Spoof detection on ASVspoof PA 2019 (eval)
0.46EER (%)CQT+SE-Res2Net50+CE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CQT+SE-Res2Net50+CEInput feature=CQT, Classifier=SE-Res2Net50, Loss function=CE2021.09 | 0.46 | 0.0116 | |
| Fbank&CQT+ResNeWt+CEInput feature=Fbank&CQT, Classifier=ResNeWt, Loss function=CE2021.09 | 0.52 | 0.0134 | |
| Spect+SE-Res2Net50+CEInput feature=Spect, Classifier=SE-Res2Net50, Loss function=CE2021.09 | 0.74 | 0.0207 | |
| logPowSpec+EABNet+CombLossInput feature=logPowSpec, Classifier=EABNet, Loss function=CombLoss2021.09 | 0.86 | 0.0239 | |
| Spect+LCGRNN+tripletInput feature=Spect, Classifier=LCGRNN, Loss function=triplet2021.09 | 0.92 | 0.0198 | |
| CQTMGD+ResNeWt+CEInput feature=CQTMGD, Classifier=ResNeWt, Loss function=CE2021.09 | 0.94 | 0.025 | |
| Spect+LCGRNN+GKDE-softmaxInput feature=Spect, Classifier=LCGRNN, Loss function=GKDE-softmax2021.09 | 1.06 | 0.0222 | |
| Joint-gram+ResNet+CEInput feature=Joint-gram, Classifier=ResNet, Loss function=CE2021.09 | 1.23 | 0.0305 | |
| Spect+ResNet+CE [20]Input feature=Spect, Classifier=ResNet, Loss function=CE2021.09 | 1.29 | 0.036 | |
| LFCC-CMR2026.07 | 1.34 | 0.0315 | |
| LFCC+SE-Res2Net50+CEInput feature=LFCC, Classifier=SE-Res2Net50, Loss function=CE2021.09 | 1.46 | 0.434 | |
| MOSAIC2026.07 | 1.98 | 0.2197 | |
| logCQT&powSpect+VGG+CEInput feature=logCQT&powSpect, Classifier=VGG, Loss function=CE2021.09 | 2.11 | 0.527 | |
| Spect+ResNet+CE [29]Input feature=Spect, Classifier=ResNet, Loss function=CE2021.09 | 3.81 | 0.994 | |
| LFCC+LCNN+A-softmaxInput feature=LFCC, Classifier=LCNN, Loss function=A-softmax2021.09 | 4.6 | 0.1053 | |
| CQCC+GMM+EMInput feature=CQCC, Classifier=GMM, Loss function=EM2021.09 | 11.04 | 0.2454 | |
| LFCC+GMM+EMInput feature=LFCC, Classifier=GMM, Loss function=EM2021.09 | 13.54 | 0.3017 |