Voice Activity Detection on LibriVAD NonConcat
6.9EERMFCC-ViT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MFCC-ViTTrain data Size=Medium, Train data Type=LibriVAD-NonConcat2025.12 | 6.9 | 0.0348 | |
| MFCC-ViTTrain data Size=Large, Train data Type=LibriVAD-NonConcat2025.12 | 7.1 | 0.0377 | |
| GFCC-ViTTrain data Size=Small, Train data Type=LibriVAD-NonConcat2025.12 | 9.3 | 0.0456 | |
| MFCC-ViTTrain data Size=Small, Train data Type=LibriVAD-Concat2025.12 | 9.48 | 0.0473 | |
| MFCC-ViTTrain data Size=Small, Train data Type=LibriVAD-NonConcat2025.12 | 10.4 | 0.0528 | |
| raw-CLDNNTrain data Size=Small, Train data Type=LibriVAD-NonConcat2025.12 | 19.29 | 0.0756 | |
| MFCC-bDNNTrain data Size=Small, Train data Type=LibriVAD-NonConcat2025.12 | 19.52 | 0.0919 | |
| GFCC-bDNNTrain data Size=Small, Train data Type=LibriVAD-NonConcat2025.12 | 22.13 | 0.0953 |