Conv-TasNet-SNR

Reported on 3 benchmarks across 1 task · 1 paper · 2 SOTA

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

Audio3 results

Speech EnhancementonDeep Noise Suppression (DNS) Challenge
PESQ· 2020-08-20
2.75
SOTA
Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks arXiv:2005.11611
Speech EnhancementonDeep Noise Suppression (DNS) Challenge
ΔPESQ· 2020-08-20
0.93
SOTA
Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks arXiv:2005.11611
Speech EnhancementonDeep Noise Suppression (DNS) Challenge
PESQ-WB· 2020-08-20
2.73
best: 3.81 (ZipEnhancer (M))
Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks arXiv:2005.11611