Spontaneous dialogue recognition with large out-of-vocabularies

Author

Lee, Hang-Seop

Author_Institution

Spoken Language Processing Sect., ETRI, Taejon, South Korea

fYear

1996

fDate

18-21 Nov 1996

Firstpage

247

Lastpage

250

Abstract

This paper describes a Korean spontaneous speech recognition system that is part of a speech-to-speech translation system being developed in ETRI. The features of this system are as follows. First, we have compiled two large human to human dialog databases. The speech database used in training and evaluation of this system was composed of two task domains: 339 dialogs collected in appointment scheduling domain and 535 dialogs collected in the travel planning domain. Second, recognition performance was evaluated not only on no out-of-vocabulary (OoV) case but also on the OoV case. From the evaluation result of spontaneous speech recognition with high OoV rate, we addressed the problem of Korean language and proposed a solution

Keywords

scheduling; speech recognition; vocabulary; ETRI; Korean spontaneous speech recognition; appointment scheduling domain; large out-of-vocabularies; recognition performance; speech database; speech-to-speech translation system; spontaneous dialogue recognition; travel planning domain; Global communication; Humans; Natural languages; Spatial databases; Speech analysis; Speech recognition; Switches; Training data; User interfaces; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1996., IEEE Asia Pacific Conference on

Conference_Location

Seoul

Print_ISBN

0-7803-3702-6

Type

conf

DOI

10.1109/APCAS.1996.569265

Filename

569265