基于小波分析与Boll改进谱减法的普米语语音增强算法  

A Pumi-speech enhancement algorithm based on wavelet analysis and improved Boll's spectral subtraction

作　　者：解雪琴 张天军[1] 潘文林[1] 王璐 和丽华 杨建香 XIE Xue-qin;ZHANG Tian-jun;PAN Wen-lin;WANG Lu;HE Li-hua;YANG Jian-xiang(School of Mathematics and Computer Science, Yunnan Minzu University, Kunming 650500,China)

机构地区：[1]云南民族大学数学与计算机科学学院,云南昆明650500

出　　处：《云南民族大学学报（自然科学版）》2019年第3期306-312,共7页Journal of Yunnan Minzu University:Natural Sciences Edition

基　　金：国家自然科学基金(61363022);云南省教育厅科学研究基金(2017YJS056);云南民族大学研究生创新基金(2018YJCXS232)

摘　　要：小波分析和Boll改进谱减法具有很好的减噪效果,但同时也存在"音乐噪声"和语音失真.为了解决这2个问题,基于这2种算法,研究了一种它们的联合算法并对加噪后的普米语进行语音减噪实验.通过对减噪后语音的试听和波形图的观察以及信噪比的对比,对实验的结果进行分析.实验结果表明,针对叠加了随机高斯白噪声的普米语,算法能有效地减少这2种方法单独使用时所产生的"音乐噪声"和语音失真.Wavelet analysis and improved Boll s spectral subtraction have good noise reduction effects, but there are also “music noise” and speech distortion. In order to solve the two problems, this paper studies a joint algorithm based on these two algorithms and carries out speech denoising experiments on the Pumi language after noise addition. The experimental results are analyzed by comparing the audition and waveform of the noise-reduced speech and the comparison of the signal-to-noise ratio. The experimental results show that for the Pumi language with random Gaussian white noise superimposed, this joint algorithm can effectively reduce the “music noise” and speech distortion produced by the two methods when they are used separately.

关 键 词：普米语 小波分析 Boll的改进谱减法 

分 类 号：TN912.34[电子电信—通信与信息系统]