Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Text-to-Speech on AISHELL-3 (test)
Loading...
4.18
MOS
Recording
4.024
4.0645
4.105
4.1455
Jun 13, 2026
MOS
Preference Rate (System A)
N/P Score (%)
Preference Rate (System B)
P-value
Updated 1mo ago
Evaluation Results
Method
Method
Links
MOS
Preference Rate (System A)
N/P Score (%)
Preference Rate (System B)
P-value
Recording
2026.06
4.18
-
-
-
-
CosyVoice(50k)
Training dataset=50k h...
2026.06
4.06
-
-
-
-
dynamic prosody prediction
Training dataset=50k h...
2026.06
4.06
-
-
-
-
CoT
Training dataset=50k h...
2026.06
4.03
-
-
-
-
Proposed model
Model A=Vevo1.5
2026.06
-
23.2
41.2
35.6
0.01
Proposed model
Model A=F5-TTS
2026.06
-
21.4
50
28.6
0.01
Proposed model
Model A=CosyVoice
2026.06
-
29.1
43.2
27.7
0.65
Feedback
Search any
task
Search any
task