Consistent Goal-Directed User Model for Realistic Man-Machine Task-Oriented Spoken Dialogue Simulation

Olivier Pietquin (1)

1 SUPELEC-Campus Metz

Because of the great variability of factors to take into account, designing a spoken dialogue system is still a tailoring task. Rapid design and reusability of previous work is made very difficult. For these reasons, the application of machine learning methods to dia-logue strategy optimization has become a leading subject of re-searches this last decade. Yet, techniques such as reinforcement learning are very demanding in training data while obtaining a substantial amount of data in the particular case of spoken dia-logues is time-consuming and therefore expansive. In order to expand existing data sets, dialogue simulation techniques are be-coming a standard solution. In this paper we describe a user modeling technique for realis-tic simulation of man-machine goal-directed spoken dialogues. This model, based on a stochastic description of man-machine communication, unlike previously proposed models, is consistent along the interaction according to its history and a predefined user goal.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Olivier Pietquin. Consistent Goal-Directed User Model for Realistic Man-Machine Task-Oriented Spoken Dialogue Simulation. 7th IEEE International Conference on Multimedia and Expo, Jul 2006, Toronto, Canada. pp.425-428, ⟨10.1109/ICME.2006.262563⟩. ⟨hal-00215968⟩