基于音节的汉语连续语音声调识别方法研究
|
摘要
声调是汉语的主要属性之一，具有构词、辨义和提高表达效果等功能，对语
音识别、语音合成和自然语言理解有重要意义。
近年来，自动语音识别研究取得了突破性的进展，出现了许多不同类型的语
音识别系统。语音识别研究也转向了大词汇非认人连续语音识别和自然语言理解。
现有的汉语语音识别系统基本上没有利用声调信息，声调识别研究也多限于孤立
字和多字词的声调识别，连续语音的声调模式和声调识别研究很少，本文在这方
面开展了一点工作。
汉语连续语音的声调识别比孤立字和多字词的声调识别更困难，本文提出了
基于音节的声调识别思想，研究了其中涉及的音节分割、声调获取、特征提取、
声调模式分析和声调识别模型等问题。论文的主要内容如下：
（1）利用分形理论和波形互相关性研究了汉语连续语音中的音节分割问题。
本文选音节做声调识别基元，这将引入音节切分问题。连续语流中的音节分割是
非常困难的。本文根据语音信号的混沌本质，利用分形理论研究了汉语连续语音
中的音节分割问题，提出了基于方差分形维数的音节分割方法，并详细分析了该
方法的性能，它能很好地解决了无声与有声、浊音与清音间的分割问题，但很难
解决浊音间的分割问题，当浊音相连且过渡段较短时，该方法无法实现它们之间
的分割。为解决浊音之间的分割问题，本文根据语音中过渡段与非过渡段语音波
形的差异，利用波形互相关性进行了研究，提出了基于波形互相关性的音节分割
方法，并进行了实例分析。
（2）基于小波变换的语音基频提取。声调是基频变化的模式，因此可通过基
频提取来获取声调信息。基频提取的方法很多，本文采用了小波变换方法，该方
法对部分语音得到了较好的结果，但对大部分语音提取的基频中含有较多错误，
经仔细分析和研究，本文对它进行了改进，提出了一种基于小波变换的语音基频
检测新算法。该算法根据基频点在小波变换的不同分辨率层具有传递性和在不同
尺度上的基频点位置相似的特性，采用投票策略选择基频点。该算法主要有以下
几步：计算在五个（或三个）尺度上的小波变换；运用投票机制进行基频点选择；
基频检查；基频点的重新定位。
（3）声调识别中的特征提取问题。特征提取是模式识别的基本问题。有效的
特征既能反映模式的重要信息，又可减少计算量和误识率。汉语声调主要由基音
曲线的调位高低和走向决定。因此，本文选择头尾差和相对调位比作为三字词音
节的声调特征；选择头尾差和音节起点调位作为连续语音中音节的声调特征。
合肥工业大学搏土论文 扬耍
（4）声调模式分析。连续语音中各音节的声调特征受前后音节的影响变化较
大，声调模式更加复杂，仅具有四声的基本特征。正确地分析其中的声调模式和
变调规则，对汉语连续语音的声调识别有重要意义。本文介绍了孤立字和二字词
的声调模式，定性和定量地分析了三字词的声调模式，在此基础上研究了连续语
音的声调模式。
（5）声调识别模型的选择与设计。汉语连续语音的声调模式复杂多变，一个
固定不变的识别模型不可能解决连续语音的声调识别问题。本文以具有在线学习
能力的模糊神经网络作为声调识别模型，提出了基于模糊自适应谐振理论映射的
声调识别方法。
（6）用三字词和连续语音实例验证了所提出的思想和方法。
论文中取得的研究成果如下：
门）根据汉语的特点，提出了基于音节的汉语连续语音声调识别思想。
（2）根据语音信号的混饨本质，提出了基于方差分形维数的音节分割方法；针对
浊音间的分割困难，提出了基于波形互相关性的音节分割方法。
（3）根据传统小波变换方法在基频检测实验中出现的问题，引入投票策略，提出
了一种基于小波变换的基频检测新算法。
（4）根据汉语声调曲线的特点，选择头尾差和相对调位比作为三字词各音节的声
调特征：选择头尾差和音节起点调位作为连续语音中音节的声调特征。
（5）定性和定量地分析了三字词的声调模式，印证了已有的三字词声调模式变化
规律，得到了一些新的三字词声调模式变调规则。对汉语连续语音的声调模
式，提出了以下观点：连续语音中的音节声调模式可以二字词和三字词的声
调模式为基础：连续语音中的音节可认为仅受前后音节的彤响，一定间隙前
后的音节声调可看成互不相关；连续语音中的音节声调模式可归为头、中和
尾三类，通过对这三类声调模式的建模，可解决连续语音的声调识别问题。
①）为了适应连续语音中的复杂情况，提出了以具有在线学习能力的模糊神经网
络作为声调识别模型的观点。在此基础上提出了基于模糊自适应谐振理论映
射的声调识别方法。
Tone is one of primary properties for Chinese. Its functions are listed as the following: constructing words, distinguishing semantic and improving expression effect. It is important to speech recognition, speech synthesis and natural language understanding.
In the recent years large progress has been made in speech auto-recognition; many voice speech systems have been developed. Now research on voice recognition turns to large vocabulary speaker-independent continuous speech recognition and natural language understanding. Tone information isn抰 basically used in current Chinese speech recognition system, study on tone recognition is limited to tone recognition of isolated-word and multi-syllabic word, and research on tone patterns and tone recognition for Chinese continuous speech is little.
In this dissertation a syllable-based method of tone recognition for continuous speech is proposed. This method includes the following procedures: syllable segmentation, pitch detection, feature extraction, tone pattern analysis, and tone recognition. Our research is show as follows:
(1) Syllable segmentation in continuous speech. Syllable is determined as tone recognition unit in this thesis, so syllable segmentation must be done. Syllable segmentation in continuous speech is very difficult. In accordance with chaotic essence of speech signals, syllable segmentation in continuous speech is researched by fractal theory. An approach of syllable segmentation using variance fractal dimension is proposed, its performance is analyzed in detail. The method can discriminate between voiced and unvoiced, between surd and sonant, but it can hardly discriminate between sonant. According to difference of speech waveform between transition segment and non-transition segment, dividing between sonant is researched using waveform cross correlation. A method of syllable segmentation is presented based on waveform cross-correlation.
(2) Pitch detection of speech signals. Mandarin tone is the patterns of pitch variation, so it may be acquired by pitch extraction. Many methods of pitch detection are developed so far, in this thesis the pitch detector using waveform transform is adopted. According to the problem appearing in pitch extraction experiment, a novel algorithm of pitch detection is presented. The pitch points in speech signal exhibit local maximum across several consecutive dyadic scales, and their positions are similar, so the improved approach selects pitch points by vote strategy, not by traditional method. Procedure of the new algorithm is as follows: (i) calculating the wavelet transform
III
2
Abstract
across 5 (or 3) consecutive scales; (ii) choosing pitch points by vote strategy; (iii) checking pitch points; (IV) relocation of pitch points.
(3) Feature extraction in tone recognition. Feature extraction is a basic problem in pattern recognition. Valid feature can reflect important information of pattern, and decrease computation and error recognition rate. Mandarin tone is characterized for tone level and tendency of pitch curve, so head-tail difference and relative tone level rate are determined as tone features for Chinese tn-syllable word; head-tail difference and tone level at the beginning of syllable are determined as tone features of syllable in continuous speech.
(4) Tone pattern analysis. Tone characteristics of syllable in continuous speech have lager variation than original tone characteristic under the influence of its preceding and posterior syllable, so tone patterns is more complicated, and have only basic characteristic of Chinese tone. Tone patterns and its variation rules are important to tone recognition for continuous speech. In this dissertation tone patterns of disyllabic word and isolated-word are introduced, tone patterns for tn-syllabic word are analyzed in detail. Tone patterns for continuous speech are researched based on the foregoing work.
(5) Selection and design of tone recognition module. Tone patterns of syllable in continuous speech
音识别、语音合成和自然语言理解有重要意义。
近年来，自动语音识别研究取得了突破性的进展，出现了许多不同类型的语
音识别系统。语音识别研究也转向了大词汇非认人连续语音识别和自然语言理解。
现有的汉语语音识别系统基本上没有利用声调信息，声调识别研究也多限于孤立
字和多字词的声调识别，连续语音的声调模式和声调识别研究很少，本文在这方
面开展了一点工作。
汉语连续语音的声调识别比孤立字和多字词的声调识别更困难，本文提出了
基于音节的声调识别思想，研究了其中涉及的音节分割、声调获取、特征提取、
声调模式分析和声调识别模型等问题。论文的主要内容如下：
（1）利用分形理论和波形互相关性研究了汉语连续语音中的音节分割问题。
本文选音节做声调识别基元，这将引入音节切分问题。连续语流中的音节分割是
非常困难的。本文根据语音信号的混沌本质，利用分形理论研究了汉语连续语音
中的音节分割问题，提出了基于方差分形维数的音节分割方法，并详细分析了该
方法的性能，它能很好地解决了无声与有声、浊音与清音间的分割问题，但很难
解决浊音间的分割问题，当浊音相连且过渡段较短时，该方法无法实现它们之间
的分割。为解决浊音之间的分割问题，本文根据语音中过渡段与非过渡段语音波
形的差异，利用波形互相关性进行了研究，提出了基于波形互相关性的音节分割
方法，并进行了实例分析。
（2）基于小波变换的语音基频提取。声调是基频变化的模式，因此可通过基
频提取来获取声调信息。基频提取的方法很多，本文采用了小波变换方法，该方
法对部分语音得到了较好的结果，但对大部分语音提取的基频中含有较多错误，
经仔细分析和研究，本文对它进行了改进，提出了一种基于小波变换的语音基频
检测新算法。该算法根据基频点在小波变换的不同分辨率层具有传递性和在不同
尺度上的基频点位置相似的特性，采用投票策略选择基频点。该算法主要有以下
几步：计算在五个（或三个）尺度上的小波变换；运用投票机制进行基频点选择；
基频检查；基频点的重新定位。
（3）声调识别中的特征提取问题。特征提取是模式识别的基本问题。有效的
特征既能反映模式的重要信息，又可减少计算量和误识率。汉语声调主要由基音
曲线的调位高低和走向决定。因此，本文选择头尾差和相对调位比作为三字词音
节的声调特征；选择头尾差和音节起点调位作为连续语音中音节的声调特征。
合肥工业大学搏土论文 扬耍
（4）声调模式分析。连续语音中各音节的声调特征受前后音节的影响变化较
大，声调模式更加复杂，仅具有四声的基本特征。正确地分析其中的声调模式和
变调规则，对汉语连续语音的声调识别有重要意义。本文介绍了孤立字和二字词
的声调模式，定性和定量地分析了三字词的声调模式，在此基础上研究了连续语
音的声调模式。
（5）声调识别模型的选择与设计。汉语连续语音的声调模式复杂多变，一个
固定不变的识别模型不可能解决连续语音的声调识别问题。本文以具有在线学习
能力的模糊神经网络作为声调识别模型，提出了基于模糊自适应谐振理论映射的
声调识别方法。
（6）用三字词和连续语音实例验证了所提出的思想和方法。
论文中取得的研究成果如下：
门）根据汉语的特点，提出了基于音节的汉语连续语音声调识别思想。
（2）根据语音信号的混饨本质，提出了基于方差分形维数的音节分割方法；针对
浊音间的分割困难，提出了基于波形互相关性的音节分割方法。
（3）根据传统小波变换方法在基频检测实验中出现的问题，引入投票策略，提出
了一种基于小波变换的基频检测新算法。
（4）根据汉语声调曲线的特点，选择头尾差和相对调位比作为三字词各音节的声
调特征：选择头尾差和音节起点调位作为连续语音中音节的声调特征。
（5）定性和定量地分析了三字词的声调模式，印证了已有的三字词声调模式变化
规律，得到了一些新的三字词声调模式变调规则。对汉语连续语音的声调模
式，提出了以下观点：连续语音中的音节声调模式可以二字词和三字词的声
调模式为基础：连续语音中的音节可认为仅受前后音节的彤响，一定间隙前
后的音节声调可看成互不相关；连续语音中的音节声调模式可归为头、中和
尾三类，通过对这三类声调模式的建模，可解决连续语音的声调识别问题。
①）为了适应连续语音中的复杂情况，提出了以具有在线学习能力的模糊神经网
络作为声调识别模型的观点。在此基础上提出了基于模糊自适应谐振理论映
射的声调识别方法。
Tone is one of primary properties for Chinese. Its functions are listed as the following: constructing words, distinguishing semantic and improving expression effect. It is important to speech recognition, speech synthesis and natural language understanding.
In the recent years large progress has been made in speech auto-recognition; many voice speech systems have been developed. Now research on voice recognition turns to large vocabulary speaker-independent continuous speech recognition and natural language understanding. Tone information isn抰 basically used in current Chinese speech recognition system, study on tone recognition is limited to tone recognition of isolated-word and multi-syllabic word, and research on tone patterns and tone recognition for Chinese continuous speech is little.
In this dissertation a syllable-based method of tone recognition for continuous speech is proposed. This method includes the following procedures: syllable segmentation, pitch detection, feature extraction, tone pattern analysis, and tone recognition. Our research is show as follows:
(1) Syllable segmentation in continuous speech. Syllable is determined as tone recognition unit in this thesis, so syllable segmentation must be done. Syllable segmentation in continuous speech is very difficult. In accordance with chaotic essence of speech signals, syllable segmentation in continuous speech is researched by fractal theory. An approach of syllable segmentation using variance fractal dimension is proposed, its performance is analyzed in detail. The method can discriminate between voiced and unvoiced, between surd and sonant, but it can hardly discriminate between sonant. According to difference of speech waveform between transition segment and non-transition segment, dividing between sonant is researched using waveform cross correlation. A method of syllable segmentation is presented based on waveform cross-correlation.
(2) Pitch detection of speech signals. Mandarin tone is the patterns of pitch variation, so it may be acquired by pitch extraction. Many methods of pitch detection are developed so far, in this thesis the pitch detector using waveform transform is adopted. According to the problem appearing in pitch extraction experiment, a novel algorithm of pitch detection is presented. The pitch points in speech signal exhibit local maximum across several consecutive dyadic scales, and their positions are similar, so the improved approach selects pitch points by vote strategy, not by traditional method. Procedure of the new algorithm is as follows: (i) calculating the wavelet transform
III
2
Abstract
across 5 (or 3) consecutive scales; (ii) choosing pitch points by vote strategy; (iii) checking pitch points; (IV) relocation of pitch points.
(3) Feature extraction in tone recognition. Feature extraction is a basic problem in pattern recognition. Valid feature can reflect important information of pattern, and decrease computation and error recognition rate. Mandarin tone is characterized for tone level and tendency of pitch curve, so head-tail difference and relative tone level rate are determined as tone features for Chinese tn-syllable word; head-tail difference and tone level at the beginning of syllable are determined as tone features of syllable in continuous speech.
(4) Tone pattern analysis. Tone characteristics of syllable in continuous speech have lager variation than original tone characteristic under the influence of its preceding and posterior syllable, so tone patterns is more complicated, and have only basic characteristic of Chinese tone. Tone patterns and its variation rules are important to tone recognition for continuous speech. In this dissertation tone patterns of disyllabic word and isolated-word are introduced, tone patterns for tn-syllabic word are analyzed in detail. Tone patterns for continuous speech are researched based on the foregoing work.
(5) Selection and design of tone recognition module. Tone patterns of syllable in continuous speech
引文
[1]张家录．汉语语言识别的声学模型和语言模型．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，13-24．
[2]W.Hess. Pitch determination of speech signals: algorithms and devices. Berlin: Sprinter-Verlag,1983.
[3]S.Kadambe and G.Faye boudreaux-Bartels.Application of the wavelet transform for pitch detection of speech signal. IEEE Trans. on IT, vol. 38, pp.917-924, 1992.
[4]S.G.Mallat and S.Zhong. Characterization of signals from mulitscale edges.IEEE Trans. of Patt. Analy. And Mach. Intell., vol.14, pp.710-732, 1992.
[5]M.M. Sondhi.New methods of pitch extraction.IEEE Trans.on Audio Electroacoust,vol.AU-16, pp.262-266, 1968.
[6]A.M. Noll. Cepstrum pitch determination.J.Acoust. Soc.Amer., 1967, 41(2): 293-309.
[7]B.Gold and L.R. Rabiner.Parallel processing for estimating pitch periods of speech in the time domain.J.A.S.A., vol. 46, pp.442-448, 1969.
[8]J.D. Markel.The SIFT algorithm for fundamental frequency estimation,IEEE Trans.Audio Electroacoust., vol. AU-20, pp.367-377,1972.
[9]M.J.Ross et al. Average magnitude difference function pitch extractor.IEEE Trans.Acoust.,Speech, Signal Processing, vol.ASSP-22, pp.353-362, 1974.
[10]周俏峰，蔡莲红．音节数据库基音自动标注工具的研究．小型微型计算机系统，1995，16(10)：12-17．
[11]曹阳，黄泰翼，基于小波变化得基频提取和连续语音中基频变化模式的分析．见：第四届全国人机语音通讯学术会议论文集．北京，1996，271-276．
[12]Y.Medan, et al.Super resolution pitch determination of speech signals, IEEE Trans.on Signal Processing,1991, 39(1 ): 40-48.
[13]Abe T,Kobayashi T,Imai S. Robust pitch estimation with harmonics enhancement in noisy environments based on instantaneous frequency. ICSLP 96, 1996(2): 1277-1280.
[14]国立新，莫福源，李昌立．基于连续高斯混合密度 HMM 的汉语全音节语音识别研究．声学学报，1995，20(5)：321-329．
[15]Lee Chin-Hui, L.R. Rabiner, A frame-synchronous network search algorithm for connected word recognition. IEEE Trans. ASSP, 1989, 37(11): 1649-1658.
[16]J.Navarro-Mesa, et al.An improved speech endpoint detection system in noisy environments by means of third-order spectra. IEEE Signal Processing Letters, 1999, 6(9): 224-226.
[17]Nemat S. Abdel Kader and Amr M. Refat. End points detection for noisy speech using a wavelet based algorithm. In the proceedings of NRSC'99, Cairo, Egypt, 1999, c18: 1-5.
[18]朱维彬，张家录．汉语语音资料库的语音学标记及人工切分．声学学报，1999，24(3)：225-235．
[19]秦兵，沙宗先．连续小波变换在语音分形特征分析及语音分割中的应用．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，56-62．
[20]柴配琪．基于本特征的汉语音节切分方法．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，288-291．
[21]陈韬，莫福源，李昌立．语音信号的自动分段方法研究．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，292-295，
[22]刘建，张渊，俞铁城．汉语连续语流中浊音音段切分对识别影响研究．见：第四届全国人机语音通讯学术会议论文集．北京，1996，46-51．
[23]赵鹤鸣，周旭东．基于知识的汉语连续语音识别研究．计算机研究与发展，1993，No．6：44-48．
[24]张向东，刘建，俞铁成．基于声韵母转移模型的汉语特定人无限词汇连续语音识别研究．见：第四届全国人机语音通讯学术会议论文集，北京，1996，58-63．
[25]Y.Hashimo and S.Nakagawa. Investigation on segmentation of continuous speech using Hidden Markov Model. Record of Spring Meeting on Acoustic Society of Japan, 3-P-2 (1988 March in Japanese).
[26]林茂灿．北京话声调分布区的知觉研究．卢学学报，1995，20(6)：437-445．
[27]徐士林，樊懋，丁曙光．基于知识的汉语语音识别系统．模式识别与人工智能，1993，6(1)：49-54．
[28]陶维青，徐士林，钟金宏．汉语二字词声调的模式分析，中文信息学报，1998，12(3)：29-35．
[29]吴宗济，林茂灿．实验语音学概要．北京：高等教育出版社，1989，153-190．
[30]钟金宏，杨善林，徐士林．三字词声调的模糊识别方法．系统工程与电子技术，2000，22(12)：69-72．
[31]钟金宏，杨善林，陶维青，徐士林．基于音节的三字词声凋神经网络识别方法．模式识别与人工智能，2000，13(4)：339-442．
[32]黄泽镇，杨行峻．普通话孤立字四声的一种模式识别方法．声学学报，1990，15(1)：36-43．
[33]赵开江，杨行峻．汉语普通话孤立字的四声识别．第三届语音、通讯与图像处理论文集．
[34]徐士林，应勇．汉语卢凋的多特征模糊识别方法．模式识别与人工智能，1994，7(1)．
[35]Y.R.Wang, J.M. Shieh and S.H.Chen. Tone recognition of continuous Mandarin speech based on neural network. IEEE Trans. on ASSP, 1994, pp.2637-2645.
[36]W.J.Yang, et al. HMM for mandarin lexical tone recognition.IEEE Trans.on ASSP, July 1988,vo136, pp.988-992.
[37]L.R.拉宾纳，R.W.谢弗著．语音信号数字处理．北京：科学出版社，1983年．
[38]胡航．语音信号处理．哈尔滨：哈尔滨工业大学出版社，2000．
[39]方棣棠，汉语语音识别的当前任务与研究方向，见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，pp3-12．
[40]陈尚勤，罗承烈，杨雪．近代语音识别．北京：电子科学出版社，1991．
[41]陈方，高升．语音识别技术及发展．电信科学，1996，12(10)：54-57．
[42]语音知识．http://go6.163.com/～mypth/yyzs.htm.
[43]林茂灿．汉语语音实验研究在中国．http://xjjp.cass.net.cn/s18 yys／yuyin／rpr-il／linmc 98.htm.
[1]H.-O. Peitgen, H. Juergens, D.Saupe.Fractals for the Classroom, Pan One, Introduction to Fractals and Chaos.New York, Berlin,Heidelberg:Springer,1992.
[2]周凌云，王瑞丽，吴光敏，段良和，非线性物理理论及应用，北京：科学出版社，2000．3．
[3]黄润生．混沌及其应用．武汉：武汉大学出版社，2000．1．
[4]李后强，汪富泉．分形理论及其在分子科学中的应用．北京：科学出版社，1997．
[5]曼德尔布洛特著．分形对象一形、机遇和维数．文志英，苏虹泽．北京：世界图书出版公司北京公司，1999．12．
[6]J.D.Farmer. Chaotic Attractors of an Infinite-dimensional Dynamical System. Physica D 4,366-393,1982.
[7]A.H.Nayfeh and B.Balachandran.Applied Nonlinear Dynamics: Analytical, Computational, and Experimental Methods. New York: Wiley, pp. 545-547, 1995.
[8]P.Grassberger and Procaccia Itamar. Measuring the Strangeness of Strange Attractors, in Physica 9D, pp.189-208, 1983.
[9]D.H.Keefe and L. Bernice. Correlation dimension of woodwind multiphonic tones, in J. Acoust. Soc. Am. 90(4) Pt.1, pp.1754-65.
[10]G.L.Baker and P.G. Jerry. Chaotic Dynamics. New York: Cambridge University Press, 1996.
[11]P.Grassberger and I.Procaccia. Characterization of Strange Attractors, Phys. Rev. Letters, 50, 346-349, 1983.
[12]J.Theiler. Estimating Fractal Dimension, J. Optical Soc. Am. A 7, 1055-1073, 1990.
[13]J.D.Farmer. Chaotic Attractors of an Infinite-dimensional Dynamical System. Physica D 4,366-393, 1982.
[14]A.N.Kolmogorov, & V.M.Tihomirov.1959-1961. Epsilon-entropy and epsilon-capacity of sets in functional spaces. Uspekhi Matematicheskikh Nauk (N.S.): 14, 3-86. Traduction dans American Mathematical Society Translations (Series 2): 17, 277-364.
[15]Gerald A. Edgar. Measure, Topology, and Fractal Geometry. New York: Springer-Verlag, 1990.
[16]B.B.Mandelbrot.The Fractal Geometry of Nature, San Francisko: W,H. Freeman, 1983
[17]J.Feder. Fractals.New York, London: Plenum Press, 1989.
[18]B.B. Mandelbrot. Self-affine fractal sets. In L. Pietronero, E.Tosatti(eds.): Fractals in Physics.Amsterdam: North Holland, 1986.
[19]H.O. Peitgen, D. Saupe.The Science of Fractal Images. New York, Berlin, Heidelberg: Springer,1988.
[20]B.B. Mandelbrot. The Fractal Geometry of Nature, San Francisco, CA: Freeman, CA 1992.
[21]H.E. Hurst, R.P. Black & Y.M. Simaika, Long-Term Storage: An Experimental Study, Constable, London, 1965.
[22]A. Einstein, Investigations on the Theory of the Brownian Motion, Dover, USA, 1956.
[23]B.B.Mandelbrot, J.W. Van Ness. Fractal Brownian Motions, Fractional Noises and Applications,SIAM Rev., 1968, 10(4): 422-437.
[24]P. Zhang, H. Barad, and A. Martinez. Fractal dimension estimators for fractional Brownian motion, IEEE Southeastcon'90 Proceedings, vol.3, pp.934-939, 1990.
[25]W. Grleder and W. Kinsner. Speech segmentation by variance fractal dimension, Proc. 1994 Canadian Conf. On Electrical and Computer Engineering, 481-485, 1994.
[26]L. Jiao, W. Moon, and W. Kinsner. Variance fractal dimension analysis of seismic refraction signal: Proc. IEEE WESCANEX 97 Conf. On Comm., Power and computing, 116-120, 1997.
[27]T.S.Parker, L.O. Chua. Practical numerical algorithms for chaotic systems[M]. New York: Springer, 1989.
[28]C.Thompson, A. Mulpur. Transition to chaos in acoustically driven flow (acoustic streaming), J Acoust Soc Am, 1991, 90: 2097-2103.
[29]H.O. Peitgen, H. Jurgens. Chaos and fractals. New York: Springer-Verlag, 1992.
[30]P.Maragos. Fractal aspects of speech signals: dimension and interpolation. Proc. IEEE ICASSP, 1991, pp.417-420.
[31]W.Grieder and W. Kinsner, Speech segmentation by variance fractal dimension, Proc.1994 Canadian Conf. On Electrial and Computer Engineering, 481-485, 1994.
[32]Lingxiu Jiao and Wooil M. Moon, Detection of seismic refraction signals using a variance fractal dimension technique, Geophysics, 2000, 65(1): 286-292.
[33]钟金宏．汉语声调的模糊神经网络识别[硕士学位论文]．合肥：合肥工业大学，1998．
[34]J.S.Bendat,and A.G.Piersol.Random Data:Analysis and Measurement Procedures.New York:John Wiley&Sons,1971.Pg.332.
[35]A.V.Oppenheim,and R.W.Schafer.Digital Signal Processing.Englewood Cliffs,NJ:Prentice Hall,1975.Pgs.63-67,746-747,839-842.
[36]黄文梅等编著．信号分析与处理：MATLAB语言及应用．长沙：国防科技大学出版社，2000．2，216-217．
[37]徐守时．信号与系统：理论·方法和应用．合肥：中国科学技术大学出版社，1999．9，34-39．
[38]李后强，汪富泉．分形理论及其发展历程．http://best .163.com/～sbw /fxrm/fxrm007.htm.
[1]Y.Meyer. Wavelets algorithms and applications (英译本译者：R.D. Ryan), Chapter 2: Wavelet from a historical perspective, Society for Industrial and Applied Mathematics, 1993.
[2]I. Daubechies, Where do wavelets come from? — A personal point of view, Proc. IEEE 84(5),510-513, 1996.
[3]秦前清，杨宗凯．实用小波分析．西安：西安电子科技大学出版社，1998．
[4]杨福生．小波变换的工程分析与应用．北京：科学出版社，1992．2．
[5]胡昌华，张军波，夏军，张伟。基于MATLAB的系统分析与设计—小波分析．西安：西安电子科技大学出版社，1999．12．
[6]刘贵忠，邸双亮．小波分析与应用．西安：西安电子科技大学出版社，1997．
[7]李建平，唐远炎．小波分析方法的应用．重庆：重庆大学出版社，1999．10．
[8]崔锦泰．小波分析导论．程正兴译，西安：西安交通大学出版社，1994．7．
[9]I. Daubechies. Orthonormal bases of compactly supported wavelets. Comm. On Pure and Appl.Math., 1988, 41(7): 909-996.
[10]S. Mallat. Multiresolution approximations and wavelet orthonormal bases of L~2(R). Trans. Amer. Math. Soc.,1989, 315: 69-87.
[11]I. Daubechies. Ten lectures on wavelets, CBMS-NSF Series in Appl, Math., SIAM, 1991.
[12]S. Mallat and S. Zhong. Characterization of signals from multiscale edges. IEEE Trans. PAMI,1992, 14(7): 710-732.
[13]S. Mallat and W.L. Hwang. Singularity detection and processing with wavelets. IEEE Trans.Information Theory, 1992, 38(2): 617-643.
[14]聂永安，崔晓峰，陈宇坤．用小波变换进行信号特征检测的原理与方法．地震地磁观察与研究，1999，Vol．20，No．3．
[15]李世雄编，小波变换及其应用．北京：高等教育出版社．1997．
[16]李后强，汪富泉，分形理论及其在分子科学中的应用．北京：科学出版社，1997．
[17]钟金宏．汉语声调的模糊神经网络识别[硕士学位论文]．合肥：合肥工业大学，1998．
[18]程正兴．小波分析算法与应用．西安：西安交通大学出版社，1997．9．
[19]S. Kadambe and G. Faye boudreaux-Bartels, Application of the wavelet transform for pitch detection of speech signal. IEEE Trans. on IT. 1992, 38(2): 917-924.
[20]王长富，林志刚，戴蓓倩，张劲松．基于小波变换的语音基音周期检测．中国科学技术大学学报，1995，25(1)：47-52．
[21]赵鹤鸣，周旭东，金延庆，翁桂荣．基于小波变换的重叠语音基频提取及声调识别．声学学报，1999，24(1)：87-93．
[22]程俊，张璞，戴善荣，易克初．小波变换用于信号突变的检测．通信学报，1995，16(3)：96-104．
[23]钟金宏，杨善林，张学应．汉语连续语音三字词声调提取方法研究．合肥工业大学学报(自然科学版)，2000，23(5)：710-714．
[24]陶维青，徐士林，钟金宏．汉语三字词声调的模式分析．中文信息学报，1998，12(3)：29-35．
[1]杨行峻，迟惠生等编．语音信号数字处理．北京：电子工业出版社，1995年．
[2]徐士林，陶维青，钟金宏．非特定人二字词声调模糊识别方法．模式识别与人工智能，1998，11(1)：80—88．
[3]陶维青，徐士林，钟金宏．汉语三字词声调的模式分析．中文信息学报，1998，12(3)：29-35．
[4]吴宗济，林茂灿主编．实验语音学概要．北京：高等教育出版社，1989年．
[5]钟金宏．汉语声调的模糊神经网络识别[硕士学位论文]．合肥：合肥工业大学，1998．
[6]曹阳，黄泰翼．基于小波变化得基频提取和连续语音中基频变化模式的分析．见：第四届全国人机语音通讯学术会议论文集．北京，1996，271-276．
[7]沈晓楠(林茂灿译)．汉语普通话的协同发音．国外语言学，1992，第二期．
[8]卢甲文．关于三个上声连续变调问题的商榷．语言教学与研究，1979，第二期．
[9]杨善林，钟金宏，刘业政，刘应玲，一种基于简化FUZZY ARTMAP的声调识别规则获取方法．模式识别与人工智能，已录用．
[1]R.P. Lippmann. An introduction to computing with neural nets.IEEE ASSP Magazine,1987,pp. 4-22.
[2]M.A.Kramer, J.A. Leonard. Diagnosis using back-propagation neural networks: analysis and criticism. Computer Chem. Engin., 1990, 14: 1328-1338.
[3]B.D. Ripley. Neural networks and related methods for classification. J.R. Statist. Soc.B., 1994,56(3): 409-456.
[4]G.A. Carpenter, S. Grossberg. A massively parallel architecture for a self-organizing neural pattern recognition machine. Computer Vision, Graphics and Image Processing, 1987, 37:54-115.
[5]G.A. Carpenter, S. Grossberg. Stable self-organization of pattern recognition codes for analog input patterns. Applied Optics, 1987, 26(23): 4919-4930.
[6]G.A. Carpenter, S. Grossberg. ART3: Hierarchical search using chemical transmitters in self-organizing pattern recognition architectures. Neural Networks, 1990, 3:129-152.
[7]G.A. Carpenter, S. Grossberg, & D. Rosen. Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system. Neural Networks, 1991, 4: 759-771.
[8]G.A. Carpenter, S. Grossberg, & J.H. Reznolds. ARTMAP: Supervised real-time learning and classification of nonstationary data by a self-organizing neural network. Neural Networks, 1991,inconsistent cases.Neural Networks,1998,11:323-336.
4: 565-588．
[9] G.A. Carpenter, S. Grossberg. Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps. IEEE Transactions on Neural Networks, 1992, 3(5) : 698-713．
[10] L. Zadeh, Fuzzy sets, Inform. Contr., 1965, 8:338-353．
[11] F.M. Ham, & S. Han. Classification of cardiac arrhythmia using fuzzy ARTMAP.IEEE Transactions on Biomedical Engineering, 1996,43: 425-430．
[12] G.A. Carpenter, M.N. Gjaja, S. Gopal, & C.E. Woodcock. ART neural networks for remote sensing: Vegetation classification from Landsat TM and terrain data. IEEE Transactions on Geoscience and Remote Sensing, 1997,35: 308-325．
[13] J. Racz, & A. Dubrawski. Artificial neural networks for mobile robot topological localization. Robotics and Autonomous Systems, 1995, 16: 73-80．
[14] S.S. Kalkunte, J.M. Kumar, & L.M. Patnaik. A neural network approach for high-resolution fault diagnosis in digital circuits. Proceedings of the International Joint Conference on Neural Networks, 1992, pp. 83-88．
[15] G.A. Carpenter, S. Grossberg, & J.H. Reynolds. A Fuzzy ARTMAP nonparametric probability estimator for nonstationary pattern recognition problems. IEEE Transaction on Neural Networks, 1995,6(6) : 1330-1336．
[16] G.A. Carpenter, & A.H. Tan. Rule extraction: From neural architecture to symbolic representation. Connection Science, 1995,7:3-27．
[17] Issam Dagher, Michael Georgiopoulos, Gregory L. Heileman, George Bebis.. An ordering algorithm for pattern presentation in Fuzzy ARTMAP that tends to improve generalization performance. IEEE Transaction on Neural Networks, 1999,10(4) : 768-778．
[18] Paul S. Wu, Ming Li. Supervised and unsupervised fuzzy-adaptive Hamming net, Pattern Recognition, 1999,32: 1801-1816．
[19] G.A. Carpenter, & A.H. Tan. Rule extraction, Fuzzy ARTMAP and medical databases. Proceedings of World Congress on Neural Networks, 1993,1: 501-506．
[20] S. Marriott, & R.F. Harrison. A modified fuzzy ARTMAP architecture for the approximation of noisy mappings. Neural Networks, 1995, 8(4) : 619-641．
[21] G.A. Carpenter, & W.D. Ross. ART-EMAP: A neural network architecture for object recognition by evidence accumulation. IEEE Transactions on Neural Networks, 1995, 6(4) : 805-818．
[22] G.A. Carpenter, F.D.M. Wilson. ARTMAP-DS: Pattern discrimination by discounting similarities. Proceedings of the International Conference on Artificial Neural Networks, 1997, pp. 607-612．
[23] G.A. Carpenter, & N. Markuzon. ARTMAP-IC and medical diagnosis: instance counting and
[24]G.A. Carpenter.Distributed learning, recognition, and prediction by ART and ARTMAP neural networks.Neural Networks, 1997, 10(8): 1473-1494.
[25]C.P.Lim, & R.F. Harrison. Probabilistic fuzzy ARTMAP: An autonomous neural network architecture for Bayesian probability estimation. Proceedings of Fourth International Conference on Artificial Neural Networks, 1995, 6:148-153.
[26]N.I.Lee. A coupled-ART neural network capable of modularized categorization of patterns. Patten, Recognition Letters, 1999, 20:131-140.
[27]陈兆乾，周戎等．一种新的自适应谐振算法FTART．软件学报，1996，7(8)：458-465．
[28]陈道文，黄泰翼．面向语音处理的神经网络发展综述．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，25-30．
[29]P.C.Chang, S.W. Sun, S.H. Chen. Mandarin tone recognition by multi-layer percepion. Proceeding of ICASSP-90, 1990.
[30]徐士林，陶维青，钟金宏．非特定人二字词声凋模糊识别方法．模式识别与人工智能，1998，7(1)：80-88．
[31]钟金宏，杨善林，张学应．汉语连续语音三字词声调提取方法研究．合肥上业大学学报(自然科学版)，2000，23(5)：710-714．
[32]钟金宏，杨善林，陶维青，徐士林．基于音节的三字词声调神经网络识别方法．模式识别与人工智能，2000，13(4)：339-442．
[33]杨毓英，史习智，李国奕．基于Fuzzy ARTMAP的室性早搏诊断规则获取．自然科学进展，1999，9(5)：448-455．
[34]P.K. Simpson. Fuzzy min-max neural networks Part 1:classification. IEEE Transactions on Neural Networks, 1992, 3(5): 776-786.
[35]G.A. Carpenter, & A.H. Tan. Rule extraction: From neural architecture to symbolic representation. Connection Science, 1995, 7(1): 3-27.
[2]W.Hess. Pitch determination of speech signals: algorithms and devices. Berlin: Sprinter-Verlag,1983.
[3]S.Kadambe and G.Faye boudreaux-Bartels.Application of the wavelet transform for pitch detection of speech signal. IEEE Trans. on IT, vol. 38, pp.917-924, 1992.
[4]S.G.Mallat and S.Zhong. Characterization of signals from mulitscale edges.IEEE Trans. of Patt. Analy. And Mach. Intell., vol.14, pp.710-732, 1992.
[5]M.M. Sondhi.New methods of pitch extraction.IEEE Trans.on Audio Electroacoust,vol.AU-16, pp.262-266, 1968.
[6]A.M. Noll. Cepstrum pitch determination.J.Acoust. Soc.Amer., 1967, 41(2): 293-309.
[7]B.Gold and L.R. Rabiner.Parallel processing for estimating pitch periods of speech in the time domain.J.A.S.A., vol. 46, pp.442-448, 1969.
[8]J.D. Markel.The SIFT algorithm for fundamental frequency estimation,IEEE Trans.Audio Electroacoust., vol. AU-20, pp.367-377,1972.
[9]M.J.Ross et al. Average magnitude difference function pitch extractor.IEEE Trans.Acoust.,Speech, Signal Processing, vol.ASSP-22, pp.353-362, 1974.
[10]周俏峰，蔡莲红．音节数据库基音自动标注工具的研究．小型微型计算机系统，1995，16(10)：12-17．
[11]曹阳，黄泰翼，基于小波变化得基频提取和连续语音中基频变化模式的分析．见：第四届全国人机语音通讯学术会议论文集．北京，1996，271-276．
[12]Y.Medan, et al.Super resolution pitch determination of speech signals, IEEE Trans.on Signal Processing,1991, 39(1 ): 40-48.
[13]Abe T,Kobayashi T,Imai S. Robust pitch estimation with harmonics enhancement in noisy environments based on instantaneous frequency. ICSLP 96, 1996(2): 1277-1280.
[14]国立新，莫福源，李昌立．基于连续高斯混合密度 HMM 的汉语全音节语音识别研究．声学学报，1995，20(5)：321-329．
[15]Lee Chin-Hui, L.R. Rabiner, A frame-synchronous network search algorithm for connected word recognition. IEEE Trans. ASSP, 1989, 37(11): 1649-1658.
[16]J.Navarro-Mesa, et al.An improved speech endpoint detection system in noisy environments by means of third-order spectra. IEEE Signal Processing Letters, 1999, 6(9): 224-226.
[17]Nemat S. Abdel Kader and Amr M. Refat. End points detection for noisy speech using a wavelet based algorithm. In the proceedings of NRSC'99, Cairo, Egypt, 1999, c18: 1-5.
[18]朱维彬，张家录．汉语语音资料库的语音学标记及人工切分．声学学报，1999，24(3)：225-235．
[19]秦兵，沙宗先．连续小波变换在语音分形特征分析及语音分割中的应用．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，56-62．
[20]柴配琪．基于本特征的汉语音节切分方法．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，288-291．
[21]陈韬，莫福源，李昌立．语音信号的自动分段方法研究．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，292-295，
[22]刘建，张渊，俞铁城．汉语连续语流中浊音音段切分对识别影响研究．见：第四届全国人机语音通讯学术会议论文集．北京，1996，46-51．
[23]赵鹤鸣，周旭东．基于知识的汉语连续语音识别研究．计算机研究与发展，1993，No．6：44-48．
[24]张向东，刘建，俞铁成．基于声韵母转移模型的汉语特定人无限词汇连续语音识别研究．见：第四届全国人机语音通讯学术会议论文集，北京，1996，58-63．
[25]Y.Hashimo and S.Nakagawa. Investigation on segmentation of continuous speech using Hidden Markov Model. Record of Spring Meeting on Acoustic Society of Japan, 3-P-2 (1988 March in Japanese).
[26]林茂灿．北京话声调分布区的知觉研究．卢学学报，1995，20(6)：437-445．
[27]徐士林，樊懋，丁曙光．基于知识的汉语语音识别系统．模式识别与人工智能，1993，6(1)：49-54．
[28]陶维青，徐士林，钟金宏．汉语二字词声调的模式分析，中文信息学报，1998，12(3)：29-35．
[29]吴宗济，林茂灿．实验语音学概要．北京：高等教育出版社，1989，153-190．
[30]钟金宏，杨善林，徐士林．三字词声调的模糊识别方法．系统工程与电子技术，2000，22(12)：69-72．
[31]钟金宏，杨善林，陶维青，徐士林．基于音节的三字词声凋神经网络识别方法．模式识别与人工智能，2000，13(4)：339-442．
[32]黄泽镇，杨行峻．普通话孤立字四声的一种模式识别方法．声学学报，1990，15(1)：36-43．
[33]赵开江，杨行峻．汉语普通话孤立字的四声识别．第三届语音、通讯与图像处理论文集．
[34]徐士林，应勇．汉语卢凋的多特征模糊识别方法．模式识别与人工智能，1994，7(1)．
[35]Y.R.Wang, J.M. Shieh and S.H.Chen. Tone recognition of continuous Mandarin speech based on neural network. IEEE Trans. on ASSP, 1994, pp.2637-2645.
[36]W.J.Yang, et al. HMM for mandarin lexical tone recognition.IEEE Trans.on ASSP, July 1988,vo136, pp.988-992.
[37]L.R.拉宾纳，R.W.谢弗著．语音信号数字处理．北京：科学出版社，1983年．
[38]胡航．语音信号处理．哈尔滨：哈尔滨工业大学出版社，2000．
[39]方棣棠，汉语语音识别的当前任务与研究方向，见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，pp3-12．
[40]陈尚勤，罗承烈，杨雪．近代语音识别．北京：电子科学出版社，1991．
[41]陈方，高升．语音识别技术及发展．电信科学，1996，12(10)：54-57．
[42]语音知识．http://go6.163.com/～mypth/yyzs.htm.
[43]林茂灿．汉语语音实验研究在中国．http://xjjp.cass.net.cn/s18 yys／yuyin／rpr-il／linmc 98.htm.
[1]H.-O. Peitgen, H. Juergens, D.Saupe.Fractals for the Classroom, Pan One, Introduction to Fractals and Chaos.New York, Berlin,Heidelberg:Springer,1992.
[2]周凌云，王瑞丽，吴光敏，段良和，非线性物理理论及应用，北京：科学出版社，2000．3．
[3]黄润生．混沌及其应用．武汉：武汉大学出版社，2000．1．
[4]李后强，汪富泉．分形理论及其在分子科学中的应用．北京：科学出版社，1997．
[5]曼德尔布洛特著．分形对象一形、机遇和维数．文志英，苏虹泽．北京：世界图书出版公司北京公司，1999．12．
[6]J.D.Farmer. Chaotic Attractors of an Infinite-dimensional Dynamical System. Physica D 4,366-393,1982.
[7]A.H.Nayfeh and B.Balachandran.Applied Nonlinear Dynamics: Analytical, Computational, and Experimental Methods. New York: Wiley, pp. 545-547, 1995.
[8]P.Grassberger and Procaccia Itamar. Measuring the Strangeness of Strange Attractors, in Physica 9D, pp.189-208, 1983.
[9]D.H.Keefe and L. Bernice. Correlation dimension of woodwind multiphonic tones, in J. Acoust. Soc. Am. 90(4) Pt.1, pp.1754-65.
[10]G.L.Baker and P.G. Jerry. Chaotic Dynamics. New York: Cambridge University Press, 1996.
[11]P.Grassberger and I.Procaccia. Characterization of Strange Attractors, Phys. Rev. Letters, 50, 346-349, 1983.
[12]J.Theiler. Estimating Fractal Dimension, J. Optical Soc. Am. A 7, 1055-1073, 1990.
[13]J.D.Farmer. Chaotic Attractors of an Infinite-dimensional Dynamical System. Physica D 4,366-393, 1982.
[14]A.N.Kolmogorov, & V.M.Tihomirov.1959-1961. Epsilon-entropy and epsilon-capacity of sets in functional spaces. Uspekhi Matematicheskikh Nauk (N.S.): 14, 3-86. Traduction dans American Mathematical Society Translations (Series 2): 17, 277-364.
[15]Gerald A. Edgar. Measure, Topology, and Fractal Geometry. New York: Springer-Verlag, 1990.
[16]B.B.Mandelbrot.The Fractal Geometry of Nature, San Francisko: W,H. Freeman, 1983
[17]J.Feder. Fractals.New York, London: Plenum Press, 1989.
[18]B.B. Mandelbrot. Self-affine fractal sets. In L. Pietronero, E.Tosatti(eds.): Fractals in Physics.Amsterdam: North Holland, 1986.
[19]H.O. Peitgen, D. Saupe.The Science of Fractal Images. New York, Berlin, Heidelberg: Springer,1988.
[20]B.B. Mandelbrot. The Fractal Geometry of Nature, San Francisco, CA: Freeman, CA 1992.
[21]H.E. Hurst, R.P. Black & Y.M. Simaika, Long-Term Storage: An Experimental Study, Constable, London, 1965.
[22]A. Einstein, Investigations on the Theory of the Brownian Motion, Dover, USA, 1956.
[23]B.B.Mandelbrot, J.W. Van Ness. Fractal Brownian Motions, Fractional Noises and Applications,SIAM Rev., 1968, 10(4): 422-437.
[24]P. Zhang, H. Barad, and A. Martinez. Fractal dimension estimators for fractional Brownian motion, IEEE Southeastcon'90 Proceedings, vol.3, pp.934-939, 1990.
[25]W. Grleder and W. Kinsner. Speech segmentation by variance fractal dimension, Proc. 1994 Canadian Conf. On Electrical and Computer Engineering, 481-485, 1994.
[26]L. Jiao, W. Moon, and W. Kinsner. Variance fractal dimension analysis of seismic refraction signal: Proc. IEEE WESCANEX 97 Conf. On Comm., Power and computing, 116-120, 1997.
[27]T.S.Parker, L.O. Chua. Practical numerical algorithms for chaotic systems[M]. New York: Springer, 1989.
[28]C.Thompson, A. Mulpur. Transition to chaos in acoustically driven flow (acoustic streaming), J Acoust Soc Am, 1991, 90: 2097-2103.
[29]H.O. Peitgen, H. Jurgens. Chaos and fractals. New York: Springer-Verlag, 1992.
[30]P.Maragos. Fractal aspects of speech signals: dimension and interpolation. Proc. IEEE ICASSP, 1991, pp.417-420.
[31]W.Grieder and W. Kinsner, Speech segmentation by variance fractal dimension, Proc.1994 Canadian Conf. On Electrial and Computer Engineering, 481-485, 1994.
[32]Lingxiu Jiao and Wooil M. Moon, Detection of seismic refraction signals using a variance fractal dimension technique, Geophysics, 2000, 65(1): 286-292.
[33]钟金宏．汉语声调的模糊神经网络识别[硕士学位论文]．合肥：合肥工业大学，1998．
[34]J.S.Bendat,and A.G.Piersol.Random Data:Analysis and Measurement Procedures.New York:John Wiley&Sons,1971.Pg.332.
[35]A.V.Oppenheim,and R.W.Schafer.Digital Signal Processing.Englewood Cliffs,NJ:Prentice Hall,1975.Pgs.63-67,746-747,839-842.
[36]黄文梅等编著．信号分析与处理：MATLAB语言及应用．长沙：国防科技大学出版社，2000．2，216-217．
[37]徐守时．信号与系统：理论·方法和应用．合肥：中国科学技术大学出版社，1999．9，34-39．
[38]李后强，汪富泉．分形理论及其发展历程．http://best .163.com/～sbw /fxrm/fxrm007.htm.
[1]Y.Meyer. Wavelets algorithms and applications (英译本译者：R.D. Ryan), Chapter 2: Wavelet from a historical perspective, Society for Industrial and Applied Mathematics, 1993.
[2]I. Daubechies, Where do wavelets come from? — A personal point of view, Proc. IEEE 84(5),510-513, 1996.
[3]秦前清，杨宗凯．实用小波分析．西安：西安电子科技大学出版社，1998．
[4]杨福生．小波变换的工程分析与应用．北京：科学出版社，1992．2．
[5]胡昌华，张军波，夏军，张伟。基于MATLAB的系统分析与设计—小波分析．西安：西安电子科技大学出版社，1999．12．
[6]刘贵忠，邸双亮．小波分析与应用．西安：西安电子科技大学出版社，1997．
[7]李建平，唐远炎．小波分析方法的应用．重庆：重庆大学出版社，1999．10．
[8]崔锦泰．小波分析导论．程正兴译，西安：西安交通大学出版社，1994．7．
[9]I. Daubechies. Orthonormal bases of compactly supported wavelets. Comm. On Pure and Appl.Math., 1988, 41(7): 909-996.
[10]S. Mallat. Multiresolution approximations and wavelet orthonormal bases of L~2(R). Trans. Amer. Math. Soc.,1989, 315: 69-87.
[11]I. Daubechies. Ten lectures on wavelets, CBMS-NSF Series in Appl, Math., SIAM, 1991.
[12]S. Mallat and S. Zhong. Characterization of signals from multiscale edges. IEEE Trans. PAMI,1992, 14(7): 710-732.
[13]S. Mallat and W.L. Hwang. Singularity detection and processing with wavelets. IEEE Trans.Information Theory, 1992, 38(2): 617-643.
[14]聂永安，崔晓峰，陈宇坤．用小波变换进行信号特征检测的原理与方法．地震地磁观察与研究，1999，Vol．20，No．3．
[15]李世雄编，小波变换及其应用．北京：高等教育出版社．1997．
[16]李后强，汪富泉，分形理论及其在分子科学中的应用．北京：科学出版社，1997．
[17]钟金宏．汉语声调的模糊神经网络识别[硕士学位论文]．合肥：合肥工业大学，1998．
[18]程正兴．小波分析算法与应用．西安：西安交通大学出版社，1997．9．
[19]S. Kadambe and G. Faye boudreaux-Bartels, Application of the wavelet transform for pitch detection of speech signal. IEEE Trans. on IT. 1992, 38(2): 917-924.
[20]王长富，林志刚，戴蓓倩，张劲松．基于小波变换的语音基音周期检测．中国科学技术大学学报，1995，25(1)：47-52．
[21]赵鹤鸣，周旭东，金延庆，翁桂荣．基于小波变换的重叠语音基频提取及声调识别．声学学报，1999，24(1)：87-93．
[22]程俊，张璞，戴善荣，易克初．小波变换用于信号突变的检测．通信学报，1995，16(3)：96-104．
[23]钟金宏，杨善林，张学应．汉语连续语音三字词声调提取方法研究．合肥工业大学学报(自然科学版)，2000，23(5)：710-714．
[24]陶维青，徐士林，钟金宏．汉语三字词声调的模式分析．中文信息学报，1998，12(3)：29-35．
[1]杨行峻，迟惠生等编．语音信号数字处理．北京：电子工业出版社，1995年．
[2]徐士林，陶维青，钟金宏．非特定人二字词声调模糊识别方法．模式识别与人工智能，1998，11(1)：80—88．
[3]陶维青，徐士林，钟金宏．汉语三字词声调的模式分析．中文信息学报，1998，12(3)：29-35．
[4]吴宗济，林茂灿主编．实验语音学概要．北京：高等教育出版社，1989年．
[5]钟金宏．汉语声调的模糊神经网络识别[硕士学位论文]．合肥：合肥工业大学，1998．
[6]曹阳，黄泰翼．基于小波变化得基频提取和连续语音中基频变化模式的分析．见：第四届全国人机语音通讯学术会议论文集．北京，1996，271-276．
[7]沈晓楠(林茂灿译)．汉语普通话的协同发音．国外语言学，1992，第二期．
[8]卢甲文．关于三个上声连续变调问题的商榷．语言教学与研究，1979，第二期．
[9]杨善林，钟金宏，刘业政，刘应玲，一种基于简化FUZZY ARTMAP的声调识别规则获取方法．模式识别与人工智能，已录用．
[1]R.P. Lippmann. An introduction to computing with neural nets.IEEE ASSP Magazine,1987,pp. 4-22.
[2]M.A.Kramer, J.A. Leonard. Diagnosis using back-propagation neural networks: analysis and criticism. Computer Chem. Engin., 1990, 14: 1328-1338.
[3]B.D. Ripley. Neural networks and related methods for classification. J.R. Statist. Soc.B., 1994,56(3): 409-456.
[4]G.A. Carpenter, S. Grossberg. A massively parallel architecture for a self-organizing neural pattern recognition machine. Computer Vision, Graphics and Image Processing, 1987, 37:54-115.
[5]G.A. Carpenter, S. Grossberg. Stable self-organization of pattern recognition codes for analog input patterns. Applied Optics, 1987, 26(23): 4919-4930.
[6]G.A. Carpenter, S. Grossberg. ART3: Hierarchical search using chemical transmitters in self-organizing pattern recognition architectures. Neural Networks, 1990, 3:129-152.
[7]G.A. Carpenter, S. Grossberg, & D. Rosen. Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system. Neural Networks, 1991, 4: 759-771.
[8]G.A. Carpenter, S. Grossberg, & J.H. Reznolds. ARTMAP: Supervised real-time learning and classification of nonstationary data by a self-organizing neural network. Neural Networks, 1991,inconsistent cases.Neural Networks,1998,11:323-336.
4: 565-588．
[9] G.A. Carpenter, S. Grossberg. Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps. IEEE Transactions on Neural Networks, 1992, 3(5) : 698-713．
[10] L. Zadeh, Fuzzy sets, Inform. Contr., 1965, 8:338-353．
[11] F.M. Ham, & S. Han. Classification of cardiac arrhythmia using fuzzy ARTMAP.IEEE Transactions on Biomedical Engineering, 1996,43: 425-430．
[12] G.A. Carpenter, M.N. Gjaja, S. Gopal, & C.E. Woodcock. ART neural networks for remote sensing: Vegetation classification from Landsat TM and terrain data. IEEE Transactions on Geoscience and Remote Sensing, 1997,35: 308-325．
[13] J. Racz, & A. Dubrawski. Artificial neural networks for mobile robot topological localization. Robotics and Autonomous Systems, 1995, 16: 73-80．
[14] S.S. Kalkunte, J.M. Kumar, & L.M. Patnaik. A neural network approach for high-resolution fault diagnosis in digital circuits. Proceedings of the International Joint Conference on Neural Networks, 1992, pp. 83-88．
[15] G.A. Carpenter, S. Grossberg, & J.H. Reynolds. A Fuzzy ARTMAP nonparametric probability estimator for nonstationary pattern recognition problems. IEEE Transaction on Neural Networks, 1995,6(6) : 1330-1336．
[16] G.A. Carpenter, & A.H. Tan. Rule extraction: From neural architecture to symbolic representation. Connection Science, 1995,7:3-27．
[17] Issam Dagher, Michael Georgiopoulos, Gregory L. Heileman, George Bebis.. An ordering algorithm for pattern presentation in Fuzzy ARTMAP that tends to improve generalization performance. IEEE Transaction on Neural Networks, 1999,10(4) : 768-778．
[18] Paul S. Wu, Ming Li. Supervised and unsupervised fuzzy-adaptive Hamming net, Pattern Recognition, 1999,32: 1801-1816．
[19] G.A. Carpenter, & A.H. Tan. Rule extraction, Fuzzy ARTMAP and medical databases. Proceedings of World Congress on Neural Networks, 1993,1: 501-506．
[20] S. Marriott, & R.F. Harrison. A modified fuzzy ARTMAP architecture for the approximation of noisy mappings. Neural Networks, 1995, 8(4) : 619-641．
[21] G.A. Carpenter, & W.D. Ross. ART-EMAP: A neural network architecture for object recognition by evidence accumulation. IEEE Transactions on Neural Networks, 1995, 6(4) : 805-818．
[22] G.A. Carpenter, F.D.M. Wilson. ARTMAP-DS: Pattern discrimination by discounting similarities. Proceedings of the International Conference on Artificial Neural Networks, 1997, pp. 607-612．
[23] G.A. Carpenter, & N. Markuzon. ARTMAP-IC and medical diagnosis: instance counting and
[24]G.A. Carpenter.Distributed learning, recognition, and prediction by ART and ARTMAP neural networks.Neural Networks, 1997, 10(8): 1473-1494.
[25]C.P.Lim, & R.F. Harrison. Probabilistic fuzzy ARTMAP: An autonomous neural network architecture for Bayesian probability estimation. Proceedings of Fourth International Conference on Artificial Neural Networks, 1995, 6:148-153.
[26]N.I.Lee. A coupled-ART neural network capable of modularized categorization of patterns. Patten, Recognition Letters, 1999, 20:131-140.
[27]陈兆乾，周戎等．一种新的自适应谐振算法FTART．软件学报，1996，7(8)：458-465．
[28]陈道文，黄泰翼．面向语音处理的神经网络发展综述．见：杨家沅主编．语音识别和语音合成．重庆：四川科学技术出版社，1994，25-30．
[29]P.C.Chang, S.W. Sun, S.H. Chen. Mandarin tone recognition by multi-layer percepion. Proceeding of ICASSP-90, 1990.
[30]徐士林，陶维青，钟金宏．非特定人二字词声凋模糊识别方法．模式识别与人工智能，1998，7(1)：80-88．
[31]钟金宏，杨善林，张学应．汉语连续语音三字词声调提取方法研究．合肥上业大学学报(自然科学版)，2000，23(5)：710-714．
[32]钟金宏，杨善林，陶维青，徐士林．基于音节的三字词声调神经网络识别方法．模式识别与人工智能，2000，13(4)：339-442．
[33]杨毓英，史习智，李国奕．基于Fuzzy ARTMAP的室性早搏诊断规则获取．自然科学进展，1999，9(5)：448-455．
[34]P.K. Simpson. Fuzzy min-max neural networks Part 1:classification. IEEE Transactions on Neural Networks, 1992, 3(5): 776-786.
[35]G.A. Carpenter, & A.H. Tan. Rule extraction: From neural architecture to symbolic representation. Connection Science, 1995, 7(1): 3-27.