本論文採用一種新的語詞辨識方法，將同一類的語詞轉換為差異子空間的方式，消除語者本身或語者之間說話時的差異性，並且應用於多語者的使用環境之中。此外，本論文亦提出一種新的端點偵測法來判斷信號中語音的成分。最後並以Microsoft Windows為作業平台，完成理論的驗證。

A novel phrase recognition method is proposed. It eliminates the speech difference between intraspeaker or interspeaker by transform phrases to difference subspace. A new endpoint detection method is also proposed, it can detection the human speech signal more effectively. All methods are test and verify at Microsoft Windows environment.

目 錄
頁 次
論文提要..…….………………………………………………………..Ⅱ
致謝…………………………………………………………………….Ⅲ
目錄…………………………………………………………………….Ⅳ
圖表目錄……………………………………………………………….Ⅴ
第一章 緒論………………………………………………..…………1

1-1 研究動機………………………………………………………1
1-2 語音辨識系統介紹……………………………………………1
1-3 論文主題………………………………………………………6
1-4 論文架構………………………………………………………7

第二章 語音訊號處理…………………………………………………8

2-1 語音處理介紹…………………………………………………8
2-1-1 端點偵測(Endpoint Detection)……………………...9
2-1-2 預強(Preemphasize)…………………………………11
2-1-3 加窗函數…………………………………………….12
2-2 端點偵測之研究………………………………………………15
2-3 倒頻譜係數(Cepstrum Coefficient)…………………………...21
2-4 梅爾-倒頻譜(Mel-Cepstrum Coefficient) …………………….23

第三章 語詞辨識之研究………………………………………………27

3-1 信號空間………………………………………………………27
3-2 語詞信號空間…………………………………………………29
3-3 共通向量與差異子空間的正交性質…………………………33
3-4 LBG分類器 ………………………………………………….34

第四章 實驗結果及系統設計…………………………………………37

4-1 英文語音資料測試結果………………………………………41
4-2 中文語音資料測試結果………………………………………46
4-3 視窗程式架構…………………………………………………49

第五章 結論與建議……………………………………………………50

[1] John R. Deller,Jr. , John G. Proakis, and John H. L. Hansen—“Discrete- Time Processing of Speech Signals”, New Jersey, Prentice Hall, 1987.

[2] 林燾，王理嘉—“語音學教程”，五南圖書出版公司，1995.

[3] Rivarol Vergin, Douglas O’Shaughnessy, and Azarshid Farhat— “Generalized Mel Frequency Cepstral Coefficients for Large-Vocabulary Specker-Independent Continuous-Speech Recognition”, IEEE Transactions on Speech and Audio Processing, Vol. 7, NO.5, September 1999.

[4] Zimer Tranter—“Principles of Communications”, Houghton Mifflin Company, 1995, 4th Edition.

[5] Steven J.Leon—“Linear Algebra with Application”, Macmillan Publishing Company, New York, 3rd, 1990.

[6] M.B.Gulmezoglu , V.Dzhafarov , M.Keskin and A.Barkana ,”A Novel Approach to Isolated Word Recognition,” IEEE Trans.on Speech and Audio Processing,Vol.7 , No.6 , pp.620-628,1999.

[7] 龍生雲, “不特定語句之中文語者辨識系統研究”, 國立中山大學電機工程研究所博士論文, pp.20, 民國88年11月17日

[8] Lawrence Rabiner, Biing-Hwang Juang—“Fundamentals of Speech Recognition”, Prentice Hall, 1993.



[9] Liang-sheng Huang and Chung-ho Yang, “A Novel Approach to Robust Speech Endpoint Detection in Car Environments”, Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on , Volume: 3 , pp.1751-1754.

[10] Joseph W.Picone, “Signal Modeling Technique in Speech Recognition”, Proceedings of The IEEE, Vol. 81, NO.9, pp.1221-1223, September 1993.

[11] Alan V. Oppenheim ,Ronald W. Schafer , with John R.Buck—1st ed .”Discrete-Time Signal Processing”, Chapter 12, Prentice Hall , 1989 .

[12] Yariv Ephraim and Harry L.Van Trees, “A Signal Subspace Approach for Speech Enhancement”, IEEE Transactions on Speech and Audio Processing, Vol.3, NO. 4, pp251-266, July 1995

[13] 林合仁, “中文語詞辨識系統之視窗軟體設計研究”, 國立中山大學電機工程研究所碩士論文, 民國88年6月30日

[14] Shallom, I.D., Haimi-Cohen, R., Rannon, Z.M., “Dynamic Time Warping with Generalized Templates for Speaker Independent Speech Recognition”, Electrical and Electronics Engineers in Israel. The Sixteenth Conference of page1-page4. 1989.