Comparing intrinsic and extrinsic evaluation of MT output in a dialogue system

Anne H Schneider, Ielka van der Sluis, Saturnino Luz

Research output: Contribution to conferenceUnpublished paperpeer-review

7 Citations (Scopus)


We present an exploratory study to assess machine transla- tion output for application in a dialogue system using an in- trinsic and an extrinsic evaluation method. For the intrinsic evaluation we developed an annotation scheme to determine the quality of the translated utterances in isolation. For the extrinsic evaluation we employed the Wizard of Oz technique to assess the quality of the translations in the context of a dialogue application. Results differ and we discuss the possible reasons for this outcome.
Original languageEnglish
Number of pages8
Publication statusPublished - Dec 2010
Event7th International Workshop on Spoken Language Translation (IWSLT 2010) - Paris, France
Duration: 2 Dec 20102 Dec 2010


Workshop7th International Workshop on Spoken Language Translation (IWSLT 2010)

Cite this