Oral task-oriented dialogue for air-traffic controller training

ABSTRACT We present here an oral dialogue system designed for air-traffic controller training. The system has to handle the dialogue with a student air-traffic controller and communicate thestudent's requests to the air-traffic simulator in the form of a sequence of instructions expres-sed in a formal command language. The system has been developed with two main objectives in mind: we tried to make it both user-friendly and robust. Different knowledge sourcesincluding the vocabulary, the phraseology, the task model and the history of dialogue, areevolving during the dialogue. These knowledge bases are represented in a unified form of hierarchical frames. The message analysis and interpretation consist of two phases: categori-sation and instantiation. Then, a validity control process uses a set of semantico-pragmaticconstraints. Correction procedures have also been developed to improve speech recognitionperformance. The speech recognition system which is being used is AMADEUS developed atthe LIMSI. It is a DTW-based continuous speech recognizer. The representation of the lan-guage model used to limit the DTW search space is described as well as the way it is modifiedby the dialogue manager in order to permit or prohibit messages and improve recognition.In system evaluation, the classical DTW approach is compared to a probabilistic approachusing dynamically updated transition probabilities.