Neural Speech Coding on LibriSpeech (test-other)
1STOIGround Truth
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Ground Truth# Param=-, # hours=-, frame rate=-, bps=-2026.05 | 1 | 4.64 | 3.5 | 1 | 2.81 | |
| FlexiCodec# Param=450M, # hours=54k, frame rate=6.25, bps=6402026.05 | 0.88 | 2.11 | 3.74 | 0.71 | 4.69 | |
| StableCodec-700# Param=950M, # hours=105k, frame rate=25, bps=7002026.05 | 0.87 | 1.91 | 3.91 | 0.58 | 12 | |
| SAC# Param=533M, # hours=20k, frame rate=12.5/25, bps=5252026.05 | 0.87 | 2.04 | 3.9 | 0.77 | 4.15 | |
| StableCodec-400# Param=950M, # hours=105k, frame rate=25, bps=4002026.05 | 0.85 | 1.79 | 3.93 | 0.53 | 14.4 | |
| ClariCodec (w/o RL)# Param=301M, # hours=50k, frame rate=12.5, bps=3002026.05 | 0.84 | 1.75 | 3.67 | 0.51 | 13.3 | |
| ClariCodec# Param=301M, # hours=50k, frame rate=12.5, bps=3002026.05 | 0.84 | 1.75 | 3.73 | 0.51 | 10.4 | |
| WavTokenizer# Param=72M, # hours=8k, frame rate=40, bps=4802026.05 | 0.82 | 1.6 | 3.16 | 0.48 | 21.1 | |
| Encodec# Param=15M, # hours=17.5k, frame rate=10, bps=7502026.05 | 0.76 | 1.27 | 1.26 | 0.25 | 36.4 | |
| SemantiCodec# Param=507M, # hours=37.6k, frame rate=12.5/12.5, bps=312.52026.05 | 0.76 | 1.39 | 2.41 | 0.38 | 40.2 | |
| SoCodec# Param=54M, # hours=7.2k, frame rate=8.3, bps=4662026.05 | 0.09 | 1.33 | 2.33 | 0.46 | 10.6 |