Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Video-to-Speech Generation on CinePile OOD film data
Loading...
58.7
WER
HiCoDiT
57.516
65.508
73.5
81.492
Apr 17, 2026
WER
MCD
DNSMOS
Emotion Matching Score
Speaker Identity Score
LSE-D
Updated 1mo ago
Evaluation Results
Method
Method
Links
WER
MCD
DNSMOS
Emotion Matching Score
Speaker Identity Score
LSE-D
HiCoDiT
2026.04
58.7
9.8
3.5
82
50.1
7.6
AlignDiT
2026.04
80.8
11.4
3.2
75.2
58.5
8.23
EmoDubber
2026.04
88.3
9.9
2.8
76.5
45.1
7.72
Feedback
Search any
task
Search any
task
