Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise

YUAN Wen-hao; LOU Ying-xi; LIANG Chun-yan; XIA Bin

doi:10.3969/j.issn.0372-2112.2019.04.004

您当前的位置：

首页 >

文章列表页 >

Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise

更新时间：2025-07-16

- Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise
- Acta Electronica Sinica Vol. 47, Issue 4, Pages: 791-797(2019)
- 作者机构：
  
  山东理工大学计算机科学与技术学院,山东,淄博,255000
- 作者简介：
- 基金信息：
- DOI：10.3969/j.issn.0372-2112.2019.04.004
  CLC： TN912.3
- Published Online：25 April 2019，
  
  Published：2019
- 稿件说明：
移动端阅览
YUAN Wen-hao, LOU Ying-xi, LIANG Chun-yan, et al. Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise[J]. Acta Electronica Sinica, 2019, 47(4): 791-797.
DOI：

YUAN Wen-hao, LOU Ying-xi, LIANG Chun-yan, et al. Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise[J]. Acta Electronica Sinica, 2019, 47(4): 791-797. DOI： 10.3969/j.issn.0372-2112.2019.04.004.

摘要

如何提高对未知噪声类型的泛化能力是有监督语音增强方法中亟待解决的重要问题，通过对大量不同类型噪声进行建模，深度神经网络成为了解决该问题的有效手段.为了进一步提高基于深度神经网络的语音增强方法的泛化能力，本文基于生成式对抗网络（Generative Adversarial Networks，GAN）设计了能够由真实噪声数据生成新的噪声类型的NoiseGAN；通过在训练集中增加生成噪声类型，提高训练集噪声类型的多样性，从而达到提高语音增强模型泛化能力的目的.不同结构的网络下的语音增强实验结果表明，本文提出的NoiseGAN能够生成新的噪声类型，具备提高训练集噪声类型多样性的能力，有效提高了语音增强模型在未知噪声类型下的泛化能力.

Abstract

How to improve the generalization ability to unknown noise types is an important problem to be solved urgently in supervised speech enhancement approaches.By modeling a large number of types of noise

the deep neural network(DNN)becomes an effective way to solve this problem.In order to further improve the generalization ability of speech enhancement approaches based on DNN

this paper designs NoiseGAN based on Generative Adversarial Networks (GAN) to generate new noise types from real noise data.By adding generated noise to training set

the diversity of noise types in training set is increased

and thereby the generalization ability of speech enhancement model is improved.The results of speech enhancement experiments under different structures of networks show that the proposed NoiseGAN can generate new noise types

increase the diversity of noise types in training set

and effectively improve the generalization ability of speech enhancement models under unknown noise types.

关键词

Keywords

references

Views

513

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Convolutional Gated Recurrent Network for Speech Enhancement

Speech Enhancement Method Based on Binaural Cues Coding Principle

Adversarial Learning and Enhanced Optimization Based Restoration Method for VC-Generated Speeches

Binary Code Similarity Detection Method Based on Cross-Modal Coordinated Representation Learning

A Hierarchical Automatic Sleep Staging Model Based on Transformer Architecture

Related Author

HU Shao-dong

SHI Yun-long

LI Zhao

CHEN Nan

BAO Chang-chun

SU Zhao-pin

ZHOU Xiao-lin

ZHANG Guo-fu

Related Institution

Faculty of Information Technology, Beijing University of Technology

Joint Laboratory of Intelligent Prevention and Recognition of Audio and Video

Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology)

Department of Physical Evidence Identification, Anhui Public Security Department

School of Computer and Information Technology, Hefei University of Technology

⁰