Graduate School and Research Center in Digital Sciences

Preserving privacy in speaker and speech characterisation

Nautsch, Andreas; Jiménez, Abelino; Treiber, Amos; Kolberg, Jascha; Jasserand, Catherine; Kindt, Els ; Delgado, Héctor; Todisco, Massimiliano; Hmani , Mohamed Amine; Mtibaa, Aymen; Abdelraheem, Mohammed Ahmed; Abad, Alberto; Teixeira, Francisco; Matrouf, Driss; Gomez-Barrero, Marta; Petrovska-Delacrétaz, Dijana; Chollet, Gérard; Evans, Nicholas; Bonastre, Jean-François; Raj, Bhiksha; Trancoso, Isabel; Busch, Christoph

Computer Speech and Language, June 2019

Speech recordings are a rich source of personal, sensitive data that can be used to support a plethora of diverse applications, from health profiling to biometric recognition. It is therefore essential that speech recordings are adequately protected so that they cannot be misused. Such protection, in the form of privacy-preserving technologies, is required to ensure that: (i) the biometric profiles of a given individual (e.g., across different biometric service operators) are unlinkable; (ii) leaked, encrypted biometric information is irreversible, and that (iii) biometric references are renewable. Whereas many privacy-preserving technologies have been developed for other biometric characteristics, very few solutions have been proposed to protect privacy in the case of speech signals. Despite privacy preservation this is now being mandated by recent European and international data protection regulations. With the aim of fostering progress and collaboration between researchers in the speech, biometrics and applied cryptography communities, this survey article provides an introduction to the field, starting with a legal perspective on privacy preservation in the case of speech data. It then establishes the requirements for effective privacy preservation, reviews generic cryptography-based solutions, followed by specific techniques that are applicable to speaker characterisation (biometric applications) and speech characterisation (non-biometric applications). Glancing at non-biometrics, methods are presented to avoid function creep, preventing the exploitation of biometric information, e.g., to single out an identity in speech-assisted health care via speaker characterisation. In promoting harmonised research, the article also outlines common, empirical evaluation metrics for the assessment of privacy-preserving technologies for speech data.

Doi Bibtex

Title:Preserving privacy in speaker and speech characterisation
Keywords:Data privacy, Voice biometrics, Standardisation, Cryptography, Legislation
Type:Journal
Language:English
City:
Date:
Department:Digital Security
Eurecom ref:5910
Copyright: © Elsevier. Personal use of this material is permitted. The definitive version of this paper was published in Computer Speech and Language, June 2019 and is available at : https://doi.org/10.1016/j.csl.2019.06.001
Bibtex: @article{EURECOM+5910, doi = {https://doi.org/10.1016/j.csl.2019.06.001}, year = {2019}, month = {06}, title = {{P}reserving privacy in speaker and speech characterisation}, author = {{N}autsch, {A}ndreas and {J}im{\'e}nez, {A}belino and {T}reiber, {A}mos and {K}olberg, {J}ascha and {J}asserand, {C}atherine and {K}indt, {E}ls and {D}elgado, {H}{\'e}ctor and {T}odisco, {M}assimiliano and {H}mani , {M}ohamed {A}mine and {M}tibaa, {A}ymen and {A}bdelraheem, {M}ohammed {A}hmed and {A}bad, {A}lberto and {T}eixeira, {F}rancisco and {M}atrouf, {D}riss and {G}omez-{B}arrero, {M}arta and {P}etrovska-{D}elacr{\'e}taz, {D}ijana and {C}hollet, {G}{\'e}rard and {E}vans, {N}icholas and {B}onastre, {J}ean-{F}ran{\'c}ois and {R}aj, {B}hiksha and {T}rancoso, {I}sabel and {B}usch, {C}hristoph}, journal = {{C}omputer {S}peech and {L}anguage, {J}une 2019}, url = {http://www.eurecom.fr/publication/5910} }
See also: