Method and equipment of pattern recognition, its program and its recording medium

作者: Tsuneo Kato

DOI:

关键词:

摘要: The present invention provides a method and equipment of pattern recognition capable efficiently pruning partial hypotheses without lowering accuracy, its program, recording medium. In second search unit, likelihood calculation unit calculates an acoustic by matching time series data feature parameters against lexical tree stored in database model third to determine accumulated accumulating the direction. A self-transition causes each hypothesis make process. An LR transition RL transition. reward attachment adds R(x) accordance with number reachable words raise likelihood. excludes less from targets.

参考文章(13)
Vyacheslav A. Barannikov, Alexandr A. Kibkalo, Method and apparatus for dynamic beam control in viterbi search ,(2001)
Eric Rolse Buhrke, Wu Chou, Method and apparatus for continuous speech recognition using a layered, self-adjusting decoded network Journal of the Acoustical Society of America. ,vol. 108, pp. 888- ,(1997) , 10.1121/1.1319436
Qifeng Zhu, Dynamic pruning for automatic speech recognition Journal of the Acoustical Society of America. ,vol. 134, pp. 2377- ,(2009) , 10.1121/1.4820207
Li Jiang, System for using silence in speech recognition Journal of the Acoustical Society of America. ,vol. 112, pp. 2520- ,(1999) , 10.1121/1.1536510
Yen-Lu Chow, Speech recognition system with multi-level pruning for acoustic matching The Journal of the Acoustical Society of America. ,vol. 104, pp. 1155- ,(1998) , 10.1121/1.424310
Jean-Claude Junqua, System and method using N-best strategy for real time recognition of continuously spelled names Journal of the Acoustical Society of America. ,vol. 103, pp. 3137- ,(1995) , 10.1121/1.423099