I will present recent work in statistical machine translation which uses Weighted Finite-State Transducers (WFSTs) to implement a variety of search and estimation algorithms. I will describe HiFST, a lattice-based decoder for hierarchical phrase-based statistical machine translation. The decoder is implemented with standard WFST operations as an alternative to the well-known cube pruning procedure.We find that the use of WFSTs in translation leads to fewer search errors, better parameter optimization, and improved translation performance. We also find that the direct generation of target language lattices under Hiero translation grammars can improve subsequent rescoring procedures, yielding further gains with long-span language models and Minimum Bayes Risk decoding.
Cite as: Byrne, W. (2010) Hierarchical phrase-based translation with weighted finite state transducers. Proc. International Workshop on Spoken Language Translation (IWSLT 2010)
@inproceedings{byrne10_iwslt,
author={William Byrne},
title={{Hierarchical phrase-based translation with weighted finite state transducers}},
year=2010,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2010)}
}