mlsanityData to 14 Oct 2026

Exploiting Context-dependent Duration Features for Voice Anonymization Attack Systems

arXiv 2507.15214 ↗First posted 21 Jul 2025cs.SD cs.CL cs.CRcited by 3

Not on the map. Why?

Abstract (arXiv)

The temporal dynamics of speech, encompassing variations in rhythm, intonation, and speaking rate, contain important and unique information about speaker identity. This paper proposes a new method for representing speaker characteristics by extracting context-dependent duration embeddings from speech temporal dynamics. We develop novel attack models using these representations and analyze the potential vulnerabilities in speaker verification and voice anonymization systems.The experimental results show that the developed attack models provide a significant improvement in speaker verification performance for both original and anonymized data in comparison with simpler representations of speech temporal dynamics reported in the literature.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 18 of its 32 references that are in the map. First posted on arXiv 21 Jul 2025.

Builds on

18 of 32 references.

References · 18

  1. 2020ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verificationcited by 628
  2. 2020Introducing the VoicePrivacy Initiativecited by 49
  3. 2019Speaker Anonymization Using X-vector and Neural Waveform Modelscited by 41
  4. 2024The VoicePrivacy 2024 Challenge Evaluation Plancited by 41
  5. 2020Speaker anonymisation using the McAdams coefficientcited by 38
  6. 2023Anonymizing Speech: Evaluating and Designing Speaker Anonymization Techniquescited by 22
  7. 2023Speaker anonymization using neural audio codec language modelscited by 22
  8. 2019The GDPR & Speech Data: Reflections of Legal and Technology Communities, First Steps towards a Common Understandingcited by 19
  9. 2023Speaker anonymization using orthogonal Householder neural networkcited by 18
  10. 2022Language-Independent Speaker Anonymization Approach using Self-Supervised Pre-Trained Modelscited by 14
  11. 2024NPU-NTU System for Voice Privacy 2024 Challengecited by 14
  12. 2022Improving speaker de-identification with functional data analysis of f0 trajectoriescited by 12
6 more
  1. 2024MUSA: Multi-lingual Speaker Anonymization via Serial Disentanglementcited by 10
  2. 2024HLTCOE JHU Submission to the Voice Privacy Challenge 2024cited by 10
  3. 2024Analysis of Speech Temporal Dynamics in the Context of Speaker Verification and Voice Anonymizationcited by 5
  4. 2025The First VoicePrivacy Attacker Challengecited by 4
  5. 2024Speech Rhythm-Based Speaker Embeddings Extraction from Phonemes and Phoneme Duration for Multi-Speaker Speech Synthesiscited by 3
  6. 2024Voice Conversion-based Privacy through Adversarial Information Hidingcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2026The Third VoicePrivacy Challenge: Preserving Emotional Expressiveness and Linguistic Content in Voice Anonymizationcited by 8
  2. 2025SegReConcat: A Data Augmentation Method for Voice Anonymization Attackcited by 1
  3. 2025Evaluating Identity Leakage in Speaker De-Identification Systemscited by 0