Pattern recognition apparatus and pattern recognition method
作者： Yuki Tachioka , Shinji Watanabe
DOI:
关键词:
摘要: Provided is a pattern recognition apparatus for creating multiple systems and combining the to improve performance, including discriminative training unit constructing model parameters of second or subsequent system based on an output tendency previously-constructed so as be different from model. Accordingly, when are combined, performance can improved without trials-and-errors.
lens.org 参考文章(28)
Zhi-Jie Yan, Qiang Huo, Yu Zhang, Normalization based discriminative training for continuous speech recognition ,(2012)
Jing Zheng, Xin Lei, Method and apparatus for computing gaussian likelihoods ,(2011)
S. Charles Brubaker, Jianxin Wu, Jie Sun, Matthew D. Mullin, James M. Rehg, Towards the Optimal Training of Cascades of Boosted Ensembles Toward Category-Level Object Recognition. pp. 301- 320 ,(2006) , 10.1007/11957959_16
Xiaodong He, Jian Wu, Speech models generated using competitive training, asymmetric training, and data boosting ,(2005)
Mitsunori Ogihara, Tao Li, Qi Li, Music feature extraction using wavelet coefficient histograms ,(2004)
Ruhi Sarikaya, Brian Edward Doorenbos Kingsbury, Yuqing Gao, Yonggang Deng, Machine translation in continuous space conference of the international speech communication association. pp. 2350- 2353 ,(2008)
Jianfeng Gao, Hisami Suzuki, Discriminative training for language modeling ,(2005)
Sibel Yaman, Integrated speech recognition and semantic classification Journal of the Acoustical Society of America. ,vol. 129, pp. 4099- ,(2007) , 10.1121/1.3600966
Carsten Meyer, Hauke Schramm, Boosting HMM acoustic models in large vocabulary speech recognition Speech Communication. ,vol. 48, pp. 532- 548 ,(2006) , 10.1016/J.SPECOM.2005.09.009
George Saon, Hagen Soltau, Boosting systems for large vocabulary continuous speech recognition Speech Communication. ,vol. 54, pp. 212- 218 ,(2012) , 10.1016/J.SPECOM.2011.07.011