The VoicePrivacy Challenge promotes the development of voice anonymisation solutions for speech technology. In this paper we present a systematic overview and analysis of the second edition held in 2022. We describe the voice anonymisation task and datasets used for system development and evaluation, present the different attack models used for evaluation, and the associated objective and subjective metrics. We describe three anonymisation baselines, provide a summary description of the anonymisation systems developed by challenge participants, and report objective and subjective evaluation results for all. In addition, we describe post-evaluation analyses and a summary of related work reported in the open literature. Results show that solutions based on voice conversion better preserve utility, that an alternative which combines automatic speech recognition with synthesis achieves greater privacy, and that a privacy-utility trade-off remains inherent to current anonymisation solutions. Finally, we present our ideas and priorities for future VoicePrivacy Challenge editions.
Evaluating voice anonymisation using similarity rank disclosure
EURECOM(欧雷康);Aalto University(阿尔托大学);Technische Universität Berlin(柏林工业大学);Ruhr-Universität Bochum(波鸿鲁尔大学);Omilia(奥米利亚);Athens University of Economics and Business(雅典经济与商业大学);Orange(法国电信)
Avignon Université(阿维尼翁大学);Inria(法国国家信息与自动化研究所);National Institute of Informatics(国立信息学研究所);Université de Lorraine(洛林大学);CNRS(法国国家科学研究中心);LORIA(洛林信息学及其应用研究实验室);EURECOM(法国通信系统工程师学校);University of Edinburgh(爱丁堡大学)
Université de Lorraine, CNRS, Inria, Loria(法国洛林大学、法国国家科学研究中心、法国国家信息与自动化研究所、Loria研究中心);Singapore Institute of Technology(新加坡科技学院);National Institute of Informatics(日本信息处理学会)
The Third VoicePrivacy Challenge: Preserving Emotional Expressiveness and Linguistic Content in Voice Anonymization
organization= Universit\'e de Lorraine, CNRS, Inria, LORIA, F-54000 , city= Nancy , country= France;organization= Duke Kunshan University , city= Kunshan , postcode= 215316 , country= China;organization= Mission Défense et Sécurité, Inria , city= Grenoble , country= France;organization= Institute for Natural Language Processing, University of Stuttgart , country= Germany;organization= National Institute of Informatics , addressline= Chiyoda-ku , city= Tokyo , postcode= 101-8340 , country= Japan
The VoicePrivacy 2020 Challenge: Results and findings
University of Avignon(阿维尼翁大学);National Institute of Informatics (NII)(国立信息学研究所);Université de Lorraine(洛林大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);EURECOM(欧洲通信学院);University of Edinburgh(爱丁堡大学);Aix-Marseille University(艾克斯-马赛大学)
Evaluation of the Speech Resynthesis Capabilities of the VoicePrivacy Challenge Baseline B1
International Audio Laboratories Erlangen(埃尔朗根国际音频实验室);Friedrich-Alexander-Universität Erlangen-Nürnberg(弗里德里希-亚历山大-埃尔朗根-纽伦堡大学);University of Erlangen-Nürnberg(埃尔朗根-纽伦堡大学);Fraunhofer IIS(弗劳恩霍夫集成电路研究所)
Avignon Université(阿维尼翁大学);National Institute of Informatics(国立信息学研究所);Université de Lorraine(洛林大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);Le Mans Université(勒芒大学);EURECOM(欧洲通信学院);University of Edinburgh(爱丁堡大学)
Inria(法国国家信息与自动化研究所);Singapore Institute of Technology(新加坡理工学院);Institute for Natural Language Processing, University of Stuttgart(斯图加特大学自然语言处理研究所);National Institute of Informatics(情报信息学研究所);EURECOM(欧洲通信学院)