SecureSpeech: Prompt-based Speaker and Content Protection
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908444693364736 |
|---|---|
| author | Hui, Belinda Soh Hui Miao, Xiaoxiao Wang, Xin |
| author_facet | Hui, Belinda Soh Hui Miao, Xiaoxiao Wang, Xin |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_07799 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SecureSpeech: Prompt-based Speaker and Content Protection Hui, Belinda Soh Hui Miao, Xiaoxiao Wang, Xin Sound Audio and Speech Processing 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. |
| title | SecureSpeech: Prompt-based Speaker and Content Protection |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2507.07799 |