FASTY – Faster and easier text generation for disabled people
暂无分享,去创建一个
The partially EU funded project FASTY will assist motor, speech, learning and language impaired persons to produce texts faster, with less physical/cognitive load and with better spelling and grammar. Higher text generation rates will be achieved on one hand by using a very adaptable UserInterface with a variety of feedback options such as different selection lists and orders for the predicted words/phrases, different audible feedback options and synthetic speech output. On the other hand a very strong prediction engine, which is independent from the used language, is developed. Based on n-gram methods the engine will use Part-of-Speech statistics, grammar-based prediction and prediction ranking and collocation-based prediction, too. Furthermore, a new technique for compound prediction will enhance the keystroke-saving rate for some languages. The paper will give a general introduction to the project and a status report on prototypes. First results of language performance tests will also be presented.
[1] Marco Baroni,et al. FASTY - A Multi-lingual Approach to Text Prediction , 2002, ICCHP.
[2] Marco Baroni,et al. Predicting the Components of German Nominal Compounds , 2002, ECAI.
[3] Marco Baroni,et al. Wordform- and Class-based Prediction of the Components of German Nominal Compounds in an AAC System , 2002, COLING.