Machine Translation and Neural Networks for a multilingual EU

Abstract

This paper presents an overview of the current developments and use of Machine Translation (MT) and Neural Machine Translation (NMT), specifically eTranslation, in the European Institutions. An insight into the state-of-the-art of NMT as currently in development in the Directorate-General for Translation (DG TRAD) of the European Parliament is provided by Pascale Chartier-Brun. Problems in machine translation support requiring further research and development for processing languages with complex morphosyntax are discussed in the outlook. This paper was developed from the presentation “IT integrated environment for optimising the translation of legislative documents in the EP“ by Pascale Chartier-Brun at the workshop “Europäische Rechtslinguistik und Digitale Möglichkeiten / EU Legal Linguistics and Digital Perspectives“, held at the University of Cologne July 7th/8th, 2017.