Can I read Transferable Attacks on Audio Deepfake Detection on EtoBox?

Transferable Attacks on Audio Deepfake Detection by kh-23 alrayiany is a document available to read on EtoBox.

What is Transferable Attacks on Audio Deepfake Detection about?

This paper presents a novel transferable adversarial attack framework targeting audio deepfake detection (ADD) systems, highlighting their vulnerabilities against such attacks. The proposed method utilizes a GAN-based approach that preserves transcription and perceptual integrity, significantly reducing the accuracy of state-of-the-art ADD systems in various attack scenarios. Experimental results demonstrate the need for enhanced robustness in ADD systems to counter advanced adversarial threats effectively.

Author

kh-23 alrayiany

Language

EN