Audio-driven 3D Talking Head Generation on VASA generated synthetic video 1.0 (test) benchmark leaderboard · SOTA2 Research

Audio-driven 3D Talking Head Generation on VASA generated synthetic video 1.0 (test)

8.121ScVASA-3D

Evaluation Results

Method	Links
VASA-3D2025.12		8.121	6.93	0.7874	4.29	93.91
VASA-3D2025.12		7.98	7.002	—	—	—
TalkingGaussian2025.12		6.701	8.1061	0.7971	2.38	0.72
GeneFace2025.12		5.922	9.6066	0.7857	1.73	0.72
ER-NeRF2025.12		5.921	8.7788	0.7732	1.82	1.08
MimicTalk2025.12		5.27	10.9368	0.7748	2.23	3.58