Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT

Marianne Starlander, Pierrette Bouillon, Glenn Flores, Manny Rayner, Nikos Tsourakis

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

Abstract

This paper reports preliminary results of an evaluation during which two different bidirectional versions of the limited-domain medical spoken language translator MedSLT were compared in a hospital setting. The more restricted version (V.1) only allows Yes-No answers and short elliptical sentences, while the less restricted version (V.2) allows Yes-No answers, short elliptical sentences and full sentences. Although WER is marginally better for V.1, task performance is marginally worse. There appear to be two main reasons for this disparity; short sentences are often badly recognised and patients tend to find it difficult to limit themselves to ellipsis, even if they receive clear instructions about not using full sentences.

Original languageEnglish (US)
Title of host publicationProceedings of the 12th European Association for Machine Translation Conference, EAMT 2008
Pages176-181
Number of pages6
StatePublished - 2008
Event12th Conference of the European Association for Machine Translation, EAMT 2008 - Hamburg, Germany
Duration: Sep 22 2008Sep 23 2008

Other

Other12th Conference of the European Association for Machine Translation, EAMT 2008
CountryGermany
CityHamburg
Period9/22/089/23/08

Fingerprint

Evaluation
Spoken Language
Translator
Ellipse

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Software

Cite this

Starlander, M., Bouillon, P., Flores, G., Rayner, M., & Tsourakis, N. (2008). Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT. In Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008 (pp. 176-181)

Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT. / Starlander, Marianne; Bouillon, Pierrette; Flores, Glenn; Rayner, Manny; Tsourakis, Nikos.

Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008. 2008. p. 176-181.

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

Starlander, M, Bouillon, P, Flores, G, Rayner, M & Tsourakis, N 2008, Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT. in Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008. pp. 176-181, 12th Conference of the European Association for Machine Translation, EAMT 2008, Hamburg, Germany, 9/22/08.
Starlander M, Bouillon P, Flores G, Rayner M, Tsourakis N. Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT. In Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008. 2008. p. 176-181
Starlander, Marianne ; Bouillon, Pierrette ; Flores, Glenn ; Rayner, Manny ; Tsourakis, Nikos. / Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT. Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008. 2008. pp. 176-181
@inproceedings{f7d32e37347f44ba922d4984833ba487,
title = "Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT",
abstract = "This paper reports preliminary results of an evaluation during which two different bidirectional versions of the limited-domain medical spoken language translator MedSLT were compared in a hospital setting. The more restricted version (V.1) only allows Yes-No answers and short elliptical sentences, while the less restricted version (V.2) allows Yes-No answers, short elliptical sentences and full sentences. Although WER is marginally better for V.1, task performance is marginally worse. There appear to be two main reasons for this disparity; short sentences are often badly recognised and patients tend to find it difficult to limit themselves to ellipsis, even if they receive clear instructions about not using full sentences.",
author = "Marianne Starlander and Pierrette Bouillon and Glenn Flores and Manny Rayner and Nikos Tsourakis",
year = "2008",
language = "English (US)",
isbn = "9783000257704",
pages = "176--181",
booktitle = "Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008",

}

TY - GEN

T1 - Comparing two different bidirectional versions of the limited-domain medical spoken language translator MedSLT

AU - Starlander, Marianne

AU - Bouillon, Pierrette

AU - Flores, Glenn

AU - Rayner, Manny

AU - Tsourakis, Nikos

PY - 2008

Y1 - 2008

N2 - This paper reports preliminary results of an evaluation during which two different bidirectional versions of the limited-domain medical spoken language translator MedSLT were compared in a hospital setting. The more restricted version (V.1) only allows Yes-No answers and short elliptical sentences, while the less restricted version (V.2) allows Yes-No answers, short elliptical sentences and full sentences. Although WER is marginally better for V.1, task performance is marginally worse. There appear to be two main reasons for this disparity; short sentences are often badly recognised and patients tend to find it difficult to limit themselves to ellipsis, even if they receive clear instructions about not using full sentences.

AB - This paper reports preliminary results of an evaluation during which two different bidirectional versions of the limited-domain medical spoken language translator MedSLT were compared in a hospital setting. The more restricted version (V.1) only allows Yes-No answers and short elliptical sentences, while the less restricted version (V.2) allows Yes-No answers, short elliptical sentences and full sentences. Although WER is marginally better for V.1, task performance is marginally worse. There appear to be two main reasons for this disparity; short sentences are often badly recognised and patients tend to find it difficult to limit themselves to ellipsis, even if they receive clear instructions about not using full sentences.

UR - http://www.scopus.com/inward/record.url?scp=84857591361&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857591361&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84857591361

SN - 9783000257704

SP - 176

EP - 181

BT - Proceedings of the 12th European Association for Machine Translation Conference, EAMT 2008

ER -