arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.14770eess.AS
收藏

语音增强中的测试时间适应 via 马克极化

Test-Time Adaptation For Speech Enhancement Via Mask Polarization

University of Stuttgart(斯图加特大学)

Tobias Raichle, Erfan Amini, Bin Yang

AI总结:

本文提出一种轻量级测试时间适应方法,通过分布比较恢复掩码双模态,提升语音增强在域偏移下的性能。

摘要:

适应语音增强(SE)模型以应对未见环境对实际部署至关重要,但SE的测试时间适应(TTA)仍因缺乏对域偏移下SE模型退化机制的理解而研究不足。我们发现基于掩码的SE模型在域偏移下失去信心,预测的掩码变得平坦,失去决定性的语音保留和噪声抑制。基于此见解,我们提出掩码极化(MPol),一种轻量级TTA方法,通过使用Wasserstein距离进行分布比较来恢复掩码双模态。MPol无需额外参数,适合资源受限的边缘部署。在多样化的域偏移和架构上的实验结果表明,MPol实现了非常一致的增益,其性能与显著更复杂的方案相竞争。

英文摘要:

Adapting speech enhancement (SE) models to unseen environments is crucial for practical deployments, yet test-time adaptation (TTA) for SE remains largely under-explored due to a lack of understanding of how SE models degrade under domain shifts. We observe that mask-based SE models lose confidence under domain shifts, with predicted masks becoming flattened and losing decisive speech preservation and noise suppression. Based on this insight, we propose mask polarization (MPol), a lightweight TTA method that restores mask bimodality through distribution comparison using the Wasserstein distance. MPol requires no additional parameters beyond the trained model, making it suitable for resource-constrained edge deployments. Experimental results across diverse domain shifts and architectures demonstrate that MPol achieves very consistent gains that are competitive with significantly more complex approaches.

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。

↑