Abstract
Intelligent MultiMedia or MultiModal systems involve the computer processing, understanding and production of inputs and outputs from at least speech, text, and visual information in terms of semantic representations. One of the central questions for these systems is what form of semantic representation should be used, which of course goes back to the age old question of knowledge representation in artificial intelligence. When a system processes multimodal input it needs to map that input into the representation and vice-versa there needs to be a mapping out of the representation for multimodal output presentation. In addition, there are related issues of synchronisation of input/output and information fusion and coordination. Here, we look at current trends in multimodal semantic representation which are mainly XML- and frame- based, relate our experiences in the development of multimodal systems (CHAMELEON and CONFUCIUS) and conclude that producer/consumer, intention (speech acts), semantic-content, and timestamps are four important components of any multimodal semantic representation. In addition, multimodal semantic representations depend on the task at hand, system architecture, will be necessary at different levels (media-independent and dependent) and will have numerous forms of representation. Semantic representations and content will need to provide for reference and spatial relations, two key recurring problems in multimodal systems.
Original language | English |
---|---|
Title of host publication | Unknown Host Publication |
Editors | H Bunt, K Lee, L Romary, E Krahmer |
Place of Publication | Tilburg, The Netherlands |
Publisher | Tilburg University |
Pages | 1-16 |
Number of pages | 16 |
Publication status | Published (in print/issue) - Jan 2003 |
Event | First Working Meeting of the SIGSEM Working Group on the Representation of MultiModal Semantic Information - Tilburg University, Tilburg, The Netherlands Duration: 1 Jan 2003 → … |
Workshop
Workshop | First Working Meeting of the SIGSEM Working Group on the Representation of MultiModal Semantic Information |
---|---|
Period | 1/01/03 → … |