1 paper · 1 filter
Jiaben Chen, Zixin Wang, Ailing Zeng +8
In this work, we present TalkCuts, a large-scale dataset designed to facilitate the study of multi-shot human speech video generation. Unlike existing datasets that focus on single…