A noise-robust voice conversion method with controllable background sounds

Abstract Background noises are usually treated as redundant or even harmful to voice conversion. Therefore, when converting noisy speech, a pretrained module of speech separation is usually deployed to estimate clean speech prior to the conversion. However, this can lead to speech distortion due to...

Full description

Bibliographic Details

Published in:	Complex & Intelligent Systems
Main Authors:	Lele Chen, Xiongwei Zhang, Yihao Li, Meng Sun, Weiwei Chen
Format:	Article
Language:	English
Published:	Springer 2024-02-01
Subjects:	Noise-robust voice conversion Dual-decoder structure Bridge module Cycle loss Speech disentanglement
Online Access:	https://doi.org/10.1007/s40747-024-01375-6