Using a Wizard of Oz as a baseline to determine which system architecture is the best for a spoken language translation system

Starlander, Marianne

Proceedings chapter

English

Using a Wizard of Oz as a baseline to determine which system architecture is the best for a spoken language translation system

Contributors

Starlander, Marianne

Presented at

Tartu (Estonia), 25-26 May

Published in

Proceedings of the 16th Nordic Conference of Computational Linguistics (NODALIDA), p. 161-164

Publication date

2007

Abstract

This paper is part of an extended study on system architectures, the long term aim being to determine if a unidirectional, a bidirectional or a fixed-phrase architecture is more suitable in the context of the spoken language translator in the medical domain (MedSLT). Our aim here is to compare data collected during a Wizard of Oz (WOz) experiment with data collected using a beta bidirectional version of our system.

UNIGE affiliation entities

Faculté de traduction et d'interprétation / Département de traitement informatique multilingue

Citation (ISO format)

STARLANDER, Marianne. Using a Wizard of Oz as a baseline to determine which system architecture is the best for a spoken language translation system. In: Proceedings of the 16th Nordic Conference of Computational Linguistics (NODALIDA). Tartu (Estonia). [s.l.] : [s.n.], 2007. p. 161–164.

Proceedings chapter

Identifiers

PID : unige:3458

586views

183downloads

Creation

02/10/2009 09:28:55

First validation

02/10/2009 09:28:55

Update

14/10/2025 07:07:44

Status update

14/03/2023 15:14:52

Last indexation

21/08/2026 11:50:56

Using a Wizard of Oz as a baseline to determine which system architecture is the best for a spoken language translation system

Contributors

Presented at

Published in

Publication date

Abstract

UNIGE affiliation entities

Citation (ISO format)

Main files (1)

Identifiers

Technical informations

Creation

First validation

Update

Status update

Last indexation