Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Text-to-Speech on LRS2 zero-shot
Loading...
8.93
WER
GT
4.8872
32.1761
59.465
86.7539
Jun 2, 2026
WER
Speaker Similarity
UTMOS
MCD-DS
Updated 1mo ago
Evaluation Results
Method
Method
Links
WER
Speaker Similarity
UTMOS
MCD-DS
GT
Type=GT
2026.06
8.93
100
3.14
7.59
Foley-Omni
Type=Text+Video
2026.06
13
64.6
2.67
8.24
LipVoicer
Type=Video
2026.06
22.5
62.1
2.16
9.52
Faces2Voices
Type=Video
2026.06
31.5
65
3.88
8.26
EmoDubber
Type=Text+Video
2026.06
110
10.3
1.7
8.68
Feedback
Search any
task
Search any
task