基于小波谱失真测度的语音质量评估

兵工学报 ›› 2008, Vol. 29 ›› Issue (1): 33-36.

基于小波谱失真测度的语音质量评估

王继祥^1,2，刘贵忠¹，刘广健²，高向东²

1．西安交通大学电子与信息学院，陕西西安710049;2．中国人民解放军63880部队，河南洛阳471003

收稿日期:2006-08-23 上线日期:2014-12-25
通讯作者: 王继祥 E-mail:wjxfy2002@163．Com
作者简介:王继祥（1965-），男，高级工程师，博士研究生。

Evaluation of Speech Quality Based on Wavelet Spectrum Distortion Measure

WANG Ji-xiang^1,2, LIU Gui-zhong¹，LIU Guang-jian²,GAO Xiang-dong²

1.School of Electronics and Information Engineering, Xi’an Jiaotong University, Xi’an 710049, Shaanxi, China; 2.Unit 63880, PLA, Luoyang 471003, Henan, China

Received:2006-08-23 Online:2014-12-25
Contact: WANG Ji-xiang E-mail:wjxfy2002@163．Com

摘要/Abstract

摘要： 语音质量评估常用于通信对抗装备试验中，但现有客观评估方法无法满足试验的需要，限制了它的适用领域。通过分析小波理论与客观评估手段，提出了利用小波谱失真测度(WSDM)客观评估语音信号通信质量的方法，定义了小波谱失真测度，并给出了小波谱失真测度的计算公式，实验结果表明，与MFCC谱失真测度和BARK谱失真测度相比，用小波谱失真测度客观评价语音信号的通信质量可以提高主客观评估的相关度。

关键词: 人工智能 , 小波谱失真测度 , 语音信号 , 通信质量评估

Abstract: The evaluation of speech quality is commonly used in the communication countermeasure equipment testing. Its application to the testing is limited due to the inefficiency of existent objective evaluation method. The method of evaluating communication quality with the wavelet spectrum distor?tion measure (WSDM) was proposed after the analysis of wavelet theory and objective evaluation means. The WSDM was defined and the correspondent calculation formula was presented. The testing results show that the correlation between subjective and objective evaluations is improved by the use of the method of evaluating communication quality with WSDM compared with MFCC and BARK spec?trum distortion measure.

Key words: artificial intelligence , wavelet spectrum distortion measure(WSDM) , speech signal , com?munication quality evaluation

中图分类号:

TN97

王继祥，刘贵忠，刘广健，高向东. 基于小波谱失真测度的语音质量评估[J]. 兵工学报, 2008, 29(1): 33-36.

WANG Ji-xiang, LIU Gui-zhong，LIU Guang-jian,GAO Xiang-dong. Evaluation of Speech Quality Based on Wavelet Spectrum Distortion Measure[J]. Acta Armamentarii, 2008, 29(1): 33-36.

参考文献

[I] 吴淑珍，Pols LCW.汉语通信系统客观音质评价方法[J].声学学报，1998, 23(2):170-174.
WU Shu-zhen, Pois L C W. A measure for objective speech quali?ty evaluation of しhmese communication channels [ J] ? Acta Acusti-
ca, 1998，23(2): 170 — 174? (in Chinese)
[2] Picone J W- Signal modeling techniques in speech recognition [C] //Proceedings of IEEE, 1993, 81:1215 - 1247.
[3] Kubichek R. Melcepstral distance measure for objective speech quality assessment[ C] // IEEE Pacific Rim Conference on Com?munications, Computers, and Signal Processing, 1993, I ： 125 — 128.
[4] Wang S, Sekey A, bersho A. An objevtive measure for predict?ing subjective quality of speech coders [ J ]. IEEE Journal on Se?lected Areas in Communications，1992,10(5) ： 819 一 829.
[5] Johnson Ihyeh Agbinya. Discrete wavelet transform techniques in speech processing [ C ] // IEEE Tencon-Digital Signal Processing Applications, 1996 ： 514 — 519.
[6] 李云鸿，胡修林，张蕴玉.基于人耳听觉模型的语音质量客观评价方法[J].华中理工大学学报,2000，28(5):63-65.
LI Yun-hong, HU Xiu-lin, ZHANG Yun-yu. Objective evalua?tion method of speech quality based on human auditory model[J]. Journal of Huazhong University of Science and Technology, 2000, 28(5):63-65. (in Chinese)
[7] 杨震，毕厚杰.一种新的用于语音主观质量评价的谱失真參数[J].电子信息学报，2001，23(7)： 669-676.
YANG Zhen, BI Hou-jie. A new paramter of spectral distortion for predicting subjective quality of speech [J] ? Journal of Elec?tronics and Information Technology, 2001, 23(7): 669 一 676. (in Chinese)
[8] 易克初，田斌，付强.语音信号处理[M].北京:国防エ业出版社，2000： 137-145.
YI Ke-chu, TIAN Bin, FU Qiang. Speech signal processing [M]. Beijing： National Defence Industry Press, 2000： 137 — 145. (in chinese)
[9] 朱学芳，徐建平.计算机语音信号处理与语音识别系统[J]. 南京邮电学院学报，1998，18(5)： 113-119.
ZHU Xue-fang, Xu Jian-ping. A speech processing and phonetic recognition system[ J ]. Journal of Nanjing Institute of Posts and Telecommunications，1998，18(5) ： 113 - 119. (in Chinese)
[10] Ta-Hsin Li, Hee-Seok Oh. Wavelet spectrum and its characteri?zation property for random property for random processes [ J ]. IEEE Transactions on Information Theory, 2002, 48 (11): 2922-2937.
[11] 张明新，沈钧毅，周耿烈，等.ニ维非平稳信号的小波谱均衡分析方法[J].兰州大学学报：自然科学版，2004, 40(3): 30-33.
ZHANG Ming-xin, SHEN Jun-yi, ZHOU Geng-lie, et al. Spec?trum equalization based on wavelet transtorm for 2D non-station- ary signal[j]. Journal of Lanzhou University： Natural Sciences, 2004，40(3) ： 30 — 33. (in Chinese)
[12] Kidae K，Dae H Y, Chulhee L. Evaluation of wavelet filters for speech recognition[J]. IEEE，2000： 2891 — 2894.
[13] Stephance Mallat.信号处理的小波导引[M].杨カ华，戴道清，黄文良，等，译.北京:机械エ业出版社，2003:111-112. Stephane Mallat. A wavelet tour of signal processing [ J ] ? YANG Li-hua, DAI Dao-qing，HUANG Wen-liang, et al, translated. Beijing： China Machine Press,2003:111 - 112. (in Chinese)

[1]	刘昊，张策，丁文韬. 基于智能对抗进化的联合火力打击任务规划方法[J]. 兵工学报, 2019, 40(6): 1287-1296.
[2]	栗大鹏, 梁伟. 针对非合作目标的自适应网格聚类算法[J]. 兵工学报, 2017, 38(11): 2166-2175.
[3]	王儒，王国新，阎艳，郝佳，石学海，明振军. 支持产品智能化设计的知识服务方法研究[J]. 兵工学报, 2016, 37(11): 2101-2113.
[4]	焦玉民1,2, 王强1, 徐婷1, 谢庆华2, 王海涛1. 基于GA-Q-learning 算法的虚拟维修作业规划模型[J]. 兵工学报, 2013, 34(5): 627-633.
[5]	董文1，方向1，范磊1，杨力1，雷志刚. 基于斯坦纳树的雷场网络大面积损坏修复策略[J]. 兵工学报, 2013, 34(2): 197-202.
[6]	陈方涵，张肃，王文生. Curvelet变换在低对比度目标识别中的应用[J]. 兵工学报, 2012, 33(6): 688-694.
[7]	徐超，王长龙，孙世宇，陈鹏，绳慧. 双小波神经网络迭代的漏磁缺陷轮廓重构技术[J]. 兵工学报, 2012, 33(6): 730-735.
[8]	杨博，王向华，邵利平，覃征，于维虎. 基于群集智能的传感器管理方法研究[J]. 兵工学报, 2012, 33(2): 155-162.
[9]	宋小杉，蒋晓瑜，罗建华，姚军. 基于类间距的径向基函数-支持向量机核参数评价方法分析[J]. 兵工学报, 2012, 33(2): 203-208.
[10]	康一梅，谢晚冬，胡江，黄琪. 目标尺度自适应的Mean Shift跟踪算法[J]. 兵工学报, 2011, 32(2): 210-216.
[11]	王红霞，潘红兵，叶晓慧. 多故障的测试序列问题研究[J]. 兵工学报, 2011, 32(12): 1518-1523.
[12]	彭颖，沈怀荣，马永一. 一种新的冲突证据融合方法[J]. 兵工学报, 2011, 32(1): 78-84.
[13]	于海全, 彭高亮, 刘文剑. 基于虚拟环境的维修性信息模型的建立[J]. 兵工学报, 2010, 31(7): 998-1002.
[14]	杨日杰₁，吴芳₁，徐俊艳₁，高青伟₂. 基于马尔可夫过程的水下运动目标启发式搜索[J]. 兵工学报, 2010, 31(5): 586-591.
[15]	童仲志₁，张媛₁，邢宗义₁，秦勇₂，贾利民₂. 某装备扫雷犁系统的遗传神经网络建模研究[J]. 兵工学报, 2010, 31(5): 650-656.