Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Speech Generation on Accent+
Loading...
0.596
JointCLAP
AUDIOBOX
0.09888
0.22794
0.357
0.48606
Dec 25, 2023
JointCLAP
Sim-σ
WER
QMOS
REL
Speaker Sim MOS
Updated 4d ago
Evaluation Results
Method
Method
Links
JointCLAP
Sim-σ
WER
QMOS
REL
Speaker Sim MOS
AUDIOBOX
Voice cond.=No
2023.12
0.596
0.141
2.6
3.54
3.61
3.03
AUDIOBOX
Voice cond.=Yes
2023.12
0.593
0.344
2.8
3.58
3.57
3.24
ground truth
Voice cond.=n/a
2023.12
0.561
0.526
13.5
3.24
3.51
3.27
VoiceLDM
Voice cond.=avg. CLAP
2023.12
0.204
0.076
3.9
-
-
-
AudioLDM2-SP
Voice cond.=avg. CLAP
2023.12
0.118
0.089
30.2
-
-
-
Feedback
Search any
task
Search any
task