MDD：一种掩码扩散检测器，用于保护说话人验证系统免受对抗性扰动
MDD: a Mask Diffusion Detector to Protect Speaker Verification Systems from Adversarial Perturbations
AI总结：
本文提出掩码扩散检测器（MDD），一种基于文本条件掩码扩散模型的对抗性检测与净化框架，无需对抗样本或大规模预训练，有效提升说话人验证系统的安全性与鲁棒性。
摘要：
说话人验证系统越来越多地部署在安全敏感的应用中，但仍然极易受到对抗性扰动的影响。在这项工作中，我们提出了掩码扩散检测器（MDD），一种基于文本条件掩码扩散模型的新型对抗性检测与净化框架。在训练期间，MDD对梅尔频谱图应用部分掩码，并通过前向扩散过程逐步添加噪声，模拟干净语音特征的退化。随后，一个反向过程根据输入转录重建干净表示。与先前方法不同，MDD不需要对抗性示例或大规模预训练。实验结果表明，MDD实现了强大的对抗性检测性能，并优于先前最先进的方法，包括基于扩散和基于神经编解码器的方法。此外，MDD有效净化了被对抗性操纵的语音，将说话人验证性能恢复到接近干净条件下观察到的水平。这些发现证明了基于扩散的掩码策略在实现安全可靠的说话人验证系统方面的潜力。
英文摘要：
Speaker verification systems are increasingly deployed in security-sensitive applications but remain highly vulnerable to adversarial perturbations. In this work, we propose the Mask Diffusion Detector (MDD), a novel adversarial detection and purification framework based on a \textit{text-conditioned masked diffusion model}. During training, MDD applies partial masking to Mel-spectrograms and progressively adds noise through a forward diffusion process, simulating the degradation of clean speech features. A reverse process then reconstructs the clean representation conditioned on the input transcription. Unlike prior approaches, MDD does not require adversarial examples or large-scale pretraining. Experimental results show that MDD achieves strong adversarial detection performance and outperforms prior state-of-the-art methods, including both diffusion-based and neural codec-based approaches. Furthermore, MDD effectively purifies adversarially-manipulated speech, restoring speaker verification performance to levels close to those observed under clean conditions. These findings demonstrate the potential of diffusion-based masking strategies for secure and reliable speaker verification systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
