We describe the TÜBİTAK submission to the IWSLT 2012 Evaluation Campaign. Our system development focused on utilizing Bayesian alignment methods such as variational Bayes and Gibbs sampling in addition to the standard GIZA++ alignments. The submitted tracks are the Arabic- English and Turkish-English TED Talks translation tasks.
Cite as: Mermer, C., Kaya, H., El-Kahlout, İ.D., Doğan, M.U. (2012) The TÜBİTAK statistical machine translation system for IWSLT 2012. Proc. International Workshop on Spoken Language Translation (IWSLT 2012), 144-148
@inproceedings{mermer12_iwslt,
author={Coşkun Mermer and Hamza Kaya and İlknur Durgar El-Kahlout and Mehmet Uğur Doğan},
title={{The TÜBİTAK statistical machine translation system for IWSLT 2012}},
year=2012,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2012)},
pages={144--148}
}