Speech Reconstruction on LibriSpeech other (test)
6.79WERUniAudio-Token
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| UniAudio-TokenFrame Rate=25Hz, BPS=3252026.05 | 6.79 | 4.18 | |
| StableToken#C=1, Frame Rate=25Hz, BPS=3252025.09 | 7.99 | 3.83 | |
| StableTokenFrame Rate=25Hz, BPS=3252026.05 | 7.99 | 3.83 | |
| GLM-4-Voice-Token.#C=1, Frame Rate=12.5Hz, BPS=1752025.09 | 9.33 | 3.99 | |
| GLM-4-Voice-TokenizerFrame Rate=12.5Hz, BPS=1752026.05 | 9.33 | 3.99 | |
| CosyVoice2#C=1, Frame Rate=25Hz, BPS=3252025.09 | 9.68 | 3.25 | |
| CosyVoice2Frame Rate=25Hz, BPS=3252026.05 | 9.68 | 3.25 | |
| WavTokenizerFrame Rate=75Hz, BPS=9002026.05 | 13.09 | 3.09 | |
| S3 Tokenizer#C=1, Frame Rate=25Hz, BPS=3002025.09 | 13.38 | 3.31 |