5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Rational Interpolation of Maximum Likelihood Predictors in Stochastic Language Modeling

Ernst GŁnter Schukat-Talamazzini (l), Florian Gallwitz(2), Stefan Harbeck (2), Volker Warnke (2)

(l) Institute for Computer Science, University of Jena, Germany
(2) Chair for Pattern Recognition, University of Erlangen-Nuremberg, Erlangen, Germany

In our paper, we address the problem of estimating stochastic language models based on n-gram statistics. We present a novel approach, rational interpolation, for the combination of a competing set of conditional n-gram word probability predictors, which consistently outperforms the traditional linea,r interpolation scheme. The superiority of rational interpolation is substantiated by experimental results from language modeling, speech recognition, dialog act classiflcation, and language identiflcation.

Full Paper

Bibliographic reference.  Schukat-Talamazzini, Ernst GŁnter / Gallwit, Florian / Harbeck, Stefan / Warnke, Volker (1997): "Rational interpolation of maximum likelihood predictors in stochastic language modeling", In EUROSPEECH-1997, 2731-2734.