1 paper
Qiangong Zhou, Nagasaka Tomohiro
This work considers merging two independent models, TTS and A2F, into a unified model to enable internal feature transfer, thereby improving the consistency between audio and facia…