An Open Archtecture for Multi-Domain Multi-Modal Conversational Systems

Multi-modal systems which can handle spontaneous speech have been popular in these days. But there are only a few multi-domain systems which consider about extensibility and scalability. This paper proposes an open architecture for the multi-domain spoken dialog systems which satis es extensibility and scalability. The key concept of the architecture is distribution and integration of the data fragments. The data fragments are information about speech input obtained from the continuous speech recognition engine. Each fragment is distributed and integrated through the hierarchy of the domain managers and the work modules. The way of the system composition is clearly de nd and the structure for the data fragment is de ned as open end. Using our architecture, a multi-domain system can be easily composed with several single domain systems and managers.