EZ-VC: Easy Zero-shot Any-to-Any Voice Conversion

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Joglekar, Advait, Singh, Divyanshu, Bhatia, Rooshil Rohit, Umesh, S.
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915299970777088
author Joglekar, Advait
Singh, Divyanshu
Bhatia, Rooshil Rohit
Umesh, S.
author_facet Joglekar, Advait
Singh, Divyanshu
Bhatia, Rooshil Rohit
Umesh, S.
contents Voice Conversion research in recent times has increasingly focused on improving the zero-shot capabilities of existing methods. Despite remarkable advancements, current architectures still tend to struggle in zero-shot cross-lingual settings. They are also often unable to generalize for speakers of unseen languages and accents. In this paper, we adopt a simple yet effective approach that combines discrete speech representations from self-supervised models with a non-autoregressive Diffusion-Transformer based conditional flow matching speech decoder. We show that this architecture allows us to train a voice-conversion model in a purely textless, self-supervised fashion. Our technique works without requiring multiple encoders to disentangle speech features. Our model also manages to excel in zero-shot cross-lingual settings even for unseen languages. For Demo: https://ez-vc.github.io/EZ-VC-Demo/
format Preprint
id arxiv_https___arxiv_org_abs_2505_16691
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EZ-VC: Easy Zero-shot Any-to-Any Voice Conversion
Joglekar, Advait
Singh, Divyanshu
Bhatia, Rooshil Rohit
Umesh, S.
Sound
Artificial Intelligence
Audio and Speech Processing
Voice Conversion research in recent times has increasingly focused on improving the zero-shot capabilities of existing methods. Despite remarkable advancements, current architectures still tend to struggle in zero-shot cross-lingual settings. They are also often unable to generalize for speakers of unseen languages and accents. In this paper, we adopt a simple yet effective approach that combines discrete speech representations from self-supervised models with a non-autoregressive Diffusion-Transformer based conditional flow matching speech decoder. We show that this architecture allows us to train a voice-conversion model in a purely textless, self-supervised fashion. Our technique works without requiring multiple encoders to disentangle speech features. Our model also manages to excel in zero-shot cross-lingual settings even for unseen languages. For Demo: https://ez-vc.github.io/EZ-VC-Demo/
title EZ-VC: Easy Zero-shot Any-to-Any Voice Conversion
topic Sound
Artificial Intelligence
Audio and Speech Processing
url https://arxiv.org/abs/2505.16691