality (FAD CLAP) around 0.7-0.8, demonstrating substantial room for improvement and the need for restoration-specific architectures.
title MSRBench: A Benchmarking Dataset for Music Source Restoration
topic Sound
url https://arxiv.org/abs/2510.10995