Automating the generation of coordinated multimedia explanations

An explanation system developed to overcome the disadvantages of conventional authoring in multimedia applications is presented. This experimental test bed for the automated generation of multimedia explanations is called COMET (coordinated multimedia explanation testbed), and has as its goal the coordinated, interactive generation of explanations that combine text and three-dimensional graphics, all of which is generated on the fly. In response to a user request for an explanation, COMET dynamically determines the explanation´s content using constraints based on the type of request, the information available in a set of underlying knowledge bases, and information about the user´s background and goals. Having determined what to say, COMET also determines how to express it at the time of generation. A brief overview of COMET´s domain and architecture is followed by a description of the specific ways in which COMET can coordinate its text and graphics. The current status and limitations of COMET are discussed.<>