MS-TDNN with global discriminant trainings

This article analyses the behavior of various hybrid architectures based on a multi-state neuro-Markovian scheme (MS-TDNN HMM) applied to online handwriting word recognition systems. We have considered different cost functions, including maximal mutual information criteria with discriminant training and maximum likelihood estimation, to train the systems globally at the word level and also we varied the number of states from one up to three to model the basic hidden Markov models at the letter level. We report experimental results for non-constrained, writer independent, word recognition obtained on the IRONOFF database.

[1]  Christian Viard-Gaudin,et al.  An analytical handwritten word recognition system with word-level discriminant training , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.

[2]  Jianying Hu,et al.  Writer independent on-line handwriting recognition using an HMM approach , 2000, Pattern Recognit..

[3]  Isabelle Guyon,et al.  On-line cursive script recognition using time-delay neural networks and hidden Markov models , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[4]  Éric Anquetil,et al.  Integration of an on-line handwriting recognition system in a smart phone device , 2002, Object recognition supported by user interaction for service robots.

[5]  Stefan Knerr,et al.  The IRESTE On/Off (IRONOFF) dual handwriting database , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[6]  Emmanuel Augustin,et al.  A neural network-hidden Markov model hybrid for cursive word recognition , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[7]  Paul D. Gader,et al.  WORD LEVEL DISCRIMINATIVE TRAINING FOR HANDWRITTEN WORD RECOGNITION , 2004 .

[8]  Lionel Prevost,et al.  An activation-verification model for on-line texts recognition , 2004, Ninth International Workshop on Frontiers in Handwriting Recognition.

[9]  Marco Gori,et al.  A survey of hybrid ANN/HMM models for automatic speech recognition , 2001, Neurocomputing.

[10]  Krishna S. Nathan,et al.  Initialization of hidden Markov models for unconstrained on-line handwriting recognition , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[11]  Alexander H. Waibel,et al.  Online handwriting recognition: the NPen++ recognizer , 2001, International Journal on Document Analysis and Recognition.