T ¨ UBTURKISH-ENGLISH SUBMISSIONS for IWSLT 2013
İlknur Durgar El-Kahlout, Burak Aydın, Coşkun Mermer · 2013
This paper describes the T ¨ UBTurkish-English submis- sions in both directions for the IWSLT'13 Evaluation Cam- paign TED Machine Translation (MT) track. We develop both phrase-based and hierarchical phrase-based statisti cal machine translation (SMT) systems based on Turkish word- and morpheme-level representations. We augment training data with content words extracted from itself and experi- ment with reverse word order for source languages. For the Turkish-to-English direction, we use Gigaword corpus as an additional language model with the training data. For the English-to-Turkish direction, we implemented a wide cov- erage Turkish word generator to generate words from the stem and morpheme sequences. Finally, we perform system combination of the different systems produced with different word alignments.