Neuromorphic Speech Enhancement with Dual-Branch Spiking Neural Networks
Online Supplement
Dataset
The VCTK+DEMAND dataset is used for demo.
Audio samples of the test set we processed are available at the repository (VCTK).
Compared Methods
| Method | Reference |
|---|---|
| DPSNN | Dual-Path Spiking Neural Network |
| Spiking-FSN | Spiking Full-Subband Network |
Audio Samples
| Model\id(noise) | p257_006(cafe) | p257_073(living) | p257_286(bus) | p232_227(office) | p232_378(psquare) |
|---|---|---|---|---|---|
| Clean | |||||
| Noisy | |||||
| DPSNN | |||||
| Spiking-FSN | |||||
| GSU-DPNet(ours) |
Spectrogram comparison (p257_073, living noise)
Version: v0.1.1