arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.16196eess.AS
收藏

面向隐私保护的自动语音识别(ASR):说话人级别的机器遗忘

Towards Privacy-Preserving ASR: Speaker-Level Machine Unlearning

Seoul National University(首尔大学)

Seaone Ok, Seungu Han, Eungbeom Kim, Kyogu Lee

AI总结:

本研究针对ASR模型中语言与说话人信息纠缠导致的隐私问题,通过分层分析SSL-ASR表示发现说话人身份编码于编码器中间层,提出特征级遗忘策略,在VCTK语料库上可消除特定说话人影响并维持转录性能。

摘要:

自动语音识别(ASR)系统通常被设计为转录语言内容,而不考虑说话人身份。然而,说话人特定信息可能会保留在ASR模型的潜在表示中,当训练数据包含敏感个体时,这会引发隐私问题。虽然机器遗忘已在多项语音任务中得到研究,但由于语音表示中语言信息与说话人相关信息存在强纠缠,将说话人遗忘应用于ASR仍具有挑战性。在本研究中,我们对基于自监督学习(SSL)的ASR表示进行分层分析,发现说话人身份主要编码在编码器中间层。基于该观察,我们探索针对这些身份结构的特征级遗忘策略。在VCTK语料库上的实验表明,所提方法可消除特定说话人的影响,同时保持稳定的转录性能。

英文摘要:

Automatic Speech Recognition (ASR) systems are commonly designed to transcribe linguistic content independently of speaker identity. However, speaker-specific information can persist within the latent representations of ASR models, raising privacy concerns when training data includes sensitive individuals. While machine unlearning has been studied in several speech tasks, applying speaker unlearning to ASR remains challenging due to the strong entanglement of linguistic and speaker-related information in speech representations. In this work, we conduct a layer-wise analysis of Self-Supervised Learning (SSL) based ASR representations, revealing that speaker identity is primarily encoded in intermediate encoder layers. Based on this observation, we explore a feature-level unlearning strategy that targets these identity structures. Experiments on the VCTK corpus show that the proposed approach removes the influence of specific speakers while maintaining robust transcription performance.

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。

↑