List your company
ResearchFormerly WizWand

Speech Reconstruction on Chinese speech

2.99UTMOSSAC
1.171.64252.1152.5875Jan 14, 2026
Updated 8mo ago

Evaluation results

MethodLinks
SACfs=12.5 Hz, fa=50 Hz, BitRate (kbps)=0.8752026.012.990.015387—
DualCodecsemantic_rate=12.5 Hz, Layers=6-layer, BitRate (kbps)=0.9252026.012.950.016280—
DSA-Tokenizerfs=25 Hz, fa=50 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=1.12026.012.92—703.49
WavTokenizerToken rate=75 Hz, BitRate (kbps)=0.92026.012.820.050862—
DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^7, Stacked_Layers=2, BitRate (kbps)=12026.012.81—705.27
DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=2, BitRate (kbps)=1.12026.012.81—705.02
DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=0.72026.012.74—673.77
DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=2, BitRate (kbps)=1.12026.012.680.019382—
DSA-Tokenizerfs=25 Hz, fa=50 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=1.12026.012.660.018881—
DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^7, Stacked_Layers=2, BitRate (kbps)=12026.012.650.018981—
DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=0.72026.012.60.017878—
MimiLayers=8-layer, BitRate (kbps)=1.12026.012.370.029373—
SpeechTokenizerLayers=2-layer, BitRate (kbps)=12026.011.750.243933—
DualCodecsemantic_rate=12.5 Hz, Layers=6-layer, BitRate (kbps)=0.9252026.011.73—3216.22
MimiLayers=8-layer, BitRate (kbps)=1.12026.011.54—53100.01
EncodecLayers=2-layer, BitRate (kbps)=1.52026.011.350.051962—
SpeechTokenizerLayers=2-layer, BitRate (kbps)=12026.011.34—15123.33
SACfs=12.5 Hz, fa=50 Hz, BitRate (kbps)=0.8752026.011.27—3071.87
EncodecLayers=2-layer, BitRate (kbps)=1.52026.011.24—1368.97