ISCA Archive Interspeech 2005
ISCA Archive Interspeech 2005

An n-gram-based statistical machine translation decoder

Josep M. Crego, José B. Mariño, Adrià de Gispert

In this paper we describe MARIE, an Ngram-based statistical machine translation decoder. It is implemented using a beam search strategy, with distortion (or reordering) capabilities. The underlying translation model is based on an Ngram approach, extended to introduce reordering at the phrase level. The search graph structure is designed to perform very accurate comparisons, what allows for a high level of pruning, improving the decoder efficiency. We report several techniques for efficiently prune out the search space.

The combinatory explosion of the search space derived from the search graph structure is reduced by limiting the number of reorderings a given translation is allowed to perform, and also the maximum distance a word (or a phrase) is allowed to be reordered. We finally report translation accuracy results on three different translation tasks.

doi: 10.21437/Interspeech.2005-728

Cite as: Crego, J.M., Mariño, J.B., Gispert, A.d. (2005) An n-gram-based statistical machine translation decoder. Proc. Interspeech 2005, 3185-3188, doi: 10.21437/Interspeech.2005-728

  author={Josep M. Crego and José B. Mariño and Adrià de Gispert},
  title={{An n-gram-based statistical machine translation decoder}},
  booktitle={Proc. Interspeech 2005},