From characters to words: Dynamical segmentation and predictive neural networks

We present the extension of a neural predictive system primitively designed for on-line character recognition to words. Feature extraction is performed after resampling the pen trajectory information, recorded by a digitizing tablet. Each word is modeled by the natural concatenation of letter-models corresponding to the letters composing it. Successive parts of a word trajectory are this way modeled by different Neural Networks and only transitions from each one to itself or to its right neighbors are permitted. A holistic and dynamical segmentation allows to adjust letter-models to the great variability of handwriting encountered in the words. Our system combines Multilayer Neural Networks and Dynamic Programming with an underlying Left-Right Hidden Markov Model (HMM). Training was performed on 7000 words from 9 writers, leading to good results in the letter-labelling process, without using any language model.