Publication detail

Single-Channel Speech Quality Enhancement in Mobile Networks Based on Generative Adversarial Networks

WU, G. HERENCSÁR, N.

Original Title

Single-Channel Speech Quality Enhancement in Mobile Networks Based on Generative Adversarial Networks

Type

journal article in Web of Science

Language

English

Original Abstract

A large amount of randomly generated noise in mobile networks leads to a lack of targeting and gaming processes in the speech enhancement process, and the enhancement process from the perspective of acoustic features alone suffers from major drawbacks. Propose a single-channel speech quality enhancement method based on generative adversarial networks in mobile networks. Explain the principle of generative adversarial network to realize single-channel speech quality enhancement in mobile networks and clarify its shortcomings. Design an improved Mel frequency cepstral coefficient extraction method to extract 12 base features as the enhancement basis. Use the relative average least squares loss instead of the traditional loss function to enhance the training efficiency, use the hybrid penalty term to enhance the generator's ability to generate single-channel speech, and optimize the discriminator through the multi-layer convolution and the addition of fully connected layers to enhance the speech quality enhancement ability of adversarial generative networks in various aspects, forming a relative average generative adversarial network (RaGAN) based on hybrid penalty term to realize single-channel speech quality enhancement processing. Through the experiment, when the discriminator is applied with the size of a 3*3 convolutional kernel, the best effect of speech quality enhancement is achieved in the mobile network. This method can complete the enhancement of single-channel speech quality in the mobile network, and the effect is significant, which can effectively reduce the noise in the original single-channel speech.

Keywords

Generative adversarial networks; RaGAN; Hybrid penalty term; Single-channel; Speech quality; Discriminator; Mobile networks

Authors

WU, G.; HERENCSÁR, N.

Released

2. 4. 2024

Publisher

SPRINGER

Location

NEW YORK

ISBN

1572-8153

Periodical

Mobile Networks and Applications

Year of study

2024

Number

neuvedeno

State

Kingdom of the Netherlands

Pages from

1

Pages to

15

Pages count

15

URL

Full text in the Digital Library

BibTex

@article{BUT188428,
  author="Guifen {Wu} and Norbert {Herencsár}",
  title="Single-Channel Speech Quality Enhancement in Mobile Networks Based on Generative Adversarial Networks",
  journal="Mobile Networks and Applications",
  year="2024",
  volume="2024",
  number="neuvedeno",
  pages="15",
  doi="10.1007/s11036-024-02300-4",
  issn="1572-8153",
  url="https://link.springer.com/article/10.1007/s11036-024-02300-4"
}