DeFT-AN

Reported on 6 benchmarks across 1 task · 1 paper · 6 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio6 results

Speech Enhancementonspatialized DNS challenge
PESQ· 2022-12-15
3.01
SOTA
DeFT-AN: Dense Frequency-Time Attentive Network for Multichannel Speech Enhancement arXiv:2212.07570
Speech Enhancementonspatialized DNS challenge
SI-SDR· 2022-12-15
9.9
SOTA
DeFT-AN: Dense Frequency-Time Attentive Network for Multichannel Speech Enhancement arXiv:2212.07570
Speech Enhancementonspatialized DNS challenge
STOI· 2022-12-15
0.924
SOTA
DeFT-AN: Dense Frequency-Time Attentive Network for Multichannel Speech Enhancement arXiv:2212.07570
Speech Enhancementonspatialized WSJCAM0
PESQ· 2022-12-15
3.63
SOTA
DeFT-AN: Dense Frequency-Time Attentive Network for Multichannel Speech Enhancement arXiv:2212.07570
Speech Enhancementonspatialized WSJCAM0
SI-SDR· 2022-12-15
15.7
SOTA
DeFT-AN: Dense Frequency-Time Attentive Network for Multichannel Speech Enhancement arXiv:2212.07570
Speech Enhancementonspatialized WSJCAM0
STOI· 2022-12-15
0.981
SOTA
DeFT-AN: Dense Frequency-Time Attentive Network for Multichannel Speech Enhancement arXiv:2212.07570