arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.09286eess.AScs.SD
收藏

FlowAVSE:基于条件流匹配的高效音视频语音增强

FlowAVSE: Efficient Audio-Visual Speech Enhancement with Conditional Flow Matching

Korea Advanced Institute of Science and Technology(韩国科学技术院)

Chaeyoung Jung, Suyeon Lee, Ji-Hoon Kim, Joon Son Chung

AI总结:

本文提出FlowAVSE,通过条件流匹配算法和优化U-net架构,在保持语音增强质量的同时实现22倍推理加速和模型减半。

摘要:

这项工作提出了一种利用声学和视觉线索来增强受损语音信号质量的高效方法。虽然现有的基于扩散的方法已展现出卓越的质量,但其适用性受到推理速度慢和计算复杂度高的限制。为了解决这一问题,我们提出了FlowAVSE,它在不降低输出质量的前提下提高了推理速度并减少了可学习参数的数量。具体而言,我们采用了一种条件流匹配算法,使得能够在单次采样步骤中生成高质量语音。此外,我们通过优化基于扩散系统的底层U-net架构来提高效率。我们的实验表明,FlowAVSE在保持输出质量的同时,实现了22倍的推理速度提升,并将模型大小减少了一半。演示页面可在以下地址获取:https://cyongong.github.io/FlowAVSE.github.io/

英文摘要:

This work proposes an efficient method to enhance the quality of corrupted speech signals by leveraging both acoustic and visual cues. While existing diffusion-based approaches have demonstrated remarkable quality, their applicability is limited by slow inference speeds and computational complexity. To address this issue, we present FlowAVSE which enhances the inference speed and reduces the number of learnable parameters without degrading the output quality. In particular, we employ a conditional flow matching algorithm that enables the generation of high-quality speech in a single sampling step. Moreover, we increase efficiency by optimizing the underlying U-net architecture of diffusion-based systems. Our experiments demonstrate that FlowAVSE achieves 22 times faster inference speed and reduces the model size by half while maintaining the output quality. The demo page is available at: https://cyongong.github.io/FlowAVSE.github.io/

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。

↑