Speech Dereverberation on Deep Noise Suppression (DNS) Challenge
| Rank | Model | PESQ | ΔPESQ | Paper | Code | Year |
|---|---|---|---|---|---|---|
| 1 | Conv-TasNet-SNR | 2.75 | 0.93 | Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks | 2020 | |
| 2 | Noisy/unprocessed | 1.82 | – | Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks | 2020 |