Picture for Zhongyi Ye

Zhongyi Ye

DiffS2UT: A Semantic Preserving Diffusion Model for Textless Direct Speech-to-Speech Translation

Add code
Oct 26, 2023
Viaarxiv icon