臺灣博碩士論文加值系統

在本論文中，我們將使用數位攝影機來建立一個數位影音資料庫，以作語者辨識之用。此資料庫包含了1500個人，於三個不同時段，所錄製之不特定語句語者的動態影音資料，及每人每段影音資料中所擷取之20張照片。寄望本資料庫之完成，能對同時以聲音及臉形兩種生物特徵，來作身份辨認之系統設計，提供一個訓練與評量之機制。

In this thesis, an audio-video database for speaker recognition is constructed using a digital camcorder. Motion pictures of fifteen hundred speakers are recorded in three different sessions in the database. For each speaker, 20 still images per session are also derived from the video data. It is hoped that this database can provide an appropriate training and testing mechanism for person identification using both voice and face features.

誌謝辭 I
論文摘要 II
目錄 III
圖目錄 V
第一章 緒論 1
1-1研究動機與目的 1
1-2 語者辨識概論 2
1-3 章節概要 5
第二章 語音處理相關技術之研究 6
2-1 語音處理相關領域 6
2-2 語音辨識相關技術 9
2-2-1 端點偵測(Endpoint Detection) 9
2-2-2 能量(Energy) 10
2-2-3 越零率(Zero Crossing Rate) 10
2-2-4 最大相似比(Maximum Likelihood Rate, MLR)11
2-3 視窗函數(Window Function) 12
2-4 特徵萃取(Feature Extraction) 13
2-4-1 線性預估編碼(Linear Predictive Coding, LPC)14
2-4-2梅爾倒頻譜係數(Mel-Frequency Cepstrum Coefficients, MFCC)16
第三章 國內、外語者辨識用影音資料庫19
3-1 國內語音資料庫 19
3-1-1國語語音資料庫MAT-160 19
3-1-2國語語音資料庫MAT-400 19
3-1-3國語語音資料庫MAT-2500 20
3-1-4麥克風語料庫 TCC-300Edu 20
3-1-5國語連續數字語音資料庫 21
3-1-6中文廣播新聞語料庫 MATBN 22
3-2 國外語音資料庫 23
3-2-1 TIMIT語音資料庫 23
3-2-2 NTIMIT語音資料庫 25
3-2-3 CTIMIT語音資料庫 26
3-2-4 NIST標準對話電話語音資料庫 26
3-2-5 AVTIMIT影音資料庫 27
3-2-6 VidTIMIT影音資料庫 27
第四章 實驗設計、數據及討論 28
4-1 資料庫建立之參數設計 28
4-2 實驗結果 30
第五章 結論與展望 33
5-1 結論 33
5-2 未來展望 33
參考文獻 35

[1] Lawrence Rabiner and Bing-Hwang Juang, “Fundamentals of speech recognition”,Prentice Hall, 1993.

[2] D.A. Reynolds, “Speaker identification and verification using Gaussian mixture speaker models,” Speech Communication 17. pp.91-108 , March 1995

[3] A. P. Dempster, N. M. Laira and D. B. Rubin, “Maximum Likelihood fromIncomplete Data via the EM Algorithm,” Harvard University and Educational Testing Service, Dec. 1976.

[4] Biing-Hwang Juang, Wu Chou, and Chin-Hui Lee, ”Minimum Classification ErrorRate Methods for Speech Recognition,” IEEE Trans. On Speech and Audio Processing. Vol. 5, NO. 3, May 1997.

[5] W. Chou, B.H. Juang and C.H Lee, “Segmental GPD Training of HMM based Speech Recognizer,” In proceedings of ICASSP, IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol. 1, page(s): 473 -476, 1992.41

[6] del Alamo, C.M.; Caminero Gil, F.J.; dela Torre Munilla, C.; Hernandez Gomez, L.“Discriminative Training of GMM for Speaker Identification,” In proceedings ofICASSP, IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol. 1, page(s): 89 -92 , 1996.

[7] Li Lee, Richard Rose, “A Frequency Warping Approach to Speaker
Normalization,” IEEE Trans. On Speech and Audio Processing. Vol. 6, NO.1,January 1998.

[8] Welling, L.; Kanthak, S.; Ney, H., “Improved Method For Vocal TractNormalization,” IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol.2, page(s): 761 –764, 1999.

[9] W.M. Fisher, G.R. Doddingdon “The DARPA Speech Recognition Research Database: Specifications And Status”, In Proc. DARPA Workshop Speech Recognition, Feb. 1986, pp93-99

[10]L.F. Lemal, J.L Cauvain “Cross-Lingual Experiments With Phone Recognition”, In Proc. Int. Conf. Acoustic Speech Signal Processing, 1993, pp507-510

[11]John R. Deller, John G. Prooakls, John H. Hansen, “Discrete-Time Processing Of Speech Signals”, Maxwell Macmillan international

[12] Y. Linde, A. Buzo & R. Gray, “An Algorithm For Vector Quantizer Design”, IEEE Transactions on Communications, Vol. 28, pp.84-95, 1980

[13]Moody. J, Slomka .S, Pelecanos. J, “On The Convergence Of Gasssain Mixture Models: Improvements Through Vector Quantization”, ICSLP98.
[14]A. P. Dempster, N. M. Laird, “Maximum-Likelihood For Incomplete Data Via The EM Algorithm”, J. Royal Statist. Soc. SerB., pp39, 1977.

[13]L. Rabiner, B. H. Juang, “Fundamentals of Speech Recognition”, Prentice Hall Signal Processing Series, 1993.

[14]王小川，”語音訊號處理”，全華，民國93年.

[15]戴顯權，”資料壓縮”，紳藍，民國91年.