Stochastically-based semantic analysis for machine translation

We report our experience of applying a stochastic method for understanding natural language to a multilingual appointment scheduling task, in particular, to the English spontaneous speech task (ESST). The aim of the spoken language systems developed for this task is to translate spontaneous conversational speech among different languages. We have investigated the portability of a stochastic semantic analyser from a setting of human–machine interactions air travel information services (ATIS) and multimodal multimedia automated service kiosk (MASK) into the more open one of human-to-human interactions (ESST).