HLTCOE Submission to the VoicePrivacy Attacker Challenge
We describe our submission to the 2024 VoicePrivacy Attacker Challenge. We propose three main categories of methods to improve ASV performance against anonymized speech: improvements to the underlying classifier, alternative distance metrics when computing ASV scores, and kNN-VC normalization. By simultaneously employing one or more of these methods, we were able to achieve a significant reduction in EER against all of the submitted anonymization systems in the VoicePrivacy Challenge.
Authors
- Matthew Wiesner (ORCID: https://orcid.org/0000-0001-8653-7738)
- Sanjeev Khudanpur (ORCID: https://orcid.org/0000-0001-5976-0897)
- Henry Li Xinyuan
- Leibny Paola Garcia (ORCID: https://orcid.org/0000-0002-7449-5726)
- Kevin Duh (ORCID: https://orcid.org/0000-0001-8107-4383)
- Zexin Cai
- Ashi Garg
- Nicholas Andrews
Institutions
- Johns Hopkins University (US)
Publication Details
- Published
- 2025-03-12
- DOI
- https://doi.org/10.1109/icassp49660.2025.10888439
- Citations
- 3
- Primary Topic
- Speech Recognition and Synthesis
- Type
- article
- Field-Weighted Citation Impact
- 7.44
Funders
- Office of the Director of National Intelligence
- Intelligence Advanced Research Projects Activity