Neural Vocoding on EARS (out-of-domain)
3.3UTMOSGT
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| GTPre-trained dataset=LibriTTS2026.03 | 3.3 | — | — | — | — | — | 89.06 | |
| BigVGANPre-trained dataset=LibriTTS, Pre-trained steps=5M2026.03 | 3.19 | 4.236 | 0.064 | 96.3 | 17.655 | 4.946 | 80.8 | |
| RNDVoC-sharedPre-trained dataset=LibriTTS2026.03 | 3.11 | 3.989 | 0.073 | 95.8 | 17.725 | 4.869 | 81.51 | |
| RNDVoC-nonsharedPre-trained dataset=LibriTTS2026.03 | 3.078 | 3.856 | 0.083 | 94.9 | 18.946 | 4.821 | 79.88 | |
| BigVGAN-basePre-trained dataset=LibriTTS, Pre-trained steps=5M2026.03 | 3.04 | 3.797 | 0.087 | 94.6 | 26.05 | 4.871 | 78.19 | |
| VocosPre-trained dataset=LibriTTS2026.03 | 3.007 | 3.522 | 0.094 | 93.7 | 25.904 | 4.803 | 74.1 | |
| HiFiGAN-V1Pre-trained dataset=LibriTTS2026.03 | 2.835 | 2.907 | 0.149 | 87.6 | 48.417 | 4.57 | 70.82 | |
| AvocodoPre-trained dataset=LibriTTS2026.03 | 2.809 | 3.065 | 0.149 | 86.6 | 48.494 | 4.641 | 69.15 | |
| APNet2Pre-trained dataset=LibriTTS2026.03 | 2.259 | 2.45 | 0.149 | 86.2 | 39.627 | 4.371 | 55.08 |