Preprint
Aug 2026
Xemo-Talker: Unlock Emotions Explicitly for Audio-Driven Talking Portrait Synthesis
Xemo-Talker is proposed, which first learns a neutral speech-to-motion mapping for stable articulation and lip synchronization, and then introduces a lightweight emotion branch guided by less-principal subspace supervision to enhance emotion control.
Chaolong Yang, Yinuo Guo, Kai Yao et al.
· 0 citations