5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Dealing With Out-of-Vocabulary Words and Speech Disfluencies in an N-Gram Based Speech Understanding System

Atsuhiko Kai, Yoshifumi Hirose, Seiichi Nakagawa

Toyohashi University of Technology, Japan

In this study, we investigate the effectiveness of an unknown word processing(UWP) algorithm, which is incorporated into an N-gram language model based speech recognition system for dealing with filled pauses and out-of-vocabulary(OOV) words. We have already been investigated the effect of the UWP algorithm, which utilizes a simple subword sequence decoder, in a spoken dialog system using a context free grammar(CFG) as a language model. The effect of the UWP algorithm was investigated using an N-based continuous speech recognition system on both a small dialog task and a large-vocabulary read speech dictation task. The experiment results showed that the UWP improves the recognition accuracy and an N-gram based system with the UWP can improve the understanding performance in compared with a CFG-based system.

Full Paper

Bibliographic reference.  Kai, Atsuhiko / Hirose, Yoshifumi / Nakagawa, Seiichi (1998): "Dealing with out-of-vocabulary words and speech disfluencies in an n-gram based speech understanding system", In ICSLP-1998, paper 0785.