Towards Understanding And Mitigating Audio Adversarial Examples For Speaker Recognition

Abstract

Speaker recognition systems (SRSs) have recently been shown to be vulnerable to adversarial attacks, raising significant security concerns. In this work, we systematically investigate transformation and adversarial training based defenses for securing SRSs. According to the characteristic of SRSs, we present 22 diverse transformations and thoroughly evaluate them using 7 recent promising adversarial attacks (4 white-box and 3 black-box) on speaker recognition. With careful regard for best practices in defense evaluations, we analyze the strength of transformations to withstand adaptive attacks. We also evaluate and understand their effectiveness against adaptive attacks when combined with adversarial training. Our study provides lots of useful insights and findings, many of them are new or inconsistent with the conclusions in the image and speech recognition domains, e.g., variable and constant bit rate speech compressions have different performance, and some non-differentiable transfo

Stats

• citations35
• S2 citations—
• github stars0
• HF likes0
• heat score11.67
• arxiv keychen2022towards

Related papers

• Adversarial Attack And Defense Strategies For Deep Speaker Recognition Systems (2020)13.39
• QFA2SR: Query-free Adversarial Transfer Attacks To Speaker Recognition Systems (2023)0.00
• Inaudible Adversarial Perturbations For Targeted Attack In Speaker Recognition (2020)12.33
• Adversarial Defense For Deep Speaker Recognition Using Hybrid Adversarial Training (2020)9.59
• Who Is Real Bob? Adversarial Attacks On Speaker Recognition Systems (2019)16.28
• Targeted Adversarial Examples For Black Box Audio Systems (2018)15.75
• Impact Of Phonetics On Speaker Identity In Adversarial Voice Attack (2025)0.00
• Ghostvec: A New Threat To Speaker Privacy Of End-to-end Speech Recognition System (2023)0.00