The VoicePrivacy 2020 Challenge: Results and findings

TL;DR

The paper reports on the VoicePrivacy 2020 Challenge, analyzing anonymization solutions for speech, evaluating various systems and attack models, and providing insights for future research in voice privacy protection.

Contribution

It offers a comprehensive overview of the challenge design, evaluation metrics, baseline systems, submitted solutions, and post-evaluation analyses for voice anonymization.

Findings

Baseline and submitted systems show varying effectiveness in anonymization.

Evaluation metrics reveal strengths and weaknesses of different approaches.

Insights will guide future voice privacy research and challenge design.

Abstract

This paper presents the results and analyses stemming from the first VoicePrivacy 2020 Challenge which focuses on developing anonymization solutions for speech technology. We provide a systematic overview of the challenge design with an analysis of submitted systems and evaluation results. In particular, we describe the voice anonymization task and datasets used for system development and evaluation. Also, we present different attack models and the associated objective and subjective evaluation metrics. We introduce two anonymization baselines and provide a summary description of the anonymization systems developed by the challenge participants. We report objective and subjective evaluation results for baseline and submitted systems. In addition, we present experimental results for alternative privacy metrics and attack models developed as a part of the post-evaluation analysis. Finally, we summarize our insights and observations that will influence the design of the next VoicePrivacy challenge edition and some directions for future voice anonymization research.