System and method for parameter estimation for pattern recognition

作者: Assaf Ben-Yishai , David Burshtein

DOI:

关键词: Estimation theoryArtificial intelligenceEquivalence (measure theory)Set functionPattern recognitionComputer scienceTraining setEstimator

摘要: A parameter estimator for estimating a set of parameters pattern recognition has recognizer receiving training having members. The performs on the members using current and based upon predetermined group elements. generator associated with generates at least one equivalence containing recognized set, which are used by target function determiner to calculate parameters. maximizer updates so as maximize calculated function.

参考文章(11)
Michael Alan Picheny, Dimitri Kanevsky, David Nahamoo, Hamed A. Ellozy, Michelle Y. Kim, Wlodek Wlodzimierz Zadrozny, Automatic indexing and aligning of audio and text using speech recognition ,(1995)
Martin J. Russell, Julie L. Wallace, Robert W. Series, Speech training aid ,(1996)
Radu Jasinschi, Thomas McGee, Lalitha Agnihotri, Nevenka Dimitrova, Streaming video bookmarks ,(2001)
Jean-claude Junqua, Michael Galler, Call routing device employing continuous speech ,(1996)
Chin H. Lee, Speech recognition employing key word modeling and non-key word modeling Journal of the Acoustical Society of America. ,vol. 101, pp. 649- ,(1990) , 10.1121/1.419407
Eiichi Tsuboka, Voice analyzing system using hidden Markov model and having plural neural network predictors Journal of the Acoustical Society of America. ,vol. 97, pp. 2627- 2627 ,(1990) , 10.1121/1.411878
Rafid A. Sukkhar, Keyword/non-keyword classification in isolated word speech recognition Journal of the Acoustical Society of America. ,vol. 99, pp. 646- ,(1992) , 10.1121/1.414608