A Probabilistic Graphical Model for Word-Level Language Modeling in P300 Spellers

Motivated by P300 spelling scenarios involving communication based on a limited vocabulary, we propose a probabilistic graphical model-based framework and an associated classification algorithm that uses learned statistical prior models of language at the level of words. Exploiting such high-level contextual information helps reduce the error rate of the speller. The proposed approach models all the variables in the P300 speller in a unified framework and has the capability to correct errors in previous letters in a word given the data for the current one. The structure of our model allows the use of efficient inference algorithms, which makes it possible to use this approach in real-time applications. Our experimental results demonstrate the advantages of the proposed method.