低信噪比下基于信号子空间的语音增强算法研究

王红; 王爱平; 邹海

Chinese

您当前的位置：

首页 >

文章列表页 >

低信噪比下基于信号子空间的语音增强算法研究

更新时间：2026-09-04

- 低信噪比下基于信号子空间的语音增强算法研究
- Speech Enhancement Algorithm based on Signal Subspace with Low Signal-to-Noise Ratio
- 微电子学与计算机 2017年34卷第2期页码：43-47
- 作者机构：
  
  安徽大学计算机科学与技术学院,安徽,合肥,230601
- 作者简介：
- 基金信息：
  
  国家自然科学基金 61573022
- DOI：
  中图分类号：
- 纸质出版：2017
- 稿件说明：
移动端阅览
王红, 王爱平, 邹海. 低信噪比下基于信号子空间的语音增强算法研究[J]. 微电子学与计算机, 2017, 34(2): 43-47.

WANG Hong, WANG Ai-ping, Zou Hai. Speech Enhancement Algorithm based on Signal Subspace with Low Signal-to-Noise Ratio[J]. 2017, 34(2): 43-47.
王红, 王爱平, 邹海. 低信噪比下基于信号子空间的语音增强算法研究[J]. 微电子学与计算机, 2017, 34(2): 43-47. DOI：

WANG Hong, WANG Ai-ping, Zou Hai. Speech Enhancement Algorithm based on Signal Subspace with Low Signal-to-Noise Ratio[J]. 2017, 34(2): 43-47. DOI：

摘要

内嵌式预白化子空间算法与其它经典语音增强算法相比有很好的去噪效果

但是在低信噪比环境下

仍存在大量残留噪声.针对这一点

本文提出信号子空间优化与维纳滤波方法相结合的改进方法.仿真结果表明

在几种常见背景噪声的低信噪比下

相比传统的谱减法、维纳滤波法、内嵌式预白化子空间算法

本文算法效果更佳

能有效的抑制背景噪声

改善语音质量

并且保证一定的语音可懂度.

Abstract

Compared with other classic speech enhancement algorithm

buit-in pre-whitening subspacealgorithm has a good denoising effect

but in a low signal-to-noise ratio environment

there is still a large amount of residual noise. For this

the method that combining signal subspace optimization andwiener filtering is put forward to improve the speech quality. Simulation results show that underseveral common background noises with low signal-to-noise ratio

the performance of the proposedapproach ismore excellent

it can effectively restrain background noise and improve the speech quality

meanwhile

in some extent

guarantee the speech intelligibility

compared with the traditional spectral-subtraction

wiener filtering method

buit-in pre-whitening subspacealgorithm.

关键词

Keywords

references

J Ma, Yi Hu, Loizou P C. Obiective measures for predi-cting speech intelligibility in noisy conditions based on n-ew band inportance functions[J]. The Journal of the Acou-stical society of America, 2009, 125(5): 3387-3405. DOI: 3387

Yi Hu, Loizou P C. A generalized subspace approach for enhancing speech corruptedby colordnoise[J]. IEEE Trans on Speech and Audio Processig, 2003, 11(4): 334-341. DOI: 334

Ephraim Y, Van Trees H L. A signal subspaceapproach for speech enhancemment[J]. IEEE Transon Speech andAudio Proc, 1995, 3(4): 251-266. DOI: 251

Boll S F. Suppression of acoustic noise in speech using spectral subtraction[J]. IEEE Trans on Acoustics, Speechand Signal Processing, 1979, 27(2): 113-120. DOI: 113

Loizou P C, Kim G. Reasons why current speech enh-ancement algorithms do not improve speech intelligibility and suggested solutions[J]. IEEE Trans on Audio, Speech and Language Processing, 2011, 19(1): 47-56. DOI: 47

Denrinos M, Bakamidis S, Carayanni G. Speechen hancement from noise: a regenerative approah[J]. Speech Communi cation, 1991, 10(2): 45-57. DOI: 45

Saadoune A, Amrouche A, Selouani S. A perceptual s-ubspace speech enhancement using variance of the reco nstruction error[J]. Digtal Signal Proc, 2014, 24(1): 187-196. DOI: 187

Rezayee A, Gsxot S. An adaptive KLT approach for s-peech enhancement[J]. IEEE Trans on Speech and Audio Processing, 2001, 9(2): 87-95. DOI: 87

Chang Jing, Li Zhibin. The generalized inverse eigen-value problem for generalized perodic jacobi matrices[C]//Proceedings of International Conference on Computer Design and Applications. Washington DC, USA: IEEE Press, 2010: 354-356.

武奕峰, 贾海荣, 郭欣. 用子空间提高有色噪声背景下的语音质量[J]. 太原理工大学学报, 2015, 46(2): 192-194. DOI: 192

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种基于特征上下文理解与切片分析的肺结核检测算法

V型沟槽SiC中子探测器的蒙特卡洛分析

面向处理器分支预测单元的Spectre攻击轻量化防御方法

基于改进RT-DETR的航拍小目标检测算法

基于区块链的AI生成内容溯源框架设计研究