Interactive spoken-language processing in a hybrid connectionist system

Author

Wermter, Stefan ; Weber, Volker

Author_Institution

Dept. of Comput. Sci., Hamburg Univ., Germany

Volume

29

Issue

7

fYear

1996

fDate

7/1/1996 12:00:00 AM

Firstpage

65

Lastpage

74

Abstract

Our SCREEN (Symbolic Connectionist Robust Enterprise for Natural Language) system analyzes real-world utterances and can be used for relatively large and complex tasks. SCREEN learns a robust flat syntax, semantics, and pragmatics representation. The system also deals with uncommon syntactic and semantic language irregularities. The system is able to produce many utterance hypotheses based on spoken input and can determine which hypotheses are most likely. SCREEN´s ability to analyze spoken language, despite encountering mistakes and uncertainties, demonstrates the system´s robustness and potential. Because SCREEN is a German spoken-language system, the example of spoken language that we analyze is in German. We provide a literal and, where appropriate, a more easily understood English translation of any German words and sentences that we use

Keywords

interactive systems; learning (artificial intelligence); natural languages; neural nets; speech recognition; symbol manipulation; German spoken-language system; SCREEN system; Symbolic Connectionist Robust Enterprise for Natural Language system; hybrid connectionist system; interactive spoken-language processing; mistakes; pragmatics representation learning; real-world utterances; semantics learning; spoken input; syntax learning; uncertainties; uncommon semantic language irregularities; uncommon syntactic language irregularities; utterance hypotheses; Active noise reduction; Computer errors; Humans; Manuals; Natural languages; Noise robustness; Speech enhancement; Speech processing; Speech recognition; Uncertainty;

fLanguage

English

Journal_Title

Computer

Publisher

ieee

ISSN

0018-9162

Type

jour

DOI

10.1109/2.511970

Filename

511970