Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/17701
標題: 利用共同向量法於地名之語音辨識
Using the Method of Common Vector to Speech Recognition for the Name of Cities
作者: 曾采蓮
Zeng, Cai-Lian
關鍵字: Speaker-Dependent;特定語者;Speech Recognition;Common Vector;Cepstral Coefficient;Zero-Crossing Rate;語音辨識;共同向量;倒頻譜係數;越零率
出版社: 應用數學系所
引用: [1] 王小川。“語音訊號處理”。台北市:全華,2004。 [2] 王國榮。“Visual Basic 6.0 實戰講座”。台北巿:旗標, 2000。 [3] 黎自奮,李宗寶,張正樺。“用K-means 方法於時域特徵之國語數字辨認”。碩士論文,國立中興大學應用數學研究所,台中,2004。 [4] 李宗寶,林峰樣。“探討共同向量法應用於國語數字辨識”。碩士論文,國立中興大學應用數學研究所,台中,2004。 [5] 李琳山,李上銘。“語音辨識中基於主成份分析之進一步技術”。碩士論文,國立台灣大學電信工程學研究所,台北,2000。 [6] M. B. Gulmezoglu, Vakif Dzhafarov, and Atalay Barkana, “A novel approach to isolated word recognition,”IEEE Trans.on Speech and Audio Processing, Vol. 7. No. 6, 1999. [7] M. Bilginer Gulmezoglu, Vakif Dzhafarov, and Atalay Barkana ,“The common vector approach and its relation to principal component analysis,”IEEE Trans. On Speech and Audio Processing,Vol. 9. No. 6. [8] M. Keskin, M. B. Gulmezoglu, O. Parlaktuna, and A. Barkana, “Isolated word recognition by extracting personal differences,” in Proc. 6 th Int.Conf. Signal Processing Applications and Technology, Boston, MA , pp.1989-1992, 1996. [9] S. Yucel, “Application of Gram-Schmidt orthogonalization method to speech recognition for different noise levels,” graduation project, Elect. Electron. Eng. Dept., Osmangazi Univ., Eskisehir, Turkey, 1996. [10] H. Angm, “Common vector obtained from linearly independent speech vectors by using LPC parameters,” graduation project, Elect. Electron. Eng. Dept., Osmangazi Univ., Eskisehir, Turkey, 1995. [11] M. R. Sambur and L. R. Rabiner, “A speaker-independent digit recognition system,” B. S. T. J., vol. 54, pp. 84-102, 1975. [12] L. Rabiner, S.E. Levinson, A. E. Rosenberg, and J. G. Wilpon, “Speaker-independent recognition of isolated words using clustering techniques,” IEEE Trans. Acoust. Speech, Signal Processing,Vol. ASSP-27, pp. 336-349, Aug. 1979. [13] G. Mercier et al., “Recognition of speaker–dependent continuous speech with KEAL,” in Readings in Speech Recognition, A. Waibel and K.-F. Lee, Eds. San Mateo, CA: Morgan Kaufmann, pp.225-234,1990.
摘要: 
由於現代科技十分進步,人類對於電腦的使用率很高,因此能提升對電腦輸入命令的速度,是非常方便的,而以語音來做為對電腦的輸入工具是一種很好的選擇。
本篇論文主要探討43種台灣各縣市地名,並針對「特定語者」所做的辨識。首先,將收集到的語音資料做前置處理和求取特徵值。前置處理的端點偵測部分主要是用能量量測法和越零率,找出語音的起點與終點。求取特徵值主要是找出代表語音特徵的倒頻譜係數。最後用K-means之共同向量法來辨識語音,實驗後得到總辨識率97.91%,辨識效果非常好。因此,未來的研究希望可以推廣至「非特定語者」的辨識。

Because the modern science and technology progress very much, and the rate of utilization of the computer is high for mankind, so we can improve the speed that the computer inputs the order, it is very convenient, and it is a kind of very good choice to regard speech as the inputting tool of the computer.
The paper discuss mainly speech recognition that 43 kinds of name of cities of Taiwan about “Speaker-Dependent”. First of all, the speech corpus is processed in advance and is found out the speech feature. To find the start-point and the end-point of the speech, we use the end-point detection that is the method of energy-measuring and zero-crossing rate. Find the speech feature is to find out the cepstral coefficient of the speech mainly. Finally, we use the method of common vector of K-means to recognize speech, and the total rate of speech recognition is 97.91%. It is very good to recognize speech. Therefore, we hope that the paper can be researched to the speech recognition about “Speaker-Independent” in the future.
URI: http://hdl.handle.net/11455/17701
其他識別: U0005-3006200615235200
Appears in Collections:應用數學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.