New Challenges for Content Privacy in Speech and Audio

TL;DR

This paper discusses emerging privacy challenges in speech and audio, emphasizing the need for technical solutions to protect content and contextual information in increasingly common audio technologies.

Contribution

It identifies three key research challenges in content privacy for speech and audio, highlighting gaps and opportunities for ethical and safer speech technology development.

Findings

Current privacy protections are insufficient for content and contextual audio information.

There are significant gaps in technical solutions for speech privacy enforcement.

Addressing these challenges can lead to more ethical and safer speech technologies.

Abstract

Privacy in speech and audio has many facets. A particularly under-developed area of privacy in this domain involves consideration for information related to content and context. Speech content can include words and their meaning or even stylistic markers, pathological speech, intonation patterns, or emotion. More generally, audio captured in-the-wild may contain background speech or reveal contextual information such as markers of location, room characteristics, paralinguistic sounds, or other audible events. Audio recording devices and speech technologies are becoming increasingly commonplace in everyday life. At the same time, commercialised speech and audio technologies do not provide consumers with a range of privacy choices. Even where privacy is regulated or protected by law, technical solutions to privacy assurance and enforcement fall short. This position paper introduces three important and timely research challenges for content privacy in speech and audio. We highlight current gaps and opportunities, and identify focus areas, that could have significant implications for developing ethical and safer speech technologies.