本論文探討中文三字詞與四字詞之語音辨識問題。我

In this thesis, a three-word and four-word Mandarin phrases speech recognition system is developed. This system contains two recordings of twenty-four thousand three-word phrases and twenty-two thousand four-word phrases in the database. And it applies MFCC, mono-syllable HMM’s and speech-text alignment scheme to select the initial phrase candidates. A wavelet transform based vowel segmentation technique and a Mandarin pitch identification method is then followed to increase the phrase correct identification rate and obtain the final answer. Experimental results indicate that 92% and 96% correct rates can be achieved for three-word and four-word phrases recognition problems respectively, under the conditions that the first recording of this database is used for training and the second one is for testing. For the speaker-dependent case, the correct phrase can be found within 1 second, using a PC with Intel Celeron 2.4 GHz CPU and RedHat Linux 9.0 Operation System.

第1章 緒論...............................................1
1-1 研究動機與目的....................................1
1-2 研究方法..........................................1
1-3 章節概要..........................................3

第2章 系統架構與語音訊號處理相關技術.....................4
2-1 系統架構..........................................4
2-2 切割單音..........................................5
2-2-1 線性預估編碼..................................5
2-2-2 訊號能量(Energy)..............................6
2-2-3 過零率(Zero Crossing Rate)....................6
2-3 特徵萃取(Feature Extraction) .....................8
2-3-1 漢明視窗(Hamming Window ).....................8
2-3-2 梅爾倒頻譜(MFCC)..............................8
2-3-3 餘弦轉換(DCT).................................9
2-3-4 倒濾波器(Low-time Lifter).....................9
2-3-5 倒頻譜係數(Delta-MFCC).......................10
2-4 四聲聲調分類.....................................13
2-4-1 中文語音特性.................................13
2-4-2 自相關函數(ACF，Autocorrelation Function)....14
2-4-3 Mallat演算法.................................16
2-4-4 小波係數能量分佈.............................18
2-5 隱藏式馬可夫模型(HMM)............................19
2-6 單音模型文字比對方法.............................21

第3章 實驗結果..........................................22
3-1 聲調實驗結果.....................................23
3-2 單音模型模擬實驗結果.............................24
3-3 錯誤分析與改善...................................27
3-4 改善策略.........................................28


第4章 結論與建議........................................30
4-1 結論.............................................30
4-2 建議.............................................31

第五章 參考資料.........................................32

[1] 賴昭華, “不特定語者中量語詞辨識系統之設計研究” , 國立中山大電機工程研究所碩士論文, 民國91年7月.
[2] 許博閔, “混合式中文人名語音辨識系統之設計研究”, 國立中山大學電機工程研究所碩士論文, 民國93年7月.
[3] 張慶勇, “中文地址語音辨識系統之設計研究”, 國立中山大學電機工程研究所碩士論文, 民國93年7月.
[4] 潘睿慈, “特定語者中文語詞辨識系統之設計研究”, 國立中山大學電機工程研究所碩士論文, 民國94年7月.
[5] L. Rabiner, B. H. Juang, Fundamentals of Speech Recognition, Prentice Hall Signal Processing Series, 1993..
[6] A. M. Kondoz, Digital Speech Coding, New York : John Wiley & Sons Inc., 1994.
[7] S. S. Stevens and J. Volkmann, The relation of pitch of frequency : A revised scale, Am. J. Psychol., 1940.
[8] John R. Deller, J. G.. Proakis, and John H. L. Hansen, Discrete-Time Processing of Speech Signals, New York: Macmillan Pub. Co., 1993
[9] L. R. Rabiner, “A tutorial on hidden Markov modles and selected application in speech recognition”, Proc. IEEE, vol.77, pp. 257-286, Feb. 1989.
[10] V. R. Algazi, K. L. Brown, M. J. Ready, D. H. Irvine, C. L. Cadwell and Sang Chung, “Transform Representation of the Spectra of Acoustic Speech Segment with Applications－I: General Approach and Application to Speech Recognition,” IEEE Trans. Speech and Audio Processing, vol.1, No.2, April 1993.
[11] S.V. Vaseghi, P.N. Conner “Speech modeling using cepstral-time feature matrices in hidden Markov models”, Proc. IEEE,vol.140,NO.5,OCTOBER 1993
[12] Jiqing Han , Wen Gao “Robust telephone speech recognition based on channel compensation”, Patten Recognition,32（1999）,106-1067
[13] Mallat, “A theory of mutiresolution signal decomposition: The wavelet transform”, IEEE Trans., PAMI- NO.7, 1989