Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Text-to-Speech Synthesis on LJSpeech (MOS, WER)
Loading...
4.48
MOS
FNH-TTS
3.7208
3.9179
4.115
4.3121
Aug 16, 2025
MOS
WER
Updated 5d ago
Evaluation Results
Method
Method
Links
MOS
WER
FNH-TTS
Model Size=47.73M, Dis...
2025.08
4.48
2.59
VITS Origin w/ CMoBD + SBD + VOCOS
Model Size=40.89M, Dis...
2025.08
4.44
2.6
StyleTTS2
Model Size=145.53M, Di...
2025.08
4.35
5.78
VITS Origin w/ CMoBD + SBD
Model Size=39.53M, Dis...
2025.08
4.32
2.69
VITS Origin
Model Size=39.53M, Dis...
2025.08
4.26
3.41
VITS w/ MoE-DP + CMoBD + SBD
Model Size=46.37M, Dis...
2025.08
4.2
2.42
FastSpeech2
Model Size=34.64M, Dis...
2025.08
4.12
7.02
SparkTTS
Model Size=506.63M
2025.08
4.1
7.36
VITS w/ MoE-DP
Model Size=46.37M, Dis...
2025.08
3.92
6.48
F5-TTS
Model Size=337.09M
2025.08
3.87
8.7
VITS w/ MoE-DP + VOCOS
Model Size=47.73M, Dis...
2025.08
3.75
9.74
Feedback
Search any
task
Search any
task