Presenting through performing: on the use of multiple lifelike characters in knowledge-based presentation systems

Abstract

In this paper, we investigate a new style for presenting information. We introduce the notion of presentation teams which -rather than addressing the user directly -convey information in the style of performances to be observed by him or her. The paper presents an approach to the automated generation of performances which has been tested in two different application scenarios, car sales dialogues and soccer commentary.

Context aware guidance for multimedia authoring: harmonizing domain and discourse knowledge

“…Research into the automated generation of multimedia presentations has resulted in a number of knowledge-based systems that are able to generate multimedia presentations with minimal or no human intervention. Some of these systems focus on innovative presentation techniques [2,6,7,18,45] that facilitate the synthesis of multimedia documents and plan how to present this material to various users. These approaches are insightful as they model the authoring process from a planning point of view but they are not sufficient for the authoring processes described in Section 2 as their top-down planning approach is too limited for scenarios where neither the individual user requirements nor the requested material can be predicted in advance.…”

“…This section demonstrates the applicability of the proposed approaches by introducing the architecture and implementation of an experimental system SampLe (Semi-Automatic Multimedia Presentation generation Environment) 2 . In this paper we focus on processes and resulting internal data structures of the system.…”

Generating Embodied Information Presentations

“…The selection, both in form and frequency, of these and the other types of nonverbal signals should be influenced by different factors, such as the personality type or emotion the agent should convey (see sections 2.3 and 3.5). Ideally, such factors should also influence the language used by the system, a matter that is ignored in most language generation work (some exceptions are Hovy, 1988, de Rosis andGrasso, 2000, and, for embodied agents, André and Rist, 2000).…”

“…There is no interaction between language generation and nonverbal signal production, however. Other research in this area is that of André and Rist (2000), who have worked on language generation for virtual presenters, focusing on the problem of projecting different agent personalities. The NECA project (Krenn et al, 2002) builds further on this work, developing components for reasoning about agents' emotions and generating combined (emotional) speech and nonverbal signals.…”

The role of high-level and low-level features in style-based retrieval and generation of multimedia presentations

“…In nearly all cases, they lack adaptive [8] or adaptable qualities [28], which could otherwise facilitate the adjustment of the multimedia presentation to the specific context of an individual user. For overcoming these problems, various attempts to explore and develop innovative presentation techniques are described by [1,2,6,20,36]. These approaches facilitate the synthesis of multimedia documents and plan how this material is presented to various users.…”