(3.238.240.197) 您好!臺灣時間:2021/04/12 03:38
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:杜思良
研究生(外文):Si-liang Du
論文名稱:利用共同向量法於特定語者中文單音辨識
論文名稱(外文):Using the Method of Common Vector to Recognize Isolate Mandarin Word for Speaker-Dependent System
指導教授:李宗寶
學位類別:碩士
校院名稱:國立中興大學
系所名稱:應用數學系所
學門:數學及統計學門
學類:數學學類
論文種類:學術論文
畢業學年度:96
語文別:中文
論文頁數:39
中文關鍵詞:共同向量K-th Nearest Neighbor分類法
外文關鍵詞:Common VectorK-th Nearest Neighbor
相關次數:
  • 被引用被引用:2
  • 點閱點閱:85
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
本篇論文主要是探討50個國字單音的辨識,首先利用共同向量來建構出語音模型,之後試著加入K-th Nearest Neighbor(K-NN)分類法觀察辨識率是否比共同向量法來的好,也嘗試使用混合模型(K-th Nearest Neighbor法加共同向量法),本論文討論的其他三個實驗因子:「音框數」、「分群數」、「語音特徵參數個數」,希望能找出在何種情況下50個字能具有不錯的鑑別度。
This paper is to discuss the speech recognition of 50 isolated mandarin words. First, we use common vector to construct the speech model, Then we will try to use K-th Nearest Neighbor, if it can improve the rate of recognition .We also attempt to join common vector and K-th Nearest Neighbor. We consider the other three experimental factors in this paper: "the number of cluster", "speech feature extraction", “the number of K-th” hoping to find out the circumstances under which the 50 characters can be a good discrimination.
第一章 緒論......................1
1.1研究動機及目的................2
1.2語音辨識研究範圍和應用範圍....2
1.2.1研究範圍.................2
1.2.2應用範圍.................3
1.3語音辨識方法概述..............4
1.3.1前處理...................4
1.3.2特徵參數.................5
1.3.3建立模型與辨識...........6
1.3.4辨識流程.................8
1.4論文架構......................9
第二章 語音訊號前處理與特徵參數..10
2.1前言..........................10
2.2語音前處理....................10
2.2.1取樣.....................10
2.2.2常態化...................11
2.2.3取音框...................12
2.2.4端點偵測.................12
2.2.5預強調...................15
2.2.6視窗化...................15
2.3特徵參數求取..................16
2.3.1導出倒頻譜參數...........16
2.3.2差倒頻譜參數.............18
第三章 語音模型建立與辨識方法....20
3.1前言..........................20
3.2音框壓縮與擴張................20
3.2.1音框數超過10組...........20
3.2.2音框數少於10組...........21
3.3建構模型方法..................22
3.3.1K-means分群法............22
3.3.2K-means分群法中心值設定..25
3.3.3共同向量法原理...........25
3.4辨識方法......................28
3.4.1待測語音的處理...........28
3.4.2比對方法.................28
3.5 K-th Nearest Neighbor(K-NN)..30
第四章 實驗操作流程與實驗結果....31
4.1操作介面......................31
4.2實驗流程......................31
4.2.1語音來源.................31
4.2.2影響辨識率的可能因素.....31
4.2.3實驗結果.................32
第五章 結論與未來研究方向........36
1. 王小川。"語音訊號處理"
2. 王國榮。" Visual Basic 6.0 實戰講座"
3. 李宗寶,吳宗憲。"探討K-means之共同向量法應用於國語數字辨識"
4. 李宗寶,林靖剛。"利用Multiple Common Vector 於國語數字之語音辨識"。
5. 吳明哲,黃世陽。Visual Basic 6.0 中文版學習範本
6. Angm H., “Common vector obtained from linearly independent speech vectors by using LPC parameters,” graduation project, Elect. Electron. Eng. Dept., Osmangazi Univ., Eskisehir, Turkey, 1995.
7. Bing, X. and Yihe, S. (1996). Research on ASIC for multi-speaker isolated word recognition, ASIC, 2nd International Conference, 21-24, 135-137.
8. Gulmezoglu M. B., Vakif Dzhafarov, and Atalay Barkana, “ A novel approach to isolated word recognition,” IEEE Trans. On Speech and Audio Processing, vol. 7. No. 6, 1999.
9. Harb, H., and Husseiny, A.H. (2000). Isolated words recognition using neural networks, The 7th IEEE International Conference on, 1, 17-20, 349-351.
10. Keskin M., Gulmezoglu M. B., Parlaktuna O., and Barkana A., “Isolated word recognition by extracting personal differences,” in Proc. 6 th Int.Conf. Signal Processing Applications and Technology, Boston, MA , pp.1989-1992, 1996.
11. Li, T. F. (2003). Speech recognition of mandarin monosyllables, Pattern Recognition 36, 2713-2721.
12. Rabiner L.R. and Sambur M.R., “An algorithm for determining the endpoints of isolated utterances”,The Bell System Technique Journal,Vol.54,pp.297-315,1975.
13. Sakoe H. and Chiba S., “Dynamic Programming Optimization for Spoken Word Recognition,” IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. 26, pp. 43-49, 1978.
14. Yucel S., “Application of Gram-Schmidt orthogonalization method to speech recognition for different noise levels” graduation project, Elect. Electron. Eng. Dept., Osmangazi Univ., Eskisehir, Turkey, 1996.
15. Rabiner L.R. and Sambur M.R., “An algorithm for determining the endpoints of isolated utterances”,The Bell System Technique ournal,Vol.54,pp.297-315,1975.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔