SegReConcat: A Data Augmentation Method for Voice Anonymization Attack
Not on the map. Why?
Abstract (arXiv)
Anonymization of voice seeks to conceal the identity of the speaker while maintaining the utility of speech data. However, residual speaker cues often persist, which pose privacy risks. We propose SegReConcat, a data augmentation method for attacker-side enhancement of automatic speaker verification systems. SegReConcat segments anonymized speech at the word level, rearranges segments using random or similarity-based strategies to disrupt long-term contextual cues, and concatenates them with the original utterance, allowing an attacker to learn source speaker traits from multiple perspectives. The proposed method has been evaluated in the VoicePrivacy Attacker Challenge 2024 framework across seven anonymization systems, SegReConcat improves de-anonymization on five out of seven systems.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 31 of its 42 references that are in the map. First posted on arXiv 26 Aug 2025.
Builds on
31 of 42 references.
References · 31
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
- 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,057
- 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
- 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,380
- 2021The VoicePrivacy 2020 Challenge: Results and findingscited by 46
- 2024The VoicePrivacy 2024 Challenge Evaluation Plancited by 41
- 2023Anonymizing Speech: Evaluating and Designing Speaker Anonymization Techniquescited by 22
- 2023Speaker anonymization using neural audio codec language modelscited by 22
- 2022Anonymizing Speech with Generative Adversarial Networks to Preserve Speaker Privacycited by 21
- 2023Speaker anonymization using orthogonal Householder neural networkcited by 18
- 2024The VoicePrivacy 2022 Challenge: Progress and Perspectives in Voice Anonymisationcited by 17
19 more
- 2022Differentially Private Speaker Anonymizationcited by 16
- 2022Language-Independent Speaker Anonymization Approach using Self-Supervised Pre-Trained Modelscited by 14
- 2024NPU-NTU System for Voice Privacy 2024 Challengecited by 14
- 2024MUSA: Multi-lingual Speaker Anonymization via Serial Disentanglementcited by 10
- 2024Privacy versus Emotion Preservation Trade-offs in Emotion-Preserving Speaker Anonymizationcited by 10
- 2024HLTCOE JHU Submission to the Voice Privacy Challenge 2024cited by 10
- 2024A Benchmark for Multi-speaker Anonymizationcited by 7
- 2024Distinctive and Natural Speaker Anonymization via Singular Value Transformation-assisted Matrixcited by 6
- 2024NTU-NPU System for Voice Privacy 2024 Challengecited by 6
- 2025EASY: Emotion-aware Speaker Anonymization via Factorized Distillationcited by 6
- 2025Private kNN-VC: Interpretable Anonymization of Converted Speechcited by 6
- 2024Anonymising Elderly and Pathological Speech: Voice Conversion Using DDSP and Query-by-Examplecited by 5
- 2024Analysis of Speech Temporal Dynamics in the Context of Speaker Verification and Voice Anonymizationcited by 5
- 2024Attacking Voice Anonymization Systems with Augmented Feature and Speaker Identity Differencecited by 5
- 2025SpecWav-Attack: Leveraging Spectrogram Resizing and Wav2Vec 2.0 for Attacking Anonymized Speechcited by 5
- 2024Adapting General Disentanglement-Based Speaker Anonymization for Enhanced Emotion Preservationcited by 4
- 2025Why disentanglement-based speaker anonymization systems fail at preserving emotions?cited by 3
- 2025Exploiting Context-dependent Duration Features for Voice Anonymization Attack Systemscited by 3
- 2025Inference Attacks for X-Vector Speaker Anonymizationcited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.