arXiv：cs.LG（机器学习，全量分类）· Steven Cho, Junghyun Koo, Raphael Lafargue, Tushar Dhyani, Eloi Moliner, Yuki Mitsufuji·· 13 小时前AI 评分41
基于潜空间流匹配的端到端历史管弦乐修复
End-to-End Historical Music Restoration in Latent Space
AI 导读
研究提出一种监督式端到端管弦乐历史音乐修复（HMR）基准，通过更忠实地模拟历史录音退化链，将多乐器管弦乐修复转化为可监督问题。在合成退化数据对上训练的潜空间流匹配模型，在侵入式、非侵入式与主观评测中均优于现有 HMR 基线。团队同时发布一个 9.3 小时免授权、非配对的古典音乐测试集及代码与音频演示。
正文
Abstract:Historical music restoration (HMR) has almost exclusively focused on constrained problems such as Super-Resolution or the restoration of solo pieces, under-exploring the general task of restoring orchestral historical music, which has multiple instruments. This under-exploration is largely because the HMR domain, early-20th-century recordings, has no pre-degradation ground-truth pairs, making the restoration task unsupervised and more challenging. This paper presents a supervised end-to-end orchestral HMR benchmark by exploring both the synthetic degradation functions and the end-to-end generative deep-learning restoration methods. We simulate the historical recording degradation chain more faithfully than prior work, which makes orchestral restoration into a tractable supervised problem. A latent flow-matching model trained on the resulting synthetic pairs outperforms existing HMR baselines on intrusive, non-intrusive, and subjective evaluations. We also curate and release a 9.3-hour license-free, unpaired, historical classical-music test set, along with code and audio demos.
| Comments: | 5 pages, 2 figures, 3 tables; submitted to ICASSP 2027. Code and audio demos available at the project repository |
| Subjects: | Audio and Speech Processing (eess.AS); Machine Learning (cs.LG); Sound (cs.SD); Signal Processing (eess.SP) |
| Cite as: | arXiv:2610.00607 [eess.AS] |
| (or arXiv:2610.00607v1 [eess.AS] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00607 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Steven Cho [view email]
[v1]
Wed, 30 Sep 2026 19:11:36 UTC (909 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org