Advanced State Clustering for Very Large Vocabulary HMM-based On-Line Handwriting Recognition

Abstract

This paper presents some novel methods for the introduction of context dependent Hidden Markov Models (HMM) to on-line handwriting recognition. The use of these so-called n-graphs can lead to substantially improved modeling accuracy, but requires some intelligent parameter reduction methods (state clustering). This is especially the case for the… (More)
DOI: 10.1109/ICDAR.1999.791819

Topics

2 Figures and Tables