Zero-shot Text-to-Speech on TTS tasks Zero-shot
3.89UTMOSDiFlow-TTS-Small
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DiFlow-TTS-Small#Params=122M, NFE=16, RTF=0.052025.09 | 3.89 | 5 | 45 | 8.58 | 0.008 | |
| DiFlow-TTS#Params=164M, NFE=16, RTF=0.072025.09 | 3.86 | 5 | 45 | 7.96 | 0.007 | |
| MaskGCT#Params=1.43B, NFE=50 + 45†, RTF=0.462025.09 | 3.83 | 9 | 67 | 14.33 | 0.007 | |
| F5-TTS#Params=336M, NFE=32, RTF=0.262025.09 | 3.72 | 9 | 66 | 12.66 | 0.01 | |
| VALL-E#Params=594M, NFE=-, RTF=0.862025.09 | 3.68 | 19 | 40 | 21.66 | 0.02 | |
| VoiceCraft#Params=830M, NFE=-, RTF=1.702025.09 | 3.55 | 18 | 51 | 17.22 | 0.01 | |
| DiFlow-TTS-Small#Params=122M, NFE=4, RTF=0.032025.09 | 3.34 | 6 | 43 | 8.31 | 0.007 | |
| DiFlow-TTS#Params=164M, NFE=4, RTF=0.032025.09 | 3.31 | 5 | 44 | 8.05 | 0.007 | |
| OZSpeech#Params=145M, NFE=1, RTF=0.032025.09 | 3.15 | 5 | 40 | 11.96 | 0.01 | |
| NaturalSpeech 2#Params=378M, NFE=200, RTF=1.662025.09 | 2.38 | 9 | 31 | 15.62 | 0.02 |