Token-level interpolation for class-based language models
Michael Levit, Andreas Stolcke, Shuangyu Chang, Sarangarajan Parthasarathy
Abstract
We describe a method for interpolation of class-based n-gram language models. Our algorithm is an extension of the traditional EM-based approach that optimizes perplexity of the training set with respect to a collection of n-gram language models linearly combined in the probability space. However, unlike prior work, it naturally supports context-dependent interpolation for class-based LMs. In addition, the method works naturally with the recently introduced wordphrase- entity (WPE) language models that unify words, phrases and entities into a single statistical framework. Applied to the Calendar scenario of the Personal Assistant domain, our method achieved significant perplexity reduction and improved word error rates.
BibTeX
@inproceedings{icassp2015_tokenlevelinterp,
title = {Token-level interpolation for class-based language models},
author = {Michael Levit and Andreas Stolcke and Shuangyu Chang and Sarangarajan Parthasarathy},
booktitle = {ICASSP 2015},
year = {2015}
}