Speech Enhancement on DNS Challenge synthetic No reverb
2.73PESQConv-TasNet (*)
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Conv-TasNet (*)# of params [million]=5.08, Time per frame [ms]=1, Time to infer a frame [ms]=351.9, Loss function=SNR, Reverberant augmentation=true2020.05 | 2.73 | 4.07 | 3.64 | 3.41 | |
| Conv-TasNet# of params [million]=5.08, Time per frame [ms]=1, Time to infer a frame [ms]=355.5, Loss function=PASEMSE, Reverberant augmentation=false2020.05 | 2.72 | 4.05 | 3.65 | 3.39 | |
| STFT-TCN# of params [million]=5.03, Time per frame [ms]=4, Time to infer a frame [ms]=348.1, Loss function=PASEMSE, Reverberant augmentation=true2020.05 | 2.24 | 3.63 | 3.12 | 2.93 | |
| STFT-TCN# of params [million]=5.03, Time per frame [ms]=4, Time to infer a frame [ms]=353.8, Loss function=PCMSE, Reverberant augmentation=false2020.05 | 2.16 | 3.54 | 3.04 | 2.84 | |
| STFT-TCN# of params [million]=5.03, Time per frame [ms]=4, Loss function=PCMSE, Reverberant augmentation=true2020.05 | 1.97 | 3.49 | 2.88 | 2.71 | |
| Baseline (NSNet)# of params [million]=1.26, Time per frame [ms]=10, Loss function=[6]2020.05 | 1.83 | 2.88 | 2.75 | 2.32 | |
| Noisy (Raw)2020.05 | 1.58 | 3.03 | 2.53 | 2.27 | |
| STFT-TCN# of params [million]=5.03, Time per frame [ms]=4, Time to infer a frame [ms]=350.1, Loss function=PASEMSE, Reverberant augmentation=false2020.05 | 1.51 | 2.99 | 2.25 | 2.22 |