Adversarial attacks and defenses in Speaker Recognition Systems: A survey

Jiahe, Lan,; Zhang, Rui; Yan, Zheng; Wang, Jie; Chen, Yu; Hou, Ronghui

doi:10.1016/j.sysarc.2022.102526

Cited by 13 publications

(2 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ever-increasing variety of attacks methods against SRSs necessitates the need for an evaluation criterion. Jiahe Lana et al [4] put forward the same from three aspects: i. Practicability-evaluated by transferability, universality, attack media, distance and commercial SRSs; ii. Imperceptibilityevaluated on the basis of types of adversarial audio, perturbation norm, human perception and signalto-noise ratio; iii.…”

Section: Related Workmentioning

confidence: 99%

Experimental & Comparative Study of Adversarial Attacks on Automatic Speech Recognition Systems

Sharma,

Malik,

Bansal

2024

Advancements in Communication and Systems

View full text Add to dashboard Cite

Automatic Speech Recognition Systems are an integral part of modern-day artificial intelligence-based machines due to advancement in smart technology. Although ASRSs have provided users with the ability to perform important tasks through voice commands, their ability to withstand adversarial attacks is uncertain. The primary objective of adversarial artificial intelligence techniques is to disrupt the functioning of machine learning models with misleading data, by exploiting their vulnerabilities in decision-making process, causing them to make misclassifications or incorrect predictions. This poses a significant threat in the realm of AI and machine learning research, particularly in areas where machine performance is crucial. Therefore, it becomes imperative to study adversarial attacks, with a particular focus on assessing and comparing their severity. This would allow distinguishment between highly detrimental attacks and those with lesser impact, enabling the development of tailored defense strategies to effectively safeguard against them. This paper encompasses an experimental and comparative study of intensity of various attack methods on wav2vec2 model from the Torchaudio hub, particularly: Fast Gradient Sign Method, Basic Iterative Method, Projected Gradient Descent, Carlini and Wagner, and Imperceptible CW attack.

show abstract

Section: Related Workmentioning

confidence: 99%

Experimental & Comparative Study of Adversarial Attacks on Automatic Speech Recognition Systems

Sharma,

Malik,

Bansal

2024

Advancements in Communication and Systems

View full text Add to dashboard Cite

show abstract

“…With the improvement in computer performance and data processing capacity, deep neural networks have demonstrated great advantages in intelligent environments such as image and speech recognition [1], autonomous driving [2], natural language processing [3,4], and network security detection [5]. However, recent studies have shown that deep neural networks are vulnerable to AEs [6][7][8].…”

Section: Introductionmentioning

confidence: 99%

An Intelligent Secure Adversarial Examples Detection Scheme in Heterogeneous Complex Environments

Wang,

Pan

et al. 2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Image-denoising techniques are widely used to defend against Adversarial Examples (AEs). However, denoising alone cannot completely eliminate adversarial perturbations. The remaining perturbations tend to amplify as they propagate through deeper layers of the network, leading to misclassifications. Moreover, image denoising compromises the classification accuracy of original examples. To address these challenges in AE defense through image denoising, this paper proposes a novel AE detection technique. The proposed technique combines multiple traditional image-denoising algorithms and Convolutional Neural Network (CNN) network structures. The used detector model integrates the classification results of different models as the input to the detector and calculates the final output of the detector based on a machine-learning voting algorithm. By analyzing the discrepancy between predictions made by the model on original examples and denoised examples, AEs are detected effectively. This technique reduces computational overhead without modifying the model structure or parameters, effectively avoiding the error amplification caused by denoising. The proposed approach demonstrates excellent detection performance against mainstream AE attacks. Experimental results show outstanding detection performance in well-known AE attacks, including Fast Gradient Sign Method (FGSM), Basic Iteration Method (BIM), DeepFool, and Carlini & Wagner (C&W), achieving a 94% success rate in FGSM detection, while only reducing the accuracy of clean examples by 4%.

show abstract

Enhancing cross-domain transferability of black-box adversarial attacks on speaker recognition systems using linearized backpropagation

Patel,

Bhilare,

Hati

2024

Pattern Anal Applic

View full text Add to dashboard Cite

Adversarial attacks and defenses in Speaker Recognition Systems: A survey

Cited by 13 publications

References 38 publications

Experimental & Comparative Study of Adversarial Attacks on Automatic Speech Recognition Systems

Experimental & Comparative Study of Adversarial Attacks on Automatic Speech Recognition Systems

An Intelligent Secure Adversarial Examples Detection Scheme in Heterogeneous Complex Environments

Enhancing cross-domain transferability of black-box adversarial attacks on speaker recognition systems using linearized backpropagation

Contact Info

Product

Resources

About