Speech Quality Evaluation on Common Voice 17
2.55Quality Score (NL)Neucodec
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| NeucodecCodebook=65536, Nq=1, Rate (Hz)=502026.01 | 2.55 | 2.361 | 2.354 | 2.494 | 2.493 | 2.491 | 2.504 | |
| LongCatCodebook=8192, Nq=4, Rate (Hz)=16.62026.01 | 2.5 | 2.341 | 2.226 | 2.364 | 2.424 | 2.396 | 2.326 | |
| Qwen3-TTS-TokenizerCodebook=4096, Nq=16, Rate (Hz)=122026.01 | 2.462 | 2.236 | 2.273 | 2.349 | 2.405 | 2.355 | 2.346 | |
| X-Codec-2.0 (25 Hz, 24 kHz)Codebook=65536, Nq=1, Rate (Hz)=252026.01 | 2.457 | 2.245 | 2.217 | 2.304 | 2.392 | 2.386 | 2.314 | |
| DistilCodecCodebook=32768, Nq=1, Rate (Hz)=932026.01 | 2.416 | 2.13 | 2.141 | 2.261 | 2.312 | 2.308 | 2.194 | |
| MimiCodebook=2048, Nq=32, Rate (Hz)=12.52026.01 | 2.336 | 2.032 | 2.081 | 2.152 | 2.296 | 2.273 | 2.201 | |
| WavTokenizer 40TPSCodebook=4096, Nq=1, Rate (Hz)=402026.01 | 2.319 | 2.15 | 2.107 | 2.189 | 2.23 | 2.241 | 2.231 | |
| DACCodebook=1024, Nq=9, Rate (Hz)=862026.01 | 2.315 | 2.037 | 2.085 | 2.2 | 2.258 | 2.231 | 2.204 | |
| UniCodecCodebook=16384, Nq=1, Rate (Hz)=752026.01 | 2.256 | 2.014 | 2.026 | 2.149 | 2.174 | 2.221 | 2.111 | |
| WavTokenizer 75TPSCodebook=4096, Nq=1, Rate (Hz)=752026.01 | 2.209 | 2.087 | 2.06 | 2.201 | 2.186 | 2.223 | 2.202 | |
| SpeechTokenizerCodebook=1024, Nq=8, Rate (Hz)=502026.01 | 2.203 | 2.065 | 2.003 | 2.062 | 2.159 | 2.202 | 2.096 | |
| SNACCodebook=4096, Nq=3, Rate (Hz)=272026.01 | 2.193 | 2.048 | 2.007 | 2.098 | 2.144 | 2.187 | 2.117 | |
| X-Codec-2.0 (baseline)Codebook=65536, Nq=1, Rate (Hz)=502026.01 | 2.168 | 2.086 | 2.062 | 2.108 | 2.122 | 2.136 | 2.139 | |
| EncodecCodebook=1024, Nq=2, Rate (Hz)=752026.01 | 1.388 | 1.464 | 1.386 | 1.414 | 1.424 | 1.405 | 1.369 |