Beyond EER: Multi-Dimensional Evaluation of Information Leakage in Speaker De-Identification

작성자

카테고리:

← 피드로
arXiv cs.AI · Seungmin Seo, Oleg Aulov, P. Jonathon Phillips, Kevin Mangold, Jonathan Eskin · 2026-09-17 AI

[Submitted on 16 Sep 2026]

View PDF HTML (experimental)

Abstract:Speaker de-identification (SDID) aims to preserve privacy by concealing speaker identity while maintaining speech utility. However, current evaluations often reduce privacy to a single dimension – biometric verification performance – typically measured by Equal Error Rate (EER). This narrow focus ignores critical leakage channels, such as soft biometric inference, embedding-level re-identification, and structural template similarity, which threaten the unlinkability and irreversibility of biometric references. We propose a holistic evaluation framework across five complementary metrics: (i) EER, (ii) soft biometric leakage score , (iii) cumulative match characteristic re-identification analysis, (iv) canonical correlation analysis and Procrustes embedding alignment, and (v) intelligibility via word error rate and semantic similarity. Evaluating five SDID systems from the IARPA ARTS program, we demonstrate that these metrics capture independent dimensions of information leakage. Our results indicate that reliance on a single metric can misrepresent the privacy properties of an SDID system.

Submission history

From: Seungmin Seo [view email]
[v1] Wed, 16 Sep 2026 13:50:55 UTC (578 KB)

원문에서 계속 ↗

추출 본문 · 출처: arxiv.org · https://arxiv.org/abs/2609.18673