Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
In-context Text-to-Speech on Librispeech clean (test)
Loading...
2
Word Error Rate (WER)
UniAudio
1.808
3.104
4.4
5.696
Dec 25, 2023
Word Error Rate (WER)
Sim-r
Sim-o
Updated 4d ago
Evaluation Results
Method
Method
Links
Word Error Rate (WER)
Sim-r
Sim-o
UniAudio
Training data=100K hours
2023.12
2
0.71
-
NS2
Training data=44K hours
2023.12
2.3
0.62
-
Voicebox
Training data=60K hours
2023.12
2.6
-
0.696
AUDIOBOX SPEECH
Training data=100K hours
2023.12
3.2
0.745
0.734
VALL-E
Training data=60K hours
2023.12
5.9
0.58
-
YourTTS
Training data=600 hours
2023.12
6.8
-
0.435
Feedback
Search any
task
Search any
task
