Hybrid Architectures for Multi-Engine Machine Translation

Andreas Eisele, Christian Federmann, Hans Uszkoreit, Hervé Saint-Amand, Martin Kay, Michael Jellinghaus, Sabine Hunsicker, Teresa Herrmann, Yu Chen

In: Proceedings of Translating and the Computer 30. Translating and the Computer Conference (TC-30) November 27-28 London United Kingdom ASLIB 11/2008.


We describe different architectures that combine rule-based and statistical machine translation (RBMT and SMT) engines into hybrid systems. One of them allows to combine many existing MT engines in a multi-engine setup, which can be done under the control of a decoder for SMT. Another architecture uses lexical entries induced via SMT technology to be included in a rule-based system. For all these approaches prototypical implementations have been done within the EuroMatrix project and some indicative results from the recent evaluation campaign are given, which help to highlight the strengths and weaknesses of these approaches.

Eisele-EtAl-TatC.pdf (pdf, 172 KB )

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz