Deepfake Speech Detection on ASVspoof
9.04EERQV-CNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| QV-CNNFeature=MFCC2026.04 | 9.04 | 94.2 | |
| MFCC + ResNetStudy=[35]2026.04 | 9.33 | — | |
| Spectrogram + CNNStudy=[34]2026.04 | 9.57 | — | |
| CQCC + GMM (B1)Study=[40]2026.04 | 9.57 | — | |
| STM (Mel FB) + LCNNStudy=[28]2026.04 | 9.79 | — | |
| QV-CNNFeature=Mel Spectrogram2026.04 | 10.84 | 94.57 | |
| Spectrogram + Vision TransformerStudy=[29]2026.04 | 11.02 | — | |
| CQCC + DNNStudy=[32]2026.04 | 13.74 | — | |
| Benford Distribution + TransformerStudy=[41]2026.04 | 16.37 | — | |
| Log-Spectrogram + Formant TransformerStudy=[42]2026.04 | 17.51 | — | |
| Compact Convolutional Transformer (CCT)Study=[30]2026.04 | — | 92.13 | |
| MFCC + LSTM-CNNStudy=[25]2026.04 | — | 88 | |
| Spectrogram + CNNStudy=[37]2026.04 | — | 85.99 | |
| STFT + VGG-TransformerStudy=[43]2026.04 | — | 93.36 |