Service interruption on Monday 11 July from 12:30 to 13:00: all the sites of the CCSD (HAL, EpiSciences, SciencesConf, AureHAL) will be inaccessible (network hardware connection).
Skip to Main content Skip to Navigation
Conference papers

Analyse de l'anonymisation du locuteur sur de la parole émotionnelle

Hubert Nourtel 1 Pierre Champion 1, 2 Denis Jouvet 1 Anthony Larcher 2 Marie Tahon 2 
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Speech data carries personal information, such as the speaker’s identity and emotional state, which can be used for malicious purposes. Among the studies that have addressed the topic of speech privacy, the VoicePrivacy initiative aims to promote the development of privacy tools for speechtechnologies. The goal of the VoicePrivacy Challenge 2020 (VPC) was to hide the source speaker’s identity while preserving the linguistic information. This paper studies the impact of the VPC speaker anonymization system on the emotional information of speech. Modifications to the system were also added in an attempt to mask the emotions while limiting signal degradation. Our results show that VPC baseline slightly masks the speakers’ emotions, but some modifications can mask the emotions more effectively while maintaining good intelligibility.
Document type :
Conference papers
Complete list of metadata
Contributor : Hubert NOURTEL Connect in order to contact the contributor
Submitted on : Monday, April 11, 2022 - 10:32:38 AM
Last modification on : Tuesday, May 24, 2022 - 10:19:30 AM


Files produced by the author(s)


  • HAL Id : hal-03636737, version 1


Hubert Nourtel, Pierre Champion, Denis Jouvet, Anthony Larcher, Marie Tahon. Analyse de l'anonymisation du locuteur sur de la parole émotionnelle. JEP 2022 - Journées d'Études sur la Parole, Jun 2022, Île de Noirmoutier, France. ⟨hal-03636737⟩



Record views


Files downloads