Business Data
Work Phone: 
+32 16 32 50 89
Personal Bio

Vincent Vandeghinste is a post-doc researcher at the Centre for Computational Linguistics of the University of Leuven, where he has been working on language technology and machine translation since 2000 and is co-teaching the courses on Natural Language Processing and on Computational Linguistics. Before 2000, he was working as a Lingware Consultant at Lernout & Hauspie Speech Products. He has been a reviewer for several MT projects of the 7th Framework Programme of the European Commission, and for journals like the Journal of Artificial Intelligence Research (JAIR) and Literary and Linguistic Computing (LLC). He is currently the project leader of the SCATE project, which is funded by the Flemish IWT (Institute for Scientific and Technological Innovation).

Presentation title: 
The SCATE Project: Envisaged Results
Presentation description: 
In the SCATE project (Smart Computer-Aided-Translation Environment) a consortium of Flemish university research groups investigates several aspects of the translation work flow and possible automation for professional translators, in collaboration with partners from Flemish industry. We wish to highlight the current results of our research and explain what we envisage for the near future, such as an approach to collaborative translation, improved fuzzy matching metrics allowing to better reuse existing translations, a better integration of Translation Memory Systems with Machine Translation (MT), integration of syntax and semantic models in MT, MT error classification and quality estimation, analysis of human post-editing practices, a methodology to extract terminology from comparable corpora, the use of speech recognition as a post-editing method, speech translation and improved and personalized work flows.