List your company
ResearchFormerly WizWand

Audio Reconstruction on LibriSpeech (other)

5.49WERX-Codec
5.06727.921110.77513.6289Sep 26, 2025
Updated 6mo ago

Evaluation results

MethodLinks
X-Codec#C=8, Frame Rate=50Hz, BPS=4000, Tokenizer Type=Semantic Distilled2025.095.493.19
X-Codec#C=3, Frame Rate=50Hz, BPS=1500, Tokenizer Type=Semantic Distilled2025.096.113.17
SpeechTokenizer#C=8, Frame Rate=50Hz, BPS=4000, Tokenizer Type=Semantic Distilled2025.096.583.1
StableToken#C=1, Frame Rate=25Hz, BPS=325, Tokenizer Type=Supervised Semantic2025.097.993.83
X-Codec#C=1, Frame Rate=50Hz, BPS=500, Tokenizer Type=Semantic Distilled2025.099.023.04
GLM-4-Voice-Token#C=1, Frame Rate=12.5Hz, BPS=175, Tokenizer Type=Supervised Semantic2025.099.333.99
CosyVoice2#C=1, Frame Rate=25Hz, BPS=325, Tokenizer Type=Supervised Semantic2025.099.683.25
Mimi#C=8, Frame Rate=12.5Hz, BPS=1100, Tokenizer Type=Semantic Distilled2025.099.843.06
SpeechTokenizer#C=3, Frame Rate=50Hz, BPS=1500, Tokenizer Type=Semantic Distilled2025.0910.722.89
S3 Tokenizer#C=1, Frame Rate=25Hz, BPS=300, Tokenizer Type=Supervised Semantic2025.0913.383.31
SpeechTokenizer#C=1, Frame Rate=50Hz, BPS=500, Tokenizer Type=Semantic Distilled2025.0916.062.49