mlsanityData to 7 Oct 2026

The Impact of Speech Anonymization on Pathology and Its Limits

arXiv 2404.08064 ↗First posted 11 Apr 2024eess.AS cs.AI cs.CRcited by 11

Not on the map. Why?

Abstract (arXiv)

Integration of speech into healthcare has intensified privacy concerns due to its potential as a non-invasive biomarker containing individual biometric information. In response, speaker anonymization aims to conceal personally identifiable information while retaining crucial linguistic content. However, the application of anonymization techniques to pathological speech, a critical area where privacy is especially vital, has not been extensively examined. This study investigates anonymization's impact on pathological speech across over 2,700 speakers from multiple German institutions, focusing on privacy, pathological utility, and demographic fairness. We explore both deep-learning-based and signal processing-based anonymization methods. We document substantial privacy improvements across disorders-evidenced by equal error rate increases up to 1933%, with minimal overall impact on utility. Specific disorders such as Dysarthria, Dysphonia, and Cleft Lip and Palate experience minimal utility changes, while Dysglossia shows slight improvements. Our findings underscore that the impact of anonymization varies substantially across different disorders. This necessitates disorder-specific anonymization strategies to optimally balance privacy with diagnostic utility. Additionally, our fairness analysis reveals consistent anonymization effects across most of the demographics. This study demonstrates the effectiveness of anonymization in pathological speech for enhancing privacy, while also highlighting the importance of customized and disorder-specific approaches to account for inversion attacks.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 24 of its 82 references that are in the map. First posted on arXiv 11 Apr 2024.

Builds on

24 of 82 references.

References · 24

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2016Deep Learning with Differential Privacycited by 2,682
  3. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  4. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,139
  5. 2019LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speechcited by 639
  6. 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
  7. 2017Generalized End-to-End Loss for Speaker Verificationcited by 290
  8. 2020Introducing the VoicePrivacy Initiativecited by 48
  9. 2021The VoicePrivacy 2020 Challenge: Results and findingscited by 45
  10. 2019Speaker Anonymization Using X-vector and Neural Waveform Modelscited by 41
  11. 2020Speaker anonymisation using the McAdams coefficientcited by 38
  12. 2019Evaluating Voice Conversion-based Privacy Protection against Informed Attackerscited by 30
12 more
  1. 2020Design Choices for X-vector Based Speaker Anonymizationcited by 27
  2. 2022Anonymizing Speech with Generative Adversarial Networks to Preserve Speaker Privacycited by 20
  3. 2022The VoicePrivacy 2022 Challenge Evaluation Plancited by 18
  4. 2023Private, fair and accurate: Training large-scale, privacy-preserving AI models in medical imagingcited by 17
  5. 2020AudVowelConsNet: A Phoneme-Level Based Deep CNN Architecture for Clinical Depression Diagnosiscited by 14
  6. 2023Federated learning for secure development of AI models for Parkinson's disease detection using speech from different languagescited by 9
  7. 2019Automated speech-based screening of depression using deep convolutional neural networkscited by 7
  8. 2023Preserving privacy in domain transfer of medical AI models comes at no performance costs: The integral role of differential privacycited by 6
  9. 2022The effect of speech pathology on automatic speaker verification -- a large-scale studycited by 5
  10. 2021On the invertibility of a voice privacy system using embedding alignementcited by 4
  11. 2020An Empirical Study on Text-Independent Speaker Verification based on the GE2E Methodcited by 3
  12. 2022Self-Supervised Speech Representations Preserve Speech Characteristics while Anonymizing Voicescited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 5; 2026: 4.

11 papers.

  1. 2026The Third VoicePrivacy Challenge: Preserving Emotional Expressiveness and Linguistic Content in Voice Anonymizationcited by 8
  2. 2024A Benchmark for Multi-speaker Anonymizationcited by 7
  3. 2025Differential privacy for medical deep learning: methods, tradeoffs, and deployment implicationscited by 7
  4. 2025Children's Voice Privacy: First Steps And Emerging Challengescited by 5
  5. 2024Differential privacy enables fair and accurate AI-based analysis of speech disorders while protecting patient datacited by 4
  6. 2026The Voice of Equity: A Systematic Evaluation of Bias Mitigation Techniques for Speech-Based Cognitive Impairment Detection Across Architectures and Demographicscited by 2
  7. 2025You Are What You Say: Exploiting Linguistic Content for VoicePrivacy Attackscited by 1
  8. 2025VoxGuard: Evaluating User and Attribute Privacy in Speech via Membership Inference Attackscited by 1
  9. 2025Perceptual implications of automatic anonymization in pathological speechcited by 0
  10. 2026Anonymization, Not Elimination: Utility-Preserved Speech Anonymizationcited by 0
  11. 2026The Text Beside the Image: Detection, Utility and Leakage for Trustworthy Multimodal Medical Data and Beyondcited by 0