Evaluation of speaker de-identification based on voice gender and age conversion

Abstract

Two basic tasks are covered in this paper. The first one consists in the design and practical testing of a new method for voice de-identification that changes the apparent age and/or gender of a speaker by multi-segmental frequency scale transformation combined with prosody modification. The second task is aimed at verification of applicability of a classifier based on Gaussian mixture models (GMM) to detect the original Czech and Slovak speakers after applied voice deidentification. The performed experiments confirm functionality of the developed gender and age conversion for all selected types of de-identification which can be objectively evaluated by the GMM-based open-set classifier. The original speaker detection accuracy was compared also for sentences uttered by German and English speakers showing language independence of the proposed method.

Speaker Discrimination Using Long-Term Spectrum of Speech

“…Furthermore, it will be useful to also investigate the influence of adverse acoustic conditions, especially non-stationary noises [29] due to their indirect influence on speech production. Contrary to speaker recognition, it may be interesting to investigate the applicability of long-term spectra for spectral normalization in the field of speaker de-identification [16]. Recently, there has been a growing need for de-identification of multimedia data [18], in order to ensure their anonymity with respect to privacy protection in the European Union.…”

Privacy-Protecting Techniques for Behavioral Biometric Data: A Survey

“…They formulate this as an optimization problem and measure the distance between two speakers with a confusion factor, for which they evaluate entropy and Gini index as metrics. Pribil et al [164] propose a speaker de-identification method that relies on modifying several features of the source speaker. In the first step, the prosodic and spectral features are extracted from the source speaker.…”