Provable Speech Attributes Conversion via Latent Independence
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909832162836480 |
|---|---|
| author | Svirsky, Jonathan Lindenbaum, Ofir Shaham, Uri |
| author_facet | Svirsky, Jonathan Lindenbaum, Ofir Shaham, Uri |
| contents | While signal conversion and disentangled representation learning have shown promise for manipulating data attributes across domains such as audio, image, and multimodal generation, existing approaches, especially for speech style conversion, are largely empirical and lack rigorous theoretical foundations to guarantee reliable and interpretable control. In this work, we propose a general framework for speech attribute conversion, accompanied by theoretical analysis and guarantees under reasonable assumptions. Our framework builds on a non-probabilistic autoencoder architecture with an independence constraint between the predicted latent variable and the target controllable variable. This design ensures a consistent signal transformation, conditioned on an observed style variable, while preserving the original content and modifying the desired attribute. We further demonstrate the versatility of our method by evaluating it on speech styles, including speaker identity and emotion. Quantitative evaluations confirm the effectiveness and generality of the proposed approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_05191 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Provable Speech Attributes Conversion via Latent Independence Svirsky, Jonathan Lindenbaum, Ofir Shaham, Uri Sound Artificial Intelligence While signal conversion and disentangled representation learning have shown promise for manipulating data attributes across domains such as audio, image, and multimodal generation, existing approaches, especially for speech style conversion, are largely empirical and lack rigorous theoretical foundations to guarantee reliable and interpretable control. In this work, we propose a general framework for speech attribute conversion, accompanied by theoretical analysis and guarantees under reasonable assumptions. Our framework builds on a non-probabilistic autoencoder architecture with an independence constraint between the predicted latent variable and the target controllable variable. This design ensures a consistent signal transformation, conditioned on an observed style variable, while preserving the original content and modifying the desired attribute. We further demonstrate the versatility of our method by evaluating it on speech styles, including speaker identity and emotion. Quantitative evaluations confirm the effectiveness and generality of the proposed approach. |
| title | Provable Speech Attributes Conversion via Latent Independence |
| topic | Sound Artificial Intelligence |
| url | https://arxiv.org/abs/2510.05191 |