Content-specific, Highly Accurate Rapid Incremental On-demand Translation Engines for the European Region (On-Demand Europe)
Currently the EU has 23 official languages. In the relevant FP7 ICT call, a considerable focus is on Machine Translation (MT), which is an important tool in sustaining our important cultural and linguistic differences which makes the EU what it is today: a continent at peace with itself with one of the world's three biggest markets. MT today is a developed and robust technology capable of building bridges across language barriers and thereby supporting citizens and industries in a globalised world. Economic considerations currently increase the need for MT: Costs of translation have to go down, and the automated process is a key factor. At the same time material in need of translation increases. Furthermore, translators fluent in some languages are somewhat short in supply. For these reasons it is necessary that future MT systems are made quickly in a flexible environement, adapted to the individual user while still securing that translation quality constantly improve. The project has three main goals:1) Considerable effectivisation of automatic translations systems (at least 500 % increase in speed), 2) translation systems that constantly improve as they are used, and 3) translation systems dynamically adapted the relevant translation task and individual. Our role is to focus on (2).
Forsknings- og Innovationsstyrelsen
Credit Suisse AG, TAUS, Gridwise Tech, Applied Language Solutions LTD, Tilburg University, Polytechnic University of Valencia