[1]郑瑞,李平,曾毓敏,等.基于DWT的AMDF基音检测改进算法[J].南京师范大学学报(工程技术版),2014,14(02):049.
　Zheng Rui,Li Ping,Zeng Yumin,et al.Modified AMDF Pitch Detection Algorithm Based on DWT[J].Journal of Nanjing Normal University(Engineering and Technology),2014,14(02):049.
点击复制

基于DWT的AMDF基音检测改进算法
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
14卷
期数:
2014年02期
页码:
049
栏目:
出版日期:
2014-06-30

文章信息/Info

Title:
Modified AMDF Pitch Detection Algorithm Based on DWT
作者:
郑瑞1; 李平2; 曾毓敏1; 高闻1; 李梦超1
(1.南京师范大学物理科学与技术学院,江苏 南京 210023) (2.泰州职业技术学院信息工程学院,江苏 泰州 225300)
Author(s):
Zheng Rui1; Li Ping2; Zeng Yumin1; Gao Wen1; Li Mengchao1
(1.School of Physics and Technology,Nanjing Normal University,Nanjing 210023,China) (2.Department of Information Engineering,Taizhou Vocational and Technical University,Taizhou 225300,China)
关键词:
基音; 离散小波变换; 平均幅度差函数; 下降趋势
Keywords:
pitch; discrete wavelet transform(DWT); average magnitude difference function(AMDF); falling trend
分类号:
TP391.42;TN912
文献标志码:
A
摘要:
平均幅度差函数(AMDF)法是一种常用的基音检测算法,但由于AMDF存在严重的均值下降趋势,使得用这种算法在提取基音的过程中易产生错误.为克服该缺点,提出了一种基于离散小波变换(DWT)的AMDF改进算法(DWAMDF),该算法首先求取每帧浊音信号的AMDF,利用DWT提取出AMDF中的下降趋势,去除下降趋势后得到DWAMDF用于检测基音.实验结果表明,与传统的AMDF算法及其几种典型的改进算法相比,该算法的检测精度有了明显的提高.
Abstract:
As a commonly used pitch detection algorithm,Average Magnitude Difference Function(AMDF)is used to estimate the pitch from the speech signal effectively.However,owing to the falling trend of the AMDF curve,it is easy to make the estimated pitch erroneous when AMDF algorithm is performed.In order to overcome the drawback,this paper presents a modified AMDF algorithm based on discrete wavelet transform(DWT)called DWAMDF.In this algorithm,conventional AMDF of each voiced frame is firstly calculated.Then the falling trend of AMDF is extracted by DWT.Finally DWAMDF is obtained by removing the falling trend from AMDF and the true pitch can be detected from DWAMDF.Lots of experiments show that the performance of the proposed DWAMDF outperforms AMDF and other AMDFbased modified algorithms.

参考文献/References:

[1]Li R,Bao C,Dou H.Pitch detection method for noisy speech signals based on prefilter and weighted wavelet coefficients[C]//2008 9th International Conference on Signal Processing.Beijing:IEEE Press,2008:530-533.
[2]Rabiner L.On the use of autocorrelation analysis for pitch detection[J].IEEE Transactions on Acoustics,Speech and Signal Processing,1977,25(1):24-33.
[3]Ross M,Shaffer H,Cohen A,et al.Average magnitude difference function pitch extractor[J].IEEE Transactions on Acoustics,Speech and Signal Processing,1974,22(5):353-362.
[4]Kadambe S,Bourdeauxbartels G F.A comparison of a wavelet functions for pitch detection of speech signals[C]//1991 International Conference on Acoustics,Speech,and Signal Processing.Toronto:IEEE Press,1991:449-452.
[5]Nadeu C,Pascual J,Hernando J.Pitch determination using the cepstrum of the onesided autocorrelation sequence[C]//1991 International Conference on Acoustics,Speech,and Signal Processing.Toronto:IEEE Press,1991:3 677-3 680.
[6]Rahman M S,Tanaka H,Shimamura T.Pitch determination using aligned AMDF[C]//Ninth International Conference on Spoken Language Processing.Pittsburgh:DUMMY Press,2006:1 714-1 717.
[7]顾良,刘润生.高性能汉语语音基音周期估计[J].电子学报,1999,27(1):8-11. Gu Liang,Liu Runsheng.Highperformance mandarin pitch estimation[J].Chinese Journal of Electronics,1999,27(1):8-11.(in Chinese)
[8]Zhang W,Xu G,Wang Y.Pitch estimation based on circular AMDF[C]//2002 IEEE International Conference on Acoustics,Speech,and Signal Processing(ICASSP).Orlando:IEEE Press,2002:I-341-I-344.
[9]Muhammad G.Noise robust pitch detection based on extended AMDF[C]//2008 IEEE International Symposium on Signal Processing and Information Technology.Sarajevo:IEEE Press,2008:133-138.
[10]宗源,曾毓敏,孙永熙,等.基于EMD的AMDF基音检测改进算法[J].南京师范大学学报:工程技术版,2013,13(1):62-67.
Zong Yuan,Zeng Yumin,Sun Yongxi,et al.A modified AMDF pitch detection algorithm based on EMD[J].Journal of Nanjing Normal University:Engineering and Technology Edition,2013,13(1):62-67.(in Chinese)
[11]Burrus C S,Gopinath R A,Guo H.Introduction to Wavelets and Wavelet Transforms:A Primer[M].Beijing:China Machine Press,2005.
[12]Pavel K.Discrete wavelet transform for automatic speaker recognition[C]//2010 3rd International Congress on Image and Signal Processing(CISP).Yantai:IEEE Press,2010:3 514-3 518.
[13]Plante F.A pitch extraction reference database[J].Children,1995,8(12):30-50.
[14]Rabiner L,Cheng M,Rosenberg A,et al.A comparative performance study of several pitch detection algorithms[J].IEEE Transactions on Acoustics,Speech and Signal Processing,1976,24(5):399-418.
[15]Chavan M S,Mastorakis N,Chavan M N,et al.Implementation of SYMLET wavelets to removal of Gaussian additive noise from speech signal[C]//Proceeding of 10th WSEAS International Conference on Signal Processing,Robotics and Automation.Wisconsin:ISBN Press,2011:37-41.
[16]Ntalampiras S,Fakotakis N.Speech/Music Discrimination Based on Discrete Wavelet Transform[M]//Artificial Intelligence:Theories,Models and Applications.Berlin Heidelberg:Springer,2008:205-211.
[17]Li R,Bao C,Dou H.Pitch detection method for noisy speech signals based on prefilter and weighted wavelet coefficients[C]//9th International Conference on Signal Processing(ICSP 2008).Beijing:IEEE Press,2008:530-533.
[18]Hariharan M,Fook C Y,Sindhu R,et al.A comparative study of wavelet families for classification of wrist motions[J].Computers and Electrical Engineering,2012,38(6):1 798-1 807.
[19]Ranjan S.A discrete wavelet transform based approach to hindi speech recognition[C]//2010 International Conference on Signal Acquisition and Processing(ICSAP’10).Bangalore:IEEE Press,2010:345-348.

相似文献/References:

[1]宗 源,曾毓敏,孙永熙,等.基于EMD的AMDF基音检测改进算法[J].南京师范大学学报(工程技术版),2013,13(01):062.
　Zong Yuan,Zeng Yumin,Sun Yongxi,et al.A Modified AMDF Pitch Detection Algorithm Based on EMD[J].Journal of Nanjing Normal University(Engineering and Technology),2013,13(02):062.

备注/Memo

备注/Memo:
收稿日期:2013-09-03.
基金项目:江苏省自然科学基金(BK2010546).
通讯联系人:曾毓敏,教授,研究方向:语音信号处理与图像处理.E-mail:zengyumin@njnu.edu.cn
更新日期/Last Update: 2014-06-30