为将最坏情况下的隐私泄露作为语音匿名化的隐私评估指标辩护
In Defense of Using Worst-case Privacy Disclosure as Privacy Evaluation Metric of Voice Anonymization
浏览论文内容
中文总结 AI 辅助
本文针对语音匿名化隐私评估指标认知空白，为隐私-ZEBRA框架辩护，阐释EER的局限性，说明基于排名的指标可转化为符合完美保密原则的等价指标，还探讨LLR估计对评估的影响，相关发现经模拟及VoicePrivacy Challenge数据验证。
中文摘要 AI 辅助
语音匿名化领域主要使用等错误率（Equal Error Rate, EER）评估语音身份保护性能，尽管已有隐私-ZEBRA框架、基于排名的指标等替代指标被提出，但其底层假设与差异可能未被充分知晓，尤其是对新手而言，本文旨在填补这一空白。基于香农的完美保密（或隐私）概念，本文为隐私-ZEBRA框架辩护，未提出新指标，而是解释了在EER方面表现“理想”的系统为何无法衡量对数似然比（log-likelihood ratio, LLR）空间中单个说话者的信息泄露；还展示了如何将基于排名的指标转化为符合完美保密原则的指标，且其最优解等价；此外，本文解释了LLR估计方法如何影响评估结果，据作者所知，现有论文未对这些讨论进行详细探索或阐释，最后，在模拟数据与VoicePrivacy Challenge数据上验证了上述发现。
英文摘要
The voice anonymization community mainly uses Equal Error Rate (EER) to evaluate the performance of voice identity protection. While alternative metrics such as privacy-ZEBRA and a rank-based metric have been proposed, their underlying assumptions and differences may not be well known, especially to newcomers. This paper is motivated to fill the gap. Based on the concept of Shannon's perfect secrecy (or privacy), this paper positions itself as a defense of the privacy-ZEBRA framework. While no new metric is proposed, this paper explains how an `ideal' system in terms of EER may fail to gauge the information leakage on individual speakers in the log-likelihood ratio (LLR) space. The paper also shows how the rank-based metric can be cast into a metric that follows the same principle of perfect secrecy and how their best solutions are equivalent. Furthermore, the paper explains how the method of estimating LLRs may affect the evaluation results. These discussions are, to the best of the authors' knowledge, not explored or explained in detail in existing papers. Last but not least, the findings are demonstrated on simulated and VoicePrivacy Challenge data.
