Using Word Posterior Probabilities in Lattice Translation
Vicente Alabau, Alberto Sanchís, Francisco Casacuberta, Departament de Sistemes · 2007
In this paper we describe the statistical machine transla-tion system developed at ITI/UPV, which aims especially at speech recognition and statistical machine translation in-tegration, for the evaluation campaign of the International Workshop on Spoken Language Translation (2007). The system we have developed takes advantage of an im-proved word lattice representation that uses word posterior probabilities. These word posterior probabilities are then added as a feature to a log-linear model. This model includes a stochastic finite-state transducer which allows an easy lat-tice integration. Furthermore, it provides a statistical phrase-based reordering model that is able to perform local reorder-ings of the output. We have tested this model on the Italian-English corpus, for clean text, 1-best ASR and lattice ASR inputs. The results and conclusions of such experiments are reported at the end of this paper. 1.