Novel Loss-Enhanced Universal Adversarial Patches for Sustainable Speaker Privacy
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912395290476544 |
|---|---|
| author | Karimov, Elvir Varlamov, Alexander Ivanov, Danil Korzh, Dmitrii Rogov, Oleg Y. |
| author_facet | Karimov, Elvir Varlamov, Alexander Ivanov, Danil Korzh, Dmitrii Rogov, Oleg Y. |
| contents | Deep learning voice models are commonly used nowadays, but the safety processing of personal data, such as human identity and speech content, remains suspicious. To prevent malicious user identification, speaker anonymization methods were proposed. Current methods, particularly based on universal adversarial patch (UAP) applications, have drawbacks such as significant degradation of audio quality, decreased speech recognition quality, low transferability across different voice biometrics models, and performance dependence on the input audio length. To mitigate these drawbacks, in this work, we introduce and leverage the novel Exponential Total Variance (TV) loss function and provide experimental evidence that it positively affects UAP strength and imperceptibility. Moreover, we present a novel scalable UAP insertion procedure and demonstrate its uniformly high performance for various audio lengths. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_19951 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Novel Loss-Enhanced Universal Adversarial Patches for Sustainable Speaker Privacy Karimov, Elvir Varlamov, Alexander Ivanov, Danil Korzh, Dmitrii Rogov, Oleg Y. Sound Artificial Intelligence Cryptography and Security Audio and Speech Processing Deep learning voice models are commonly used nowadays, but the safety processing of personal data, such as human identity and speech content, remains suspicious. To prevent malicious user identification, speaker anonymization methods were proposed. Current methods, particularly based on universal adversarial patch (UAP) applications, have drawbacks such as significant degradation of audio quality, decreased speech recognition quality, low transferability across different voice biometrics models, and performance dependence on the input audio length. To mitigate these drawbacks, in this work, we introduce and leverage the novel Exponential Total Variance (TV) loss function and provide experimental evidence that it positively affects UAP strength and imperceptibility. Moreover, we present a novel scalable UAP insertion procedure and demonstrate its uniformly high performance for various audio lengths. |
| title | Novel Loss-Enhanced Universal Adversarial Patches for Sustainable Speaker Privacy |
| topic | Sound Artificial Intelligence Cryptography and Security Audio and Speech Processing |
| url | https://arxiv.org/abs/2505.19951 |