Inference Attacks for X-Vector Speaker Anonymization
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909609634037760 |
|---|---|
| author | Bauer, Luke Bao, Wenxuan Jadhav, Malvika Bindschaedler, Vincent |
| author_facet | Bauer, Luke Bao, Wenxuan Jadhav, Malvika Bindschaedler, Vincent |
| contents | We revisit the privacy-utility tradeoff of x-vector speaker anonymization. Existing approaches quantify privacy through training complex speaker verification or identification models that are later used as attacks. Instead, we propose a novel inference attack for de-anonymization. Our attack is simple and ML-free yet we show experimentally that it outperforms existing approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_08978 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Inference Attacks for X-Vector Speaker Anonymization Bauer, Luke Bao, Wenxuan Jadhav, Malvika Bindschaedler, Vincent Cryptography and Security Sound Audio and Speech Processing We revisit the privacy-utility tradeoff of x-vector speaker anonymization. Existing approaches quantify privacy through training complex speaker verification or identification models that are later used as attacks. Instead, we propose a novel inference attack for de-anonymization. Our attack is simple and ML-free yet we show experimentally that it outperforms existing approaches. |
| title | Inference Attacks for X-Vector Speaker Anonymization |
| topic | Cryptography and Security Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2505.08978 |