[1]詹 展,秦会斌.基于新阈值函数的小波阈值去噪算法[J].计算机技术与发展,2019,29(11):47-51.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 010]
　ZHAN Zhan,QIN Hui-bin.A Wavelet Threshold Denoising Algorithm Based on New Threshold Function[J].,2019,29(11):47-51.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 010]
点击复制

基于新阈值函数的小波阈值去噪算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年11期
页码:
47-51
栏目:
智能、算法、系统工程
出版日期:
2019-11-10

文章信息/Info

Title:
A Wavelet Threshold Denoising Algorithm Based on New Threshold Function
文章编号:
1673-629X(2019)11-0047-05
作者:
詹 展; 秦会斌
杭州电子科技大学 新型电子器件与应用研究所,浙江 杭州 310018
Author(s):
ZHAN Zhan; QIN Hui-bin
Institute of New Electronic Device &Application,Hangzhou Dianzi University,Hangzhou 310018,China
关键词:
语音增强; 阈值函数; 去噪; 小波
Keywords:
speech enhancement; threshold function; denoising; wavelet
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 11. 010
摘要:
在小波阈值语音去噪中,阈值函数设定直接决定着语音增强的去噪效果。 以 Donoho 等最先提出的软、硬阈值函数为基础,分析了该算法存在不连续、产生固定偏差的缺点,以及文中基于该算法做出改进的现有阈值函数仍存在不足。为了进一步提高去噪性能,文中提出了一个改进的带参数的阈值函数。 该阈值函数不仅克服了传统函数的缺点,而且具有更好的灵活性。 并提出了一种算法来选取最优分解层数,最大程度消除噪声,保留有用信号的信息。 将带噪语音信号的小波系数经过该阈值函数的处理、重构后,得到增强的语音信号。 仿真与真实环境的实验结果表明,该阈值函数与传统阈值函数以及现有的阈值函数去噪相比,输出信噪比和均方误差等性能指标均得到了提升;通过观察时域波形,可以看出该阈值函数处理后的效果更接近原信号,提高了增强语音后的可懂度与整体质量。
Abstract:
In the wavelet threshold speech denoising,the threshold function setting directly determines the denoising effect of speech enhancement. Based on the soft and hard threshold functions proposed by Donoho et al,the shortcomings of discontinuity and fixed deviation of the algorithm are analyzed,and the existing threshold functions based on the algorithm are still insufficient. In order to further improve the denoising performance,an improved threshold function with parameters is proposed,which not only overcomes the shortcomings of traditional functions,but also has better flexibility. And an algorithm is proposed to select the optimal number of decomposition layers to eliminate noise and preserve the information of useful signals. After the wavelet coefficients of the noisy speech signal are processed and reconstructed by the threshold function,an enhanced speech signal is obtained. The results of simulation and real-world experiments show that compared with the traditional threshold function denoising and the existing threshold function denoising,the new threshold function has improved output signal-to-noise ratio and mean-square error. By bserving the time domain waveform,it can be seen that the effect of the new threshold function is closer to the original signal,which improves the intelligibility and overall quality of the enhanced speech.

相似文献/References:

[1]丁伟 吴小培.基于改进谱减方法的语音增强研究[J].计算机技术与发展,2008,(09):98.
　DING Wei,WU Xiao-pei.Implementation of Speech Enhancement Based on Improved Spectral Subtraction[J].,2008,(11):98.
[2]汝振 李昕 陈飞 杨李箭 李翔.一种基于HHT的语音增强算法研究与仿真[J].计算机技术与发展,2010,(08):116.
　RU Zhen,LI Xin,CHEN Fei,et al.Research and Simulation of Speech Enhancement Algorithm Based on HHT[J].,2010,(11):116.
[3]周述畅 宋亚男 吴光波.基于麦克风阵列的语音增强研究[J].计算机技术与发展,2012,(07):204.
　ZHOU Shu-chang,SONG Ya-nan,WU Guang-bo.Speech Enhancement Research Based on Microphone Array[J].,2012,(11):204.
[4]白青,王军锋,王涛,等.一种改进的小波阈值图像去噪综合模型[J].计算机技术与发展,2013,(05):50.
　BAI Qing,WANG Jun-feng,WANG Tao,et al.An Improved Comprehensive Model of Wavelet Threshold for Image Denoising[J].,2013,(11):50.
[5]李克粉,王直.改进的小波阈值去噪在语音识别中的应用[J].计算机技术与发展,2013,(05):231.
　LI Ke-fen,WANG Zhi.Application of Improved Wavelet Threshold Denoising in Speech Recognition[J].,2013,(11):231.
[6]陈欢,邱晓晖.改进谱减法语音增强算法的研究[J].计算机技术与发展,2014,24(04):69.
　CHEN Huan,QIU Xiao-hui.Research on Speech Enhancement of Improved Spectral Subtraction Algorithm[J].,2014,24(11):69.
[7]李智,张根耀,王蓓. 基于一种新的阈值函数的小波图像去噪[J].计算机技术与发展,2014,24(11):100.
　LI Zh,ZHANG Gen-yao,WANG Bei. Wavelet Image Denoising Based on a New Threshold Function[J].,2014,24(11):100.
[8]董胡. 低信噪比环境下改进的语音端点检测算法[J].计算机技术与发展,2016,26(03):71.
　DONG Hu. Improved Speech Endpoint Detection under Low SNR Environment[J].,2016,26(11):71.
[9]成谢锋,姚鹏飞.基于双阈值的心音快速分段算法及其应用研究[J].计算机技术与发展,2018,28(05):140.[doi:10．3969/j．issn．1673－629X．2018．05．032]
　CHENG Xiefeng,YAO Pengfei.A Fast Heart Sound Segmentation Based on Double Threshold and Its Application[J].,2018,28(11):140.[doi:10．3969/j．issn．1673－629X．2018．05．032]
[10]张建明,邱晓晖.基于Curvelet 变换的指纹图像去噪[J].计算机技术与发展,2018,28(05):164.[doi:10．3969/j．issn．1673－629X．2018．05．037]
　ZHANG Jianming,QIU Xiaohui.Fingerprint Image Denoising Based on Curvelet Transform[J].,2018,28(11):164.[doi:10．3969/j．issn．1673－629X．2018．05．037]

更新日期/Last Update: 2019-11-10