Saved in:
Bibliographic Details
Main Authors: Seo, Seungmin, Aulov, Oleg, Godil, Afzal, Mangold, Kevin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.14012
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918127314403328
author Seo, Seungmin
Aulov, Oleg
Godil, Afzal
Mangold, Kevin
author_facet Seo, Seungmin
Aulov, Oleg
Godil, Afzal
Mangold, Kevin
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2508_14012
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating Identity Leakage in Speaker De-Identification Systems
Seo, Seungmin
Aulov, Oleg
Godil, Afzal
Mangold, Kevin
Sound
Artificial Intelligence
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.
title Evaluating Identity Leakage in Speaker De-Identification Systems
topic Sound
Artificial Intelligence
url https://arxiv.org/abs/2508.14012