上海交通大学X-LANCE实验室系统用于MSR挑战2025
The SJTU X-LANCE Lab System for MSR Challenge 2025
AI总结：
X-LANCE实验室提出了一种基于BS-RoFormers的音乐源恢复系统，通过预训练模型和多种训练方案实现8种乐器分离，取得挑战赛多项指标第一名。
摘要：
本报告描述了提交给2025年音乐源恢复（MSR）挑战的系统。我们的方法由一系列顺序BS-RoFormers组成，每个模块处理一个任务，包括音乐源分离（MSS）、去噪和去回声。为了支持任务中给出的8种乐器，我们利用了MSS社区的预训练检查点，并通过几种训练方案微调MSS模型，包括（1）数据集的混合与清洗；（2）音乐片段的随机混合用于数据增强；（3）音频长度的扩展。我们的系统在所有三个主观和三个客观评估指标中均取得第一名，包括MMSNR得分为4.4623和FAD得分为0.1988。我们已开源所有代码和检查点，地址为https://github.com/ModistAndrew/xlance-msr。
英文摘要：
This report describes the system submitted to the music source restoration (MSR) Challenge 2025. Our approach is composed of sequential BS-RoFormers, each dealing with a single task including music source separation (MSS), denoise and dereverb. To support 8 instruments given in the task, we utilize pretrained checkpoints from MSS community and finetune the MSS model with several training schemes, including (1) mixing and cleaning of datasets; (2) random mixture of music pieces for data augmentation; (3) scale-up of audio length. Our system achieved the first rank in all three subjective and three objective evaluation metrics, including an MMSNR score of 4.4623 and an FAD score of 0.1988. We have open-sourced all the code and checkpoints at https://github.com/ModistAndrew/xlance-msr.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
