Attacking deep learning based biometric systems has drawn more and more attention with the wide deployment of fingerprint/face/speaker recognition systems, given the fact that the neural networks are vulnerable to the adversarial examples, which have been intentionally perturbed to remain almost imperceptible for human. In this paper, we demonstrated the existence of the universal adversarial perturbations~(UAPs) for the speaker recognition systems. We proposed a generative network to learn the mapping from the low-dimensional normal distribution to the UAPs subspace, then synthesize the UAPs to perturbe any input signals to spoof the well-trained speaker recognition model with high probability. Experimental results on TIMIT and LibriSpeech datasets demonstrate the effectiveness of our model.

中文翻译：

---

用于说话人识别的通用对抗性扰动生成网络

随着指纹/人脸/说话者识别系统的广泛部署，攻击基于深度学习的生物识别系统越来越受到关注，因为神经网络容易受到对抗性示例的影响，这些对抗性示例被故意扰乱以保持人类几乎察觉不到. 在本文中，我们证明了说话人识别系统的普遍对抗性扰动（UAP）的存在。我们提出了一个生成网络来学习从低维正态分布到 UAP 子空间的映射，然后合成 UAP 来扰乱任何输入信号，以高概率欺骗训练有素的说话人识别模型。在 TIMIT 和 LibriSpeech 数据集上的实验结果证明了我们模型的有效性。