Using Unsupervised Learning For Engineering of Spoken Dialogues

Major steps towards dialogue models is to know about the basic units that are used to construct a dialogue model and possible sequences. In this approach a set of dialogue acts is not predefined by any theory or manually described during the engineering process, but is learned integrating different kind of data that are available in an avised spoken dialogue system. For this purpose an existing unsupervised learning algorithm was adopted and extended to the demands in natural language processing and speech processing, called CLASSITALL. It is embedded in a tool kit for dialogue engineering, defining basic modules. The DIAlogue MOdel Learning Environment supports an engineeringoriented approach towards dialogue modelling for a spoken-language interface. The architecture is outlined and the approach is applied to the domain of appointment scheduling.