Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2605.07291 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915991765647360 |
|---|---|
| author | Chandra, Shilpa Pettenò, Matteo Evans, Nicholas Panariello, Michele Todisco, Massimiliano Bäckström, Tom Kolossa, Dorothea Martin, Rainer Stafylakis, Themos Gengembre, Nicolas |
| author_facet | Chandra, Shilpa Pettenò, Matteo Evans, Nicholas Panariello, Michele Todisco, Massimiliano Bäckström, Tom Kolossa, Dorothea Martin, Rainer Stafylakis, Themos Gengembre, Nicolas |
| contents | The evaluation of voice anonymisation remains challenging. Current practice relies on automatic speaker verification metrics such as the equal error rate (EER). Performance estimates dependent on the classifier and operating point provide an incomplete or even misleading characterisation of privacy risk. We investigate the use of similarity rank disclosure (SRD), an information-theoretic metric, which operates on feature representations rather than classifier decisions, providing a threshold-independent assessment of privacy and analysis of both average and worst-case disclosure. We report its application to speaker embeddings, fundamental frequency, and phone embeddings using 2024 VoicePrivacy Challenge systems. The SRD reveals privacy leaks and system-specific weaknesses missed by EER-based evaluation. Findings highlight the merit of representation-level metrics and demonstrate the potential of SRD as a flexible and interpretable tool for the evaluation of voice anonymisation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_07291 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evaluating voice anonymisation using similarity rank disclosure Chandra, Shilpa Pettenò, Matteo Evans, Nicholas Panariello, Michele Todisco, Massimiliano Bäckström, Tom Kolossa, Dorothea Martin, Rainer Stafylakis, Themos Gengembre, Nicolas Audio and Speech Processing The evaluation of voice anonymisation remains challenging. Current practice relies on automatic speaker verification metrics such as the equal error rate (EER). Performance estimates dependent on the classifier and operating point provide an incomplete or even misleading characterisation of privacy risk. We investigate the use of similarity rank disclosure (SRD), an information-theoretic metric, which operates on feature representations rather than classifier decisions, providing a threshold-independent assessment of privacy and analysis of both average and worst-case disclosure. We report its application to speaker embeddings, fundamental frequency, and phone embeddings using 2024 VoicePrivacy Challenge systems. The SRD reveals privacy leaks and system-specific weaknesses missed by EER-based evaluation. Findings highlight the merit of representation-level metrics and demonstrate the potential of SRD as a flexible and interpretable tool for the evaluation of voice anonymisation. |
| title | Evaluating voice anonymisation using similarity rank disclosure |
| topic | Audio and Speech Processing |
| url | https://arxiv.org/abs/2605.07291 |