arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.08831eess.AScs.CLcs.MMcs.SD

词错误率够用吗?重新思考语音隐私评估中的实体感知指标

Is Word Error Rate Enough? Rethinking Privacy Evaluation in Speech with Entity-Aware Metrics

Anjana Rajasekhar, Jule Pohlhausen, Nayana Jacob Alappattu, Anna Leschanowsky

首次发表
浏览论文内容

中文总结 AI 辅助

本研究将实体感知隐私指标从自然语言处理适配到语音隐私领域,评估两种混淆技术对命名实体的保护效果,并发现实体丰富数据微调对不同实体类别的攻击性能影响不一,最后给出基于时间对齐的指标选择建议。

中文摘要 AI 辅助

随着智能设备使用量的持续增加,它们捕获敏感语音内容的潜力引发了日益增长的隐私担忧。因此,开发既能防止信息泄露又能保留音频效用的技术,以及能够准确量化隐私水平而不高估它的评估指标,变得至关重要。在这项工作中,我们通过将自然语言处理领域的实体感知隐私指标适配到语音隐私领域,评估了两种混淆技术在保护语音内容(尤其是命名实体)方面的有效性。此外,我们研究了多种攻击场景,并表明在富含实体的数据上进行微调能提高某些实体类别的攻击性能,但对其他类别则不然。最后,我们根据混淆方法是否保留时间对齐,提供了指标选择的指导。

英文摘要

As the use of smart devices continues to increase, their potential to capture sensitive speech content raises growing privacy concerns. It is therefore critical to develop techniques that prevent information leakage while preserving the utility of the audio, and evaluation metrics that accurately quantify the level of privacy without overestimating it. In this work, we evaluate the effectiveness of two obfuscation techniques in protecting speech content, with particular emphasis on named entities, by adapting entity-aware privacy metrics from the Natural Language Processing field to the speech privacy domain. Further, we investigate several attack scenarios and show that fine-tuning on entity-rich data improves attack performance for some entity categories but not others. Finally, we provide guidance on metric selection based on whether the obfuscation method preserves temporal alignment.

发表机构

  • Fraunhofer Institute for Integrated Circuits (IIS)(弗劳恩霍夫集成电路研究所 (IIS))
  • Jade University of Applied Sciences(亚德应用科学大学)
  • Carl von Ossietzky Universität Oldenburg(奥尔登堡卡尔·冯·奥西茨基大学)

机构由 AI 辅助整理,请以论文原文为准。

↑