Figure 9
A diagram shows four radar charts labeled (a) to (d), each comparing Bona fide performance across eight anti-spoofing countermeasures labeled A 0 0 1 through A 0 0 8.

Accuracies achieved by the representative countermeasures in the model generalization evaluation, divided by attack type. Methods: CQT-LCNN, CQT-ResNet, RawSpeech-AASIST, and SSL-AASIST.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal