Audio Tokenization on Seed-TTS-Eval EN
3.02PESQ (NB)MiMo-Audio-Tokenizer
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MiMo-Audio-TokenizerkBPS=1.552025.12 | 3.02 | 2.43 | 0.85 | 0.92 | |
| BigCodeckBPS=1.042025.12 | 2.8 | 2.22 | 0.8 | 0.91 | |
| XY-TokenizerkBPS=1.02025.12 | 2.69 | 2.14 | 0.82 | 0.9 | |
| MimikBPS=1.12025.12 | 2.6 | 2.07 | 0.74 | 0.89 | |
| XCodec2.0kBPS=0.82025.12 | 2.57 | 2.01 | 0.78 | 0.89 | |
| Baichuan-Audio-TokenizerkBPS=1.02025.12 | 2.11 | 1.62 | 0.69 | 0.85 | |
| GLM-4-Voice-TokenizerkBPS=0.1752025.12 | 1.11 | 1.05 | 0.12 | 0.6 |