Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Speech Generation on Expr
Loading...
0.548
JointCLAP
ground truth
0.04776
0.17763
0.3075
0.43737
Dec 25, 2023
JointCLAP
Sim-σ
WER
Quality MOS
REL
Speaker Similarity MOS
Updated 4d ago
Evaluation Results
Method
Method
Links
JointCLAP
Sim-σ
WER
Quality MOS
REL
Speaker Similarity MOS
ground truth
Voice cond.=n/a
2023.12
0.548
0.395
5.8
4
4.01
3.38
AUDIOBOX
Voice cond.=Yes
2023.12
0.48
0.377
7.7
3.86
3.99
3.36
AUDIOBOX
Voice cond.=No
2023.12
0.387
0.181
4.5
3.82
3.94
3.02
VoiceLDM
Voice cond.=avg. CLAP
2023.12
0.093
0.115
4.8
-
-
-
AudioLDM2-SP
Voice cond.=avg. CLAP
2023.12
0.067
0.045
34.6
-
-
-
Feedback
Search any
task
Search any
task