基于子带谱熵的仿生小波语音增强

doi:10.11772/j.issn.1001-9081.2015.03.868

计算机应用 ›› 2015, Vol. 35 ›› Issue (3): 868-871.DOI: 10.11772/j.issn.1001-9081.2015.03.868

基于子带谱熵的仿生小波语音增强

刘艳, 倪万顺

大连大学信息工程学院, 辽宁大连 116622

收稿日期:2014-10-17 修回日期:2014-11-25 出版日期:2015-03-10 发布日期:2015-03-13
通讯作者: 刘艳
作者简介:刘艳(1967-),女,辽宁大连人,副教授,博士,主要研究方向:电力传动系统建模、故障诊断与容错控制;倪万顺(1989-),男,江苏泰州人,硕士研究生,主要研究方向:模式识别、语音信号处理
基金资助:
辽宁省教育厅科学计划项目(L2013463)

Speech enhancement based on bionic wavelet transform of subband spectrum entropy

LIU Yan, NI Wanshun

College of Information Engineering, Dalian University, Dalian Liaoning 116622, China

Received:2014-10-17 Revised:2014-11-25 Online:2015-03-10 Published:2015-03-13

摘要/Abstract

摘要：

前端噪声处理直接关系着语音识别的准确性和稳定性,针对小波去噪算法所分离出的信号不是原始信号的最佳估计,提出一种基于子带谱熵的仿生小波变换(BWT)去噪算法。充分利用子带谱熵端点检测的精确性,区分含噪语音部分和噪声部分,实时更新仿生小波变换中的阈值,精确地区分出噪声信号小波系数,达到语音增强目的。实验结果表明,提出的基于子带谱熵的仿生小波语音增强方法与维纳滤波方法相比,信噪比(SNR)平均提高约8%,所提方法对噪声环境下语音信号有显著的增强效果。

关键词: 语音增强, 子带谱熵, 仿生小波变换, 去噪, 阈值

Abstract:

Front end noise processing has a direct impact upon the accuracy and stability of the speech recognition. According to the fact that the signal separated by wavelet denoising algorithm isn't its optimal estimation, a novel Bionic Wavelet Transform (BWT) de-noising algorithm based on subband spectrum entropy was proposed. To achieve the purpose of speech enhancement, the subband spectrum entropy, which has a good accuracy of the endpoint detection, was taken full advantage to distinguish the parts of speech and noise, to real-timely update the threshold of BWT, and to precisely determine the noise signal wavelet coefficients. The experimental results indicate that the Signal-to-Noise Ratio (SNR) of the proposed algorithm is 8% higher than the Wiener filter algorithm. The proposed method has significant enhancement effect on speech signal in noisy environments.

Key words: speech enhancement, subband spectrum entropy, Bionic Wavelet Transform (BWT), denoising, threshold

中图分类号:

TN912.3

刘艳, 倪万顺. 基于子带谱熵的仿生小波语音增强[J]. 计算机应用, 2015, 35(3): 868-871.

LIU Yan, NI Wanshun. Speech enhancement based on bionic wavelet transform of subband spectrum entropy[J]. Journal of Computer Applications, 2015, 35(3): 868-871.

参考文献

[1] LI F. Based on wavelet transform speech recognition of research and application[D]. Guilin: Guilin University of Technology, 2011.(李法强. 基于小波变换的语音识别的研究和应用[D].桂林:桂林理工大学,2011.)
[2] DAI Y, ZHANG J. Reduction of 1/f noise in semiconductor devices based on wavelet transform and Wiener filter[J]. Acta Physica Sinica, 2011,60(11):185-190.(代煜,张建勋. 基于小波变换和维纳滤波的半导体器件1/f噪声滤波[J].物理学报,2011,60(11):185-190.)
[3] GENG Y, ZHANG D. Survey of adaptive filtering algorithms[J]. Information and Electroic Engineering, 2008,6(4):315-320.(耿妍,张端金. 自适应滤波算法综述[J].信息与电子工程, 2008,6(4):315-320.)
[4] ZHANG T, HE L, HUANG H, et al. Speech endpoint detection based on critical band and energy entropy[J]. Journal of Computer Applications, 2013,33(1):175-178.(张婷,何凌,黄华,等.基于临界频带及能量熵的语音端点检测[J].计算机应用,2013,33(1):175-178.)
[5] DONOHO D L, JOHNSTONE I M. Adapting to unknown smoothness via wavelet shrinkage[J]. Journal of the American Statistical Association, 1995, 90(432):1200-1224.
[6] REN Y, ZHANG X, JIA H. New threshold function of wavelet packet speech enhancement algorithm[J]. Application Research of Computers, 2013,30(1):114-116.(任永梅,张雪英,贾海荣. 一种新阈值函数的小波包语音增强算法[J].计算机应用研究,2013,30(1):114-116.)
[7] YAO J, ZHANG Y T. Bionic wavelet transform: a new time-frequency method based on all auditory model[J]. IEEE Transactions on Biomedical Engineering, 2001,48(8):856-863.
[8] SHEN J-L, HUNG J-W, LEE L-S. Robust entropy-based endpoint detection for speech recognition in noisy environments[EB/OL].[2014-08-12]. http://wenku.baidu.com/view/e676ae0979563c1ec 5da71d5.html.
[9] LI Y, WANG T, CUI H, et al. Voice activity detection in non-stationary noise[EB/OL].[2014-09-01]. http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=4281886&url=http%3A%2F%2Fieeexplore.ieee.org%2Fstamp%2Fstamp.jsp%3Ftp%3D%26arnumber%3D4281886.
[10] WU B, WANG K. Robust endpoint detection algorithm based on the adaptive band-partitioning spectral entropy in adverse environments[J]. IEEE Transactions on Speech and Audio Processing, 2005,13(5):762-775.
[11] LI R, HU C, YU J. Research of speech endpoint detection based on spectral entropy algorithm[J]. Journal of Wuhan University of Technology, 2013,35(7):134-139.(李荣荣,胡昌奎,余娟.基于谱熵的语音端点检测算法改进研究[J].武汉理工大学学报.2013,35(7):134-139.)
[12] YANG X, FAN X. Speech enhancement based on bionic wavelet transform and adaptive threshold functions[J]. Control and Decision, 2006,21(9):1033-1036.(杨玺,樊晓平.基于仿生小波变换和自适应阈值的语音增强方法[J].控制与决策,2006,21(9):1033-1036.)
[13] CHEN X, WANG Y, LIU L. Deep study on wavelet threshold method for image noise removing[J]. Laser and Infrared, 2012,42(1):3-5.(陈小曦,王延杰,刘恋. 小波阈值去噪法的深入研究[J].激光与红外,2012,42(1):3-5.)

[1]	刘晓光, 靳少康, 韦子辉, 梁铁, 王洪瑞, 刘秀玲. 基于阈值和极端随机树的实时跌倒检测方法[J]. 计算机应用, 2021, 41(9): 2761-2766.
[2]	贾鹤鸣, 李瑶, 姜子超, 孙康健. 基于改进共生生物搜索算法的林火图像多阈值分割[J]. 计算机应用, 2021, 41(5): 1465-1470.
[3]	卓禹心, 韩素雅, 张榆锋, 李支尧, 董毅峰. 基于超声谐波包络Nakagami参数图像的微波消融区域自动分割方法[J]. 计算机应用, 2021, 41(10): 3089-3096.
[4]	王思鹏, 杜昌平, 叶志贤, 宋广华, 郑耀. 基于多步神经网络观测器的扑翼飞行器缓变故障检测[J]. 计算机应用, 2020, 40(8): 2449-2454.
[5]	龙超, 曾庆宁, 罗瀛. 基于噪声抵消与波束形成的小阵语音增强[J]. 计算机应用, 2020, 40(8): 2386-2391.
[6]	胡雪影, 郭海儒, 朱蓉. 基于混合深度卷积网络的图像超分辨率重建[J]. 计算机应用, 2020, 40(7): 2069-2076.
[7]	曹建芳, 赵爱迪, 张自邦. 融合阈值寻优的卷积神经网络在图像标注中的应用[J]. 计算机应用, 2020, 40(6): 1587-1592.
[8]	张皓然, 王学渊, 李小霞. 基于自适应阈值活动语音检测和最小均方误差对数谱幅度估计的低信噪比降噪算法[J]. 计算机应用, 2020, 40(6): 1763-1768.
[9]	吴庆贺, 吴海锋, 沈勇, 曾玉. 工业噪声环境下多麦状态空间模型语音增强算法[J]. 计算机应用, 2020, 40(5): 1476-1482.
[10]	王忠锋, 徐志远, 宋纯贺, 张宏宇, 蔡颖凯. 基于梯度的深度网络剪枝算法[J]. 计算机应用, 2020, 40(5): 1253-1259.
[11]	余敦辉, 袁旭, 张万山, 王晨旭. 基于动态阈值的时空众包在线分配算法[J]. 计算机应用, 2020, 40(3): 658-664.
[12]	王永彪, 张文喜, 王亚慧, 孔新新, 吕彤. 拉普拉斯分布下的MMSE谱减语音增强算法[J]. 计算机应用, 2020, 40(3): 878-882.
[13]	杨立波, 蒋铁钢, 徐志强. 基于混合梯度的硬阈值追踪算法[J]. 计算机应用, 2020, 40(3): 912-916.
[14]	傅洪亮, 雷沛之. 基于去噪自编码器和长短时记忆网络的语音测谎算法[J]. 计算机应用, 2020, 40(2): 589-594.
[15]	王海鹏, 降爱莲, 李鹏翔. 牛顿-软阈值迭代鲁棒主成分分析算法[J]. 计算机应用, 2020, 40(11): 3133-3138.

基于子带谱熵的仿生小波语音增强

Speech enhancement based on bionic wavelet transform of subband spectrum entropy

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics