Fetching the paper…

DiffTalk: Crafting Diffusion Models for Generalized Audio-Driven Portraits Animation · Around