Name Translation in Statistical Machine Translation - Learning When to Transliterate

We present a method to transliterate names in the framework of end-to-end statistical machine translation. The system is trained to learn when to transliterate. For Arabic to English MT, we developed and trained a transliterator on a bitext of 7 million sentences and Google's English terabyte ngrams and achieved better name translation accuracy than 3 out of 4 professional translators. The paper also includes a discussion of challenges in name translation evaluation.

• Type: Conference
• Year: 2008
• Where: ACL
• Authors: Ulf Hermjakob, Kevin Knight, Hal Daumé III
• Keywords: ACL 2008, Computational Linguistics, Google's English Terabyte, Statistical Machine Translation, Transliterate