This letter introduces a new synthesis-based defense algorithm for counteracting with a varieties of adversarial attacks developed for challenging the performance of the cutting-edge speech-to-text transcription systems. Our algorithm implements a Sobolev-based GAN and proposes a novel regularizer for effectively controlling over the functionality of the entire generative model, particularly the discriminator network during training. Our achieved results upon carrying out numerous experiments on the victim DeepSpeech, Kaldi, and Lingvo speech transcription systems corroborate the remarkable performance of our defense approach against a comprehensive range of targeted and non-targeted adversarial attacks.

中文翻译：

---

RSD-GAN：针对语音到文本对抗攻击的正则化 Sobolev 防御 GAN

这封信介绍了一种新的基于合成的防御算法，用于对抗为挑战尖端语音到文本转录系统的性能而开发的各种对抗性攻击。我们的算法实现了基于 Sobolev 的 GAN，并提出了一种新颖的正则化器，用于有效控制整个生成模型的功能，特别是训练期间的鉴别器网络。我们在对受害者 DeepSpeech、Kaldi 和 Lingvo 语音转录系统进行大量实验后取得的结果证实了我们的防御方法在抵御全面的目标和非目标对抗性攻击方面的卓越性能。