Optimal Transport Maps are Good Voice Converters

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Asadulaev, Arip, Korst, Rostislav, Shutov, Vitalii, Korotin, Alexander, Grebnyak, Yaroslav, Egiazarian, Vahe, Burnaev, Evgeny
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929576319385600
author Asadulaev, Arip
Korst, Rostislav
Shutov, Vitalii
Korotin, Alexander
Grebnyak, Yaroslav
Egiazarian, Vahe
Burnaev, Evgeny
author_facet Asadulaev, Arip
Korst, Rostislav
Shutov, Vitalii
Korotin, Alexander
Grebnyak, Yaroslav
Egiazarian, Vahe
Burnaev, Evgeny
contents Recently, neural network-based methods for computing optimal transport maps have been effectively applied to style transfer problems. However, the application of these methods to voice conversion is underexplored. In our paper, we fill this gap by investigating optimal transport as a framework for voice conversion. We present a variety of optimal transport algorithms designed for different data representations, such as mel-spectrograms and latent representation of self-supervised speech models. For the mel-spectogram data representation, we achieve strong results in terms of Frechet Audio Distance (FAD). This performance is consistent with our theoretical analysis, which suggests that our method provides an upper bound on the FAD between the target and generated distributions. Within the latent space of the WavLM encoder, we achived state-of-the-art results and outperformed existing methods even with limited reference speaker data.
format Preprint
id arxiv_https___arxiv_org_abs_2411_02402
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Optimal Transport Maps are Good Voice Converters
Asadulaev, Arip
Korst, Rostislav
Shutov, Vitalii
Korotin, Alexander
Grebnyak, Yaroslav
Egiazarian, Vahe
Burnaev, Evgeny
Sound
Machine Learning
Audio and Speech Processing
Recently, neural network-based methods for computing optimal transport maps have been effectively applied to style transfer problems. However, the application of these methods to voice conversion is underexplored. In our paper, we fill this gap by investigating optimal transport as a framework for voice conversion. We present a variety of optimal transport algorithms designed for different data representations, such as mel-spectrograms and latent representation of self-supervised speech models. For the mel-spectogram data representation, we achieve strong results in terms of Frechet Audio Distance (FAD). This performance is consistent with our theoretical analysis, which suggests that our method provides an upper bound on the FAD between the target and generated distributions. Within the latent space of the WavLM encoder, we achived state-of-the-art results and outperformed existing methods even with limited reference speaker data.
title Optimal Transport Maps are Good Voice Converters
topic Sound
Machine Learning
Audio and Speech Processing
url https://arxiv.org/abs/2411.02402