My current project is called Vox Populi in which we try to predict the outcome of elections on the basis of tweets using different filters.
Before I have been involved in projects of many different kinds.
Database collection: SoNaR (large Dutch text corpus from various domains), Jasmin (speech corpus with elderly, children and non-native speakers), Corpus Gesproken Nederland (large speech corpus from various domains) and various speech database collections (also mobile phone, car) for companies (KPN, Philips, Temic)
Database validation: speech databases in many languages, recorded in various environments in various SpeechDat projects
Database curation: CARE, VALID, LESLLA
Speech processing: My Pronunciation Coach, Repetitor, Automatic segmentation of dialect speech, Automatic recognition of Dutch dysarthric speech