用扩散模型精修分离音源，加速后效果更优。
Improving Music Source Separation with Diffusion and Consistency Refinement
- 在确定性分离器后加扩散模型，逐步去噪提升音源质量。
- 一致性蒸馏使推理仅需一步，速度大幅提升且性能更佳。
- 方法不依赖具体架构，适配多种模型与数据集。
本文提出一种音乐源分离方法，将生成式扩散模型作为确定性分离器的后期精修，通过迭代去噪逐步提升分离音源质量。虽然扩散精修带来可衡量的质量提升，但推理需多步迭代，计算成本较高。为此，我们引入一致性蒸馏，将推理简化为单步，同时保持高质量；使用两步或以上时，蒸馏模型甚至超越原始扩散方法。关键优势在于该方法架构无关：我们在Slakh2100上对自研U-Net分离器和MUSDB18上对当前最优的BS-RoFormer模型均取得领先性能，证明其在不同主干网络上的泛化能力。音频示例见：https://consistency-separation.github.io/。
原文摘要 · Abstract (English)
In this work, we propose an approach to music source separation that uses a generative diffusion model as a last-stage refinement on top of a deterministic separator, progressively enhancing the separated sources through iterative denoising. While the diffusion refinement yields measurable quality gains, it requires iterative steps at inference, increasing computational cost. To speed up the inference process, we apply consistency distillation, reducing inference to a single step while maintaining quality; with two or more steps, the distilled model even surpasses the diffusion-based approach. Crucially, our method is architecture-agnostic: we demonstrate state-of-the-art results when applied to both a custom U-Net-based separator on Slakh2100 and the state-of-the-art BS-RoFormer model on MUSDB18, showing that the refinement generalizes across backbone architectures. Sound examples are available at: https://consistency-separation.github.io/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。