Out-of-vocabulary word recognition with a hierarchical doubly Markov language model

Hiroaki Kokubo, Hirofumi Yamamoto, Yoshihiko Ogawa, Yoshinori Sagisaka, Genichiro Kikui

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We describe a novel language model for task-dependent out-of-vocabulary (OOV) words. OOV words, such as personal names and place names in a new task can make the language model adaptation difficult. To cope with this problem, we propose a hierarchical, 2-layered language model consisting of inter-word constraints and intra-word constraints. Stochastic properties of OOV words in the two constraints are represented by multi-class modeling and trained as independent Markov models. Occurrence probabilities of an OOV word are expressed by statistics of two Markov Models (namely, doubly Markov model). The proposed model has been tested in a Japanese conversational speech database of appointment making. The word correct rate has been achieved 7.5% improvement from 78.2% to 86.7% when the new language model was used to recognize sentences with OOV words.

Original languageEnglish
Title of host publication2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages543-547
Number of pages5
ISBN (Electronic)0780379802, 9780780379800
DOIs
Publication statusPublished - 2003
EventIEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003 - St. Thomas, United States
Duration: 2003 Nov 302003 Dec 4

Publication series

Name2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003

Other

OtherIEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003
Country/TerritoryUnited States
CitySt. Thomas
Period03/11/3003/12/4

ASJC Scopus subject areas

  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Out-of-vocabulary word recognition with a hierarchical doubly Markov language model'. Together they form a unique fingerprint.

Cite this