![]() |
|
以作者查詢圖書館館藏 、以作者查詢臺灣博碩士 、以作者查詢全國書目 、勘誤回報 、線上人數:73 、訪客IP:3.135.182.160
姓名 黃啟祥(Chi-Xiang Huang) 查詢紙本館藏 畢業系所 電機工程學系 論文名稱 結合高斯混合及支撐向量機模型之語者確認研究
(Speaker Verification based on Combinations of GMM and SVM)相關論文 檔案 [Endnote RIS 格式]
[Bibtex 格式]
[相關文章]
[文章引用]
[完整記錄]
[館藏目錄]
[檢視]
[下載]
- 本電子論文使用權限為同意立即開放。
- 已達開放權限電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
- 請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
摘要(中) 本研究主要針對語者確認系統提出新的辨識流程,係利用動態微分核函數結合高斯混合模型及支撐向量機模型,藉以提升系統效能。
此系統主要是將原始語音特徵向量及已經由UBM-MAP所調適出的高斯混合語者模型參數透過微分核函數中的Fisher Kernel及Likelihood Ratio Kernel做映射,藉此得到不同語者新的特徵參數,藉以建出一個超級向量,在訓練階段中,需將超級向量做正規化,之後利用正規化後的超級向量訓練SVM模型,而在仿冒語音的選取上,則是選取與目標語者模型分數最高的前20名仿冒語音,使得訓練出來的SVM模型更有鑑別力,而測試時,則將測試語音做映射,所得到的超級向量在正規化後與對應的SVM模型計算距離值。
從實驗結果顯示,高斯混合模型選定為128-mixture且選取20位仿冒語者的語音,系統可達最好的相等錯誤率及決策成本函數分別為8.69%及0.1023,比起使用Fisher Kernel做映射的語者確認系統的效能11.92%及0.1500改善了3.23%及0.0477,而比起傳統語者確認模型的效能15.87%及0.1911,改善了7.18%及0.0888。
摘要(英) This thesis proposes a new recognition system to improve performance for speaker verification. The proposed system combines the Gaussian Mixture Model (GMM) and Support Vector Machine (SVM) with Derivative Kernel.
Target GMM for each target speaker was adapted from UBM by using target speech. we take Imposter speech feature selection method to choose specific imposter speech features and we can get new features in the target speaker GMM space by Fisher and likelihood ratio kernel mapping. Then we used the new features to establish target supervector and imposter supervector. In the train stage, we used the target supervector and imposter supervector to train SVM model. In the testing stage, we take the supervector into SVM to calculate the distance.
The experimental results shows that with a 128-mixture GMM and choose twenty imposter speech, the proposed system obtains a 3.23% EER and 4.77% DCF improvement over the SVM-fisher speaker verification system, and a 7.18% EER and 8.88% DCF improvement over the baseline system.
關鍵字(中) ★ 語者確認
★ 支撐向量機
★ 高斯混合模型
★ 微分核函數關鍵字(英) ★ Support vector machine
★ Derivative Kernel
★ speaker verification
★ Gaussian mixture model論文目次 摘要.........................i
目錄.........................iii
附圖目錄.......................vi
附表目錄.......................vii
第一章 緒論
1.1 研究動機...................... 1
1.2 語者辨識概述及分類................. 2
1.3 研究方向...................... 4
1.4 文獻回顧...................... 5
1.4 章節概要...................... 6
第二章 語音處理與語者辨識基本技術
2.1 語音特徵參數擷取.................. 6
2.2 語者模型建立.................... 9
2.2.1 高斯混合模型..................9
2.2.2 向量量化...................11
2.2.3 期望值最大化演算法........... .. 14
2.3 貝式調適法..................... 15
2.4 語者識別...................... 17
2.5 語者確認...................... 18
2.6 語者確認效能評估.................. 19
第三章 系統架構
3.1 支撐向量機.....................21
3.2 微分核函數.....................28
3.3 高斯混合結合支撐向量機之語者模型訓練........31
3.4 特定仿冒語音選取..................32
3.5 高斯混合結合支撐向量機之語者確認系統........33
第四章 語者辨識實驗之研究
4.1 語音資料庫..................... 37
4.2微分核函數結合GMM&SVM之語者確認系統......... 39
4.2.1 實驗一 使用Fisher kernel............. 39
4.2.2 實驗二 使用Likelihood ratio kernel........ 42
4.2.3 實驗三 使用Likelihood ratio kernel+仿冒語音挑選..44
第五章 結論與未來展望
5.1 結論........................ 48
5.2 未來展望...................... 49
參考文獻........................50
參考文獻 [1]T. S. Jaakkola, D. Haussler, “Exploiting generative Models in discriminative classifiers,” in Advances in Neural Information Processing Systems 11,M. S. Kearns, S. A. Solla, and D. A. Cohn, Eds. Cambridge,U.K.: MIT Press, 1998.
[2]L. R. Rabiner and B. H. Juang, Fundamentals of Speech Recognition, Prentice Hall, New Jersey, 1993.
[3]X. Huang, A. Acero and H. W. Hon, Spoken Language Processing, Prentice Hall, 2001.
[4]R. Vergin and D. O’Shaughnessy and A. Farhat, “Generalized Mel Frequency Coefficients for Large-Vocabulary Speaker-Independent Continuous-SpeechRecognition,” IEEE Trans. Speech and Audio Processing, vol. 7, no. 5, pp. 525-532, September 1999.
[5]D. A. Reynolds and R. C. Rose, “Robust Text-Independent Speaker Identification Using Gaussian Mixture Models,” IEEE Trans. Speech and Audio Processing, vol. 3, no. 1, pp. 72-83, January 1995.
[6]T. K. Moon, “The Expectation-Maximization Algorithm,” IEEE Signal Processing Magazine, vol. 13, no. 6, pp. 47-60, November 1996.
[7]D. Reynolds and T. Quatieri, “Speaker Verification Using Adapted Gaussian Mixture Models,” Digital Signal Processing 10, PP. 19-41, 2000.
[8]A. Martin, G. Doddington, T. Kamn, M. Ordowski, and M, Przybocki, “The DET curve in assessment of detection task performance,” in Proceedings of European Conference on Speech Communication and Technology, pp. 1895-1898, 1997.
[9]Johan A.K. Suykens, Tony Van Gestel, Jos De Brabanter, Bart De Moor and Joos Vandewalle, Least Squares Support Vector Machines, World Scientific, 2002
[10]V. N. Vapnik, The Nature of Statistical Learning Theory, Springer, 1995.
[11]Minqiang Xu, Beiqian Dai, Dongxing Xu, Shiqing Yang, Qingsong Liu “SVM-based Text-independent Speaker Verification using Derivative Kernel in the Reference GMM Space”,IEEE International Symposiums
[12]S. Raghavan, G.. Lazarou and J. Picone, “Speaker Verification Using Support Vector Machines,” in Proc. IEEE, 2006.
[13]C. P. Chen and J.Bilmes,“MVA Processing of Speech Features”, Audio, Speech and Language Processing, vol. 15, pp257-270, 2007.
[14]M.H. Liu, B.Q. Dai, Y.L. Xie, Z.Q. Yao, “Improved GMM-UBM/SVM for speaker verification,” in Proc. IEEE ICASSP, 2006.
[15]W. M. Campbell, “SVM based speaker verification using a GMM Supervector kernel and NAP variability compensation”, ICASSP 2006.
[16]M.H. Liu, B.Q. Dai, Y.L. Xie, Z.Q. Yao, “A New Hybrid GMM/SVM for speaker verification,” in Proc. IEEE ICPR, 2006.
[17]C. Longworth and M. J. Gales, “Derivative and Parametric Kernels for Speaker Verification,” InterSpeech2007
[18]V. Wan and S.Renals, “Speaker verification using sequence discriminant support vector machines,” IEEE Trans. Speech and Audio Processing, 2004
[19]The NIST Year 2001 Speaker Recognition Evaluation Plan”, http://www.nist.gov/speech/tests/spk/2001/
[20]Chih-Wei Hsu, Chih-Chung Chang, and Chih-Jen Lin, “A Practical Guide to Support Vector Classification”, available
http://www.csie.ntu.edu.tw/~cjlin/libsvm.
[21]吳金池, “語者辨識系統之研究",國立中央大學電機工程研究所碩士論文,民國九十一年。
[22]賴彥輔, “語者辨識之研究",國立中央大學電機工程研究所碩士論文,民國九十二年。
[23]陳柏仁,“應用投票演算法之語者確認系統研究",國立中央大學電機工程研究所碩士論文,民國九十六年。
[24]游智翔,“整合高斯混合與具性能指標支撐向量機模型之語者確認研究",國立中央大學電機工程研究所碩士論文,民國九十七年。
指導教授 莊堯棠(Yau-Tarng Juang) 審核日期 2009-7-15 推文 plurk
funp
live
udn
HD
myshare
netvibes
friend
youpush
delicious
baidu
網路書籤 Google bookmarks
del.icio.us
hemidemi
myshare