首页 | 本学科首页   官方微博 | 高级检索  
     

基于数学形态学的非线性语音增强方法
引用本文:王光艳,赵晓群,王霞.基于数学形态学的非线性语音增强方法[J].计算机工程与应用,2006,42(34):53-56.
作者姓名:王光艳  赵晓群  王霞
作者单位:1. 河北工业大学,信息学院,天津,300130;天津商学院,信息工程学院,天津,300134
2. 同济大学,电子与信息工程学院,上海,200092
3. 河北工业大学,信息学院,天津,300130
摘    要:为了更好地滤除语音信号中的宽带噪声,使语音增强向着非线性方向发展。综合考虑语音和噪声特性,选用合理的结构元素,设计了一种新的形态学滤波算法,构造出用于语音信号滤波增强的非线性形态滤波器。仿真实验和数据分析结果论证了这种基于形态学的非线性语音增强方法的可行性,尤其是对于正负脉冲噪声的滤除效果相对较好。

关 键 词:语音增强  非线性  形态学滤波  结构元素
文章编号:1002-8331(2006)34-0053-04
收稿时间:2006-03
修稿时间:2006-03

Non-linear Speech Enhancement Method Based on Mathematical Morphology
WANG Guang-yan,ZHAO Xiao-qun,WANG Xia.Non-linear Speech Enhancement Method Based on Mathematical Morphology[J].Computer Engineering and Applications,2006,42(34):53-56.
Authors:WANG Guang-yan  ZHAO Xiao-qun  WANG Xia
Affiliation:1.School of Information Engineering, Hebei University of Technology,Tianjin 300130, China; 2.School of Information Engineering, Tianjin University of Commerce,Tianjin 300134, China; 3.School of Electronics and Information Engineering,Tongji University, Shanghai 200092, China
Abstract:Non-linear method is the trend of speech enhancement,especially for the speech signal corrupted by wideband noise.In this paper,considering the properties both of the speech and noise signal,a non-llnear morphological filter is designed for suppressing the noises embedded in speech signal.This new tilter,known from the conventional ones, is based on a new morphological filtering algorithm with appropriate structuring element.Simulations results confirm the feasibility and superiority over removing the pulse noise of this new approach.
Keywords:speech enhancement  non-linear  morphological filtering  structuring element
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号