2024
Generating High-Quality Adversarial Examples with Universal Perturbation-Based Adaptive Network and Improved Perceptual Loss
ICASSP 2024accepted
Deep neural network-based speaker identification systems are vulnerable to adversarial attacks. However, the distortions of the adversarial examples are still obvious in most cases. In this work, we therefore propose a universal perturbation-based adaptive network (UPAN) to generate high-quality adv…