SecureSpeech: Prompt-based Speaker and Content Protection

TL;DR

SecureSpeech introduces a prompt-based speech generation method that anonymizes both speaker identity and spoken content, enhancing privacy while maintaining speech quality and content fidelity.

Contribution

The paper presents a novel prompt-based pipeline for dual speaker and content anonymization in speech synthesis, addressing privacy concerns in speech data.

Findings

Achieves significant privacy protection against speaker re-identification

Maintains high speech quality and content retention

Analyzes bias introduced by different speaker descriptions

Abstract

Given the increasing privacy concerns from identity theft and the re-identification of speakers through content in the speech field, this paper proposes a prompt-based speech generation pipeline that ensures dual anonymization of both speaker identity and spoken content. This is addressed through 1) generating a speaker identity unlinkable to the source speaker, controlled by descriptors, and 2) replacing sensitive content within the original text using a name entity recognition model and a large language model. The pipeline utilizes the anonymized speaker identity and text to generate high-fidelity, privacy-friendly speech via a text-to-speech synthesis model. Experimental results demonstrate an achievement of significant privacy protection while maintaining a decent level of content retention and audio quality. This paper also investigates the impact of varying speaker descriptions on the utility and privacy of generated speech to determine potential biases.