Inaudible Adversarial Perturbations For Targeted Attack In Speaker Recognition

Abstract

Speaker recognition is a popular topic in biometric authentication and many deep learning approaches have achieved extraordinary performances. However, it has been shown in both image and speech applications that deep neural networks are vulnerable to adversarial examples. In this study, we aim to exploit this weakness to perform targeted adversarial attacks against the x-vector based speaker recognition system. We propose to generate inaudible adversarial perturbations achieving targeted white-box attacks to speaker recognition system based on the psychoacoustic principle of frequency masking. Specifically, we constrict the perturbation under the masking threshold of original audio, instead of using a common l_p norm to measure the perturbations. Experiments on Aishell-1 corpus show that our approach yields up to 98.5% attack success rate to arbitrary gender speaker targets, while retaining indistinguishable attribute to listeners. Furthermore, we also achieve an effective speaker att

Stats

• citations43
• S2 citations—
• github stars0
• HF likes0
• heat score12.33
• arxiv keywang2020inaudible

Related papers

• Adversarial Attack And Defense Strategies For Deep Speaker Recognition Systems (2020)13.39
• Adversarial Defense For Deep Speaker Recognition Using Hybrid Adversarial Training (2020)9.59
• Impact Of Phonetics On Speaker Identity In Adversarial Voice Attack (2025)0.00
• Foolhd: Fooling Speaker Identification By Highly Imperceptible Adversarial Disturbances (2020)10.07
• Targeted Adversarial Examples For Black Box Audio Systems (2018)15.75
• Hiddenspeaker: Generate Imperceptible Unlearnable Audios For Speaker Verification System (2024)2.26
• Adversarial Attacks Against Automatic Speech Recognition Systems Via Psychoacoustic Hiding (2018)16.45
• Universal Adversarial Perturbations Generative Network For Speaker Recognition (2020)12.33