Discriminative model combination and language model selection in a reading tutor for children

Abstract

In this paper, we suggest the use of general acoustic and language models to deal with the mismatch between the training and testing data of a reading tutor for children. The testing data consist of isolated real and non-existing (pseudo) words, while the training data consist of continuous readings of Dutch sentences. General acoustic (e.g. context independent) and language models (e.g. bigram phone language models) are proposed as they implicitly better model the hesitant nature of the testing data. Discriminative model combination (DMC) is modified to provide different weights for different phones and was utilized to combine the new models into the baseline system. Combination of general acoustic and language models into the baseline system using DMC significantly lowers the system phone error rate, by 3.5% relative to the baseline system for the non-existing (pseudo) words.