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 2507.04879eess.AScs.LGcs.SD
收藏

自适应瘦身以实现可扩展和高效的语音增强

Adaptive Slimming for Scalable and Efficient Speech Enhancement

University of Illinois at Urbana-Champaign, USA(伊利诺伊大学厄巴纳-香槟分校);Technical University of Denmark, Denmark(丹麦技术大学)

Riccardo Miccini, Minje Kim, Clément Laroche, Luca Pezzarossa, Paris Smaragdis

AI总结:

本文提出动态瘦身技术,使DEMUcs架构在不同利用因子下自适应运行,实现高效语音增强,减少计算量并保持语音质量。

摘要:

语音增强(SE)能够实现稳健的语音识别、实时通信、助听器等应用,其中语音质量至关重要。然而,在资源受限的设备上部署此类系统涉及在性能和计算效率之间选择静态的权衡。在本文中,我们介绍了动态瘦身技术,应用于流行的SE架构DEMUcs,使其具有可扩展性和输入自适应性。瘦身允许模型在不同的利用因子(UF)下运行,每个对应不同的性能/效率权衡,从而有效模拟多个模型大小而无需额外的存储成本。此外,一个路由子网络,与主干网络端到端训练,确定当前输入的最佳UF。因此,系统通过在额外复杂性不必要的时候自适应选择较小的UF来节省资源。我们证明了解决方案在个体UF上是帕累托最优的,确认了动态路由的好处。当训练所提出的动态瘦身模型以平均使用10%的容量进行训练时,我们获得与等效静态25%利用率相同或更好的语音质量,同时将MACs减少29%。

英文摘要:

Speech enhancement (SE) enables robust speech recognition, real-time communication, hearing aids, and other applications where speech quality is crucial. However, deploying such systems on resource-constrained devices involves choosing a static trade-off between performance and computational efficiency. In this paper, we introduce dynamic slimming to DEMUCS, a popular SE architecture, making it scalable and input-adaptive. Slimming lets the model operate at different utilization factors (UF), each corresponding to a different performance/efficiency trade-off, effectively mimicking multiple model sizes without the extra storage costs. In addition, a router subnet, trained end-to-end with the backbone, determines the optimal UF for the current input. Thus, the system saves resources by adaptively selecting smaller UFs when additional complexity is unnecessary. We show that our solution is Pareto-optimal against individual UFs, confirming the benefits of dynamic routing. When training the proposed dynamically-slimmable model to use 10% of its capacity on average, we obtain the same or better speech quality as the equivalent static 25% utilization while reducing MACs by 29%.

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。

↑