Speaker anonymisation using the McAdams coefficient

TL;DR

This paper introduces a training-free, efficient speech anonymisation method using the McAdams coefficient to modify spectral envelopes, effectively degrading speaker recognition while maintaining speech intelligibility.

Contribution

It presents a novel, signal processing-based anonymisation technique that does not require training data and outperforms existing methods in privacy protection.

Findings

Outperforms competing solutions in anonymisation effectiveness.

Maintains speech intelligibility with modest degradation.

Effective even against semi-informed adversaries.

Abstract

Anonymisation has the goal of manipulating speech signals in order to degrade the reliability of automatic approaches to speaker recognition, while preserving other aspects of speech, such as those relating to intelligibility and naturalness. This paper reports an approach to anonymisation that, unlike other current approaches, requires no training data, is based upon well-known signal processing techniques and is both efficient and effective. The proposed solution uses the McAdams coefficient to transform the spectral envelope of speech signals. Results derived using common VoicePrivacy 2020 databases and protocols show that random, optimised transformations can outperform competing solutions in terms of anonymisation while causing only modest, additional degradations to intelligibility, even in the case of a semi-informed privacy adversary.