MUSA: Multi-lingual Speaker Anonymization via Serial Disentanglement
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917723345256448 |
|---|---|
| author | Yao, Jixun Wang, Qing Guo, Pengcheng Ning, Ziqian Yang, Yuguang Pan, Yu Xie, Lei |
| author_facet | Yao, Jixun Wang, Qing Guo, Pengcheng Ning, Ziqian Yang, Yuguang Pan, Yu Xie, Lei |
| contents | Speaker anonymization is an effective privacy protection solution designed to conceal the speaker's identity while preserving the linguistic content and para-linguistic information of the original speech. While most prior studies focus solely on a single language, an ideal speaker anonymization system should be capable of handling multiple languages. This paper proposes MUSA, a Multi-lingual Speaker Anonymization approach that employs a serial disentanglement strategy to perform a step-by-step disentanglement from a global time-invariant representation to a temporal time-variant representation. By utilizing semantic distillation and self-supervised speaker distillation, the serial disentanglement strategy can avoid strong inductive biases and exhibit superior generalization performance across different languages. Meanwhile, we propose a straightforward anonymization strategy that employs empty embedding with zero values to simulate the speaker identity concealment process, eliminating the need for conversion to a pseudo-speaker identity and thereby reducing the complexity of speaker anonymization process. Experimental results on VoicePrivacy official datasets and multi-lingual datasets demonstrate that MUSA can effectively protect speaker privacy while preserving linguistic content and para-linguistic information. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_11629 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MUSA: Multi-lingual Speaker Anonymization via Serial Disentanglement Yao, Jixun Wang, Qing Guo, Pengcheng Ning, Ziqian Yang, Yuguang Pan, Yu Xie, Lei Audio and Speech Processing Speaker anonymization is an effective privacy protection solution designed to conceal the speaker's identity while preserving the linguistic content and para-linguistic information of the original speech. While most prior studies focus solely on a single language, an ideal speaker anonymization system should be capable of handling multiple languages. This paper proposes MUSA, a Multi-lingual Speaker Anonymization approach that employs a serial disentanglement strategy to perform a step-by-step disentanglement from a global time-invariant representation to a temporal time-variant representation. By utilizing semantic distillation and self-supervised speaker distillation, the serial disentanglement strategy can avoid strong inductive biases and exhibit superior generalization performance across different languages. Meanwhile, we propose a straightforward anonymization strategy that employs empty embedding with zero values to simulate the speaker identity concealment process, eliminating the need for conversion to a pseudo-speaker identity and thereby reducing the complexity of speaker anonymization process. Experimental results on VoicePrivacy official datasets and multi-lingual datasets demonstrate that MUSA can effectively protect speaker privacy while preserving linguistic content and para-linguistic information. |
| title | MUSA: Multi-lingual Speaker Anonymization via Serial Disentanglement |
| topic | Audio and Speech Processing |
| url | https://arxiv.org/abs/2407.11629 |