Connected Text Recognition Using Layered HMMs and Token Passing

We present a novel approach to lexical error recovery on textual input. An advanced robust tokenizer has been implemented that can not only correct spelling mistakes, but also recover from segmentation errors. Apart from the orthographic considerations taken, the tokenizer also makes use of linguistic expectations extracted from a training corpus. The idea is to arrange Hidden Markov Models (HMM) in multiple layers where the HMMs in each layer are responsible for different aspects of the processing of the input. We report on experimental evaluations with alternative probabilistic language models to guide the lexical error recovery process.