StarGAN-VC+ASR: StarGAN-based Non-Parallel Voice Conversion Regularized by Automatic Speech Recognition

TL;DR

This paper introduces a novel approach combining StarGAN-VC with automatic speech recognition to enhance non-parallel voice conversion, especially in low-resource scenarios, by better preserving linguistic content.

Contribution

It proposes integrating automatic speech recognition into StarGAN-VC to improve linguistic content preservation in low-resource voice conversion tasks.

Findings

Enhanced linguistic content retention in low-resource settings

StarGAN-VC+ASR outperforms vanilla StarGAN-VC in experiments

Effective regularization of voice conversion model

Abstract

Preserving the linguistic content of input speech is essential during voice conversion (VC). The star generative adversarial network-based VC method (StarGAN-VC) is a recently developed method that allows non-parallel many-to-many VC. Although this method is powerful, it can fail to preserve the linguistic content of input speech when the number of available training samples is extremely small. To overcome this problem, we propose the use of automatic speech recognition to assist model training, to improve StarGAN-VC, especially in low-resource scenarios. Experimental results show that using our proposed method, StarGAN-VC can retain more linguistic information than vanilla StarGAN-VC.