一种用于区分语音与冲击噪声的改进算法  

An Improved Algorithm for Distinguishment Speech from Impact Noise

作　　者：李荣荣[1] 胡昌奎[1] 余娟[1] 

机构地区：[1]武汉理工大学理学院,湖北武汉430070

出　　处：《武汉理工大学学报（信息与管理工程版）》2013年第6期812-815,共4页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：中央高校基本科研业务费专项基金资助项目(2011-Ia-005)

摘　　要：为了将语音与冲击噪声进行准确地区分,讨论了传统检测方法的缺陷,选取汽车噪声为冲击噪声,结合语音浊音信号的周期性以及短时平均幅度差这一时域特征参数,对语音和汽车冲击噪声进行了分析,并在此基础上对算法进行改进,提出了一种新的检测参数。经Matlab仿真结果显示,新参数能够在高信噪比的环境下区分出语音与冲击噪声,为后续的语音处理奠定了基础。In order to distinguish voice from impact noise accurately,the disadvantage of traditional detection methods was discussed and an improved algorithm with a new parameter was combined with the voice periodicity and short-time average magnitude difference function to detect speech and car impact noise.The results show that the new method can make a distinction between speech and impact noise in a quiet environment.It will lay a foundation for the subsequent speech work.

关 键 词：语音检测 冲击噪声 浊音 短时平均幅度差 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]