Miguel Domingo, Mercedes García-Martínez, Amando Estela, Laurent Bié, Alexandre Helle, Álvaro Peris, Francisco Casacuberta, Manuel Herranz. Demonstration of a Neural Machine Translation System with Online Learning for Translators. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics-System Demonstrations, 2019. pp. 70-74. ACL.

We introduce a demonstration of our system,which implements online learning for neuralmachine translation in a production environ-ment. These techniques allow the system tocontinuously learn from the corrections pro-vided by the translators. We implemented anend-to-end platform integrating our machinetranslation servers to one of the most com-mon user interfaces for professional transla-tors: SDL Trados Studio. Our objective was tosave post-editing effort as the machine is con-tinuously learning from human choices andadapting the models to a specific domain oruser style