Accession Number:

ADA576644

Title:

The MIT-LL/AFRL IWSLT-2011 MT System

Descriptive Note:

Conference paper

Corporate Author:

MASSACHUSETTS INST OF TECH LEXINGTON LINCOLN LAB

Personal Author(s):

• Aminzadeh, A. R.
• Anderson, Tim
• Slyh, Ray
• Ore, Brian
• Hansen, Eric
• Shen, Wade
• Drexler, Jennifer
• Gleason, Terry

Report Date:

2011-10-27

Pagination or Media Count:

8.0

Abstract:

This paper describes the MIT-LLAFRL statistical MT system and the improvements that were developed during the IWSLT 2011 evaluation campaign. As part of these efforts, we experimented with a number of extensions to the standard phrase-based model that improve performance on the Arabic to English and English to French TED-talk translation tasks. We also applied our existing ASR system to the TED-talk lecture ASR task. We discuss the architecture of the MIT-LLAFRL MT system, improvements over our 2010 system, and experiments we ran during the IWSLT-2011 evaluation. Specifically, we focus on 1 speech recognition for lecture-like data, 2 cross-domain translation using MAP adaptation, and 3 improved Arabic morphology for MT preprocessing.

Descriptors:

• *MACHINE TRANSLATION
• ARABIC LANGUAGE
• MORPHOLOGY
• SPEECH RECOGNITION

Subject Categories:

• Linguistics
• Voice Communications

Distribution Statement:

APPROVED FOR PUBLIC RELEASE