Representing New Voice Services and Their Features
Kenneth J. TURNER
Computing Science and Mathematics, University of Stirling, Scotland
kjt@cs.stir.ac.uk
Abstract. New voice services are investigated in the fields of Internet
telephony (SIP – Session Initiation Protocol) and interactive voice systems
(VoiceXML – Voice Extended Markup Language). It is explained how CRESS (Chisel
Representation Employing Systematic Specification) can graphically represent
services and features in these domains. CRESS is a front-end for detecting
feature interactions and for implementing features. The nature of service
architecture and feature composition are presented. CRESS descriptions are
automatically compiled into LOTOS (Language Of Temporal Ordering Specification)
and SDL (Specification and Description Language), allowing automated analysis
of service behaviour and feature interaction. For implementation, CRESS diagrams
can be compiled into Perl (for SIP) and VoiceXML. The approach combines the
benefits of an accessible graphical notation, underlying formalisms, and practical
realisation.