论文标题
在扬声器验证系统中检测自适应对抗攻击
On the Detection of Adaptive Adversarial Attacks in Speaker Verification Systems
论文作者
论文摘要
扬声器验证系统已被广泛用于智能手机和物联网设备以识别合法用户。在最近的工作中，已经表明，诸如FakeBob之类的对抗性攻击可以有效地针对说话者验证系统。本文的目的是设计一个可以将原始音频与受对抗攻击污染的音频区分开的检测器。具体而言，我们设计的检测器（称为MEH-Fest）从音频的短时傅立叶变换中计算出高频的最小能量，并将其用作检测度量。通过分析和实验，我们表明我们提出的检测器易于实现，快速处理输入音频，并且有效地确定音频是否被假屁股攻击损坏。实验结果表明，检测器非常有效：在高斯混合模型（GMM）和I-vector Speaker验证系统中检测假雄性攻击的较接近零的假阳性和假阴性率。此外，讨论和研究了对我们提议的探测器的自适应对抗性攻击，并研究了他们的对策，并显示了攻击者和后卫之间的比赛。
Speaker verification systems have been widely used in smart phones and Internet of things devices to identify legitimate users. In recent work, it has been shown that adversarial attacks, such as FAKEBOB, can work effectively against speaker verification systems. The goal of this paper is to design a detector that can distinguish an original audio from an audio contaminated by adversarial attacks. Specifically, our designed detector, called MEH-FEST, calculates the minimum energy in high frequencies from the short-time Fourier transform of an audio and uses it as a detection metric. Through both analysis and experiments, we show that our proposed detector is easy to implement, fast to process an input audio, and effective in determining whether an audio is corrupted by FAKEBOB attacks. The experimental results indicate that the detector is extremely effective: with near zero false positive and false negative rates for detecting FAKEBOB attacks in Gaussian mixture model (GMM) and i-vector speaker verification systems. Moreover, adaptive adversarial attacks against our proposed detector and their countermeasures are discussed and studied, showing the game between attackers and defenders.
