Neural Speech Coding on LibriSpeech clean (test)
4.64PESQGround Truth
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Ground Truth# Param=-, # hours=-, frame rate=-, bps=-2026.05 | 4.64 | 100 | 4.09 | 100 | 1.5 | — | — | — | — | — | |
| DualCodecTraining updates=300k2026.06 | 2.954 | — | 3.95 | 0.823 | — | 12.75 | 7.49 | 4.61 | 3.93 | 3.87 | |
| HC-SED-AEDTraining updates=300k2026.06 | 2.894 | — | 3.99 | 0.816 | — | 12.96 | 7.32 | 4.4 | 3.83 | 3.63 | |
| DAC (Distill)Training updates=300k, Process=Distill2026.06 | 2.834 | — | 3.926 | 0.776 | — | 18.01 | 10.55 | 5.34 | 4.82 | 4.14 | |
| DualCodec (open-source)Training updates=300k, Source=open-source2026.06 | 2.833 | — | 3.771 | 0.92 | — | 17.23 | 14.04 | 8.22 | 5.02 | 6.32 | |
| Mimi (open-source)Training updates=300k, Source=open-source2026.06 | 2.589 | — | 3.7745 | 0.8374 | — | 35.77 | 8.58 | 6.32 | 4.15 | 4.14 | |
| FlexiCodec# Param=450M, # hours=54k, frame rate=6.25, bps=6402026.05 | 2.2 | 90 | 4.15 | 71 | 2.57 | — | — | — | — | — | |
| SAC# Param=533M, # hours=20k, frame rate=12.5/25, bps=5252026.05 | 2.16 | 90 | 4.27 | 78 | 2 | — | — | — | — | — | |
| StableCodec-700# Param=950M, # hours=105k, frame rate=25, bps=7002026.05 | 1.92 | 89 | 4.31 | 58 | 3.91 | — | — | — | — | — | |
| StableCodec-400# Param=950M, # hours=105k, frame rate=25, bps=4002026.05 | 1.92 | 89 | 4.31 | 53 | 4.88 | — | — | — | — | — | |
| ClariCodec (w/o RL)# Param=301M, # hours=50k, frame rate=12.5, bps=3002026.05 | 1.88 | 87 | 4.12 | 50 | 4.64 | — | — | — | — | — | |
| ClariCodec# Param=301M, # hours=50k, frame rate=12.5, bps=3002026.05 | 1.87 | 87 | 4.16 | 50 | 3.55 | — | — | — | — | — | |
| WavTokenizer# Param=72M, # hours=8k, frame rate=40, bps=4802026.05 | 1.63 | 85 | 3.57 | 51 | 7.38 | — | — | — | — | — | |
| SemantiCodec# Param=507M, # hours=37.6k, frame rate=12.5/12.5, bps=312.52026.05 | 1.38 | 78 | 2.72 | 34 | 22.7 | — | — | — | — | — | |
| SoCodec# Param=54M, # hours=7.2k, frame rate=8.3, bps=4662026.05 | 1.28 | 9 | 2.5 | 39 | 5.59 | — | — | — | — | — | |
| Encodec# Param=15M, # hours=17.5k, frame rate=10, bps=7502026.05 | 1.25 | 77 | 1.25 | 25 | 16.1 | — | — | — | — | — |