We give an overview of the RWTH phrase-based statistical machine translation system that was used in the evaluation campaign of the International Workshop on Spoken Language Translation 2005. We use a two pass approach. In the first pass, we generate a list of the N best translation candidates. The second pass consists of rescoring and reranking this N-best list. We will give a description of the search algorithm as well as the models that are used in each pass. We participated in the supplied data tracks for manual transcriptions for the following translation directions: Arabic-English, Chinese-English, English-Chinese and Japanese-English. For Japanese-English, we also participated in the C-Star track. In addition, we performed translations of automatic speech recognition output for Chinese- English and Japanese-English. For both language pairs, we translated the single-best ASR hypotheses. Additionally, we translated Chinese ASR lattices.
Cite as: Zens, R., Bender, O., Hasan, S., Khadivi, S., Matusov, E., Xu, J., Zhang, Y., Ney, H. (2005) The RWTH phrase-based statistical machine translation system. Proc. International Workshop on Spoken Language Translation (IWSLT 2005), 145-152
@inproceedings{zens05_iwslt,
author={Richard Zens and Oliver Bender and Saša Hasan and Shahram Khadivi and Evgeny Matusov and Jia Xu and Yuqi Zhang and Hermann Ney},
title={{The RWTH phrase-based statistical machine translation system}},
year=2005,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2005)},
pages={145--152}
}