Speech Synthesis on Speech Academic Setting
3.65MOS PredictionBigCodec
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| BigCodec# Param=412M2025.12 | 3.65 | 35.27 | 0.0148 | 82.06 | |
| Pupu-Codec# Param=119M, Size=large2025.12 | 3.61 | 35.15 | 0.0111 | 81.63 | |
| DAC# Param=154M2025.12 | 3.6 | 32.93 | 0.0128 | 79.72 | |
| Pupu-Codec# Param=32M, Size=small2025.12 | 3.54 | 38.83 | 0.0116 | 76.53 | |
| Pupu-Vocoder# Param=122M, Size=large2025.12 | 3.53 | 48.88 | 0.029 | 73.06 | |
| Pupu-Vocoder# Param=14M, Size=small2025.12 | 3.49 | 58.25 | 0.0272 | 70.97 | |
| BigVGAN# Param=122M, Size=large2025.12 | 3.47 | 50.3 | 0.0246 | 67.53 | |
| EnCodec# Param=59M2025.12 | 3.47 | 49.6 | 0.0206 | 66.84 | |
| BigVGAN# Param=14M, Size=small2025.12 | 3.39 | 61.35 | 0.0296 | 61.13 | |
| HiFi-GAN# Param=14M2025.12 | 3.29 | 80.4 | 0.0296 | 55.56 | |
| Vocos# Param=14M2025.12 | 3.22 | 68.67 | 0.0291 | 58.94 |