Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity

Open paper

Benchmark	Model	Rank	Results
gesture-generation-on-beat	Trimodal	#2	FID: 177.2
gesture-generation-on-beat2	Trimodal	#9	FGD: 1.241
gesture-generation-on-ted-gesture-dataset	Trimodal	#5	FGD: 3.729