Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Speech Generation on LibriSpeech (test-clean)
Loading...
3.43
MOS
MS-TLM
2.6292
2.8371
3.045
3.2529
Sep 7, 2021
MOS
MOS (Model/Multi-ref)
MOS (Perceptual/Predictive)
MOS (Style similarity)
MOS (Quality)
Updated 4d ago
Evaluation Results
Method
Method
Links
MOS
MOS (Model/Multi-ref)
MOS (Perceptual/Predictive)
MOS (Style similarity)
MOS (Quality)
MS-TLM
ID=12, Input=(u,d,lf),...
2021.09
3.43
4.04
3.75
-
-
MS-TLM
ID=11, Input=(u,d,lf),...
2021.09
3.31
3.76
3.78
-
-
resynthesized
2021.09
3.21
3.95
3.87
-
-
MS-TLM
ID=9, Input=u, Output=...
2021.09
3.16
3.8
3.69
-
-
MS-TLM
ID=10, Input=u, Output...
2021.09
2.66
3.36
3.15
-
-
Feedback
Search any
task
Search any
task