Audio-driven Talking Face Generation on User Study benchmark leaderboard · SOTA2 Research

Audio-driven Talking Face Generation on User Study

4.6Lip-sync AccuracyPC-Talk

Evaluation Results

Method	Links
PC-Talk2025.03		4.6	4.8	4.6	4.9	4.8
Hallo-v22025.03		3.8	4	3.4	—	3.8
Wav2Lip2025.03		3.3	1.9	1.9	—	1.9
ED-Talk2025.03		3.1	1.4	2.4	2.3	1.6
MuseTalk2025.03		2.2	2.6	3	—	3.4
Echomimic2025.03		1.9	3.4	3.5	—	3
EAT2025.03		1.3	2.9	2.6	2.6	2.1