Identity leakage through accent cues in voice anonymisation

TL;DR

This paper investigates how residual accent cues in voice anonymisation can lead to re-identification, highlighting the importance of accent-related information and proposing a system to reduce accent-based re-identification.

Contribution

It demonstrates the extent of residual accent information in anonymised voices and introduces a character-level conditioning system to better obfuscate accent cues.

Findings

Some anonymisation systems leave residual accent cues facilitating re-identification.

A character-level conditioning system reduces accent-identification accuracy by 68%.

The proposed system improves overall anonymisation performance by 11%.

Abstract

Voice anonymisation is used to conceal voice identity while preserving linguistic content. Even if anonymisation seems strong, non-timbral cues such as accent that remain post-anonymisation can help re-identification and reveal sensitive socio-demographic traits. We report a study of residual accent information involving multiple anonymisation systems. We highlight the role of accent using speaker verification, accent verification, and accent classification using a set of embeddings focusing on timbral, non-timbral and accent-related information and show the extent to which related cues facilitate reidentification post anonymisation. Results show that, while some systems are robust to reidentification attempts using accent cues, others leave residual, speaker-dependent, accentrelated cues which can be used to reveal the voice identity. We also highlight accent-dependent variation in anonymisation performance, raising fairness concerns, and show that a system with characterlevel conditioning can help obfuscate identity-revealing accent cues, reducing accent-identification accuracy by 68% on average and improving overall anonymisation performance by 11% relative.