{"ID":2880798,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.14012","arxiv_id":"2508.14012","title":"Evaluating Identity Leakage in Speaker De-Identification Systems","abstract":"Speaker de-identification aims to conceal a speaker's identity while preserving intelligibility of the underlying speech. We introduce a benchmark that quantifies residual identity leakage with three complementary error rates: equal error rate, cumulative match characteristic hit rate, and embedding-space similarity measured via canonical correlation analysis and Procrustes analysis. Evaluation results reveal that all state-of-the-art speaker de-identification systems leak identity information. The highest performing system in our evaluation performs only slightly better than random guessing, while the lowest performing system achieves a 45% hit rate within the top 50 candidates based on CMC. These findings highlight persistent privacy risks in current speaker de-identification technologies.","short_abstract":"Speaker de-identification aims to conceal a speaker's identity while preserving intelligibility of the underlying speech. We introduce a benchmark that quantifies residual identity leakage with three complementary error rates: equal error rate, cumulative match characteristic hit rate, and embedding-space similarity me...","url_abs":"https://arxiv.org/abs/2508.14012","url_pdf":"https://arxiv.org/pdf/2508.14012v1","authors":"[\"Seungmin Seo\",\"Oleg Aulov\",\"Afzal Godil\",\"Kevin Mangold\"]","published":"2025-08-19T17:20:25Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\"]","methods":"[]","has_code":false}