Jaspis is a framework for adaptive speech applications. Jaspis is designed to support distributed, highly context-sensitive applications that adapt to the user and the environment. It is designed multilingual applications in mind. The architecture is based on the shared information management and on the agent-evaluator-manager -paradigm.
The most comprehensive information can be found from the following PhD dissertation:
Markku Turunen. "Jaspis - A Spoken Dialogue Architecture and its Applications" (PDF, 324 pages). Ph.D. Thesis, University of Tampere, Department of Computer Sciences, 2004. Report A-2004-2 & Acta Electronica Universitatis Tamperensis, vol. 325.
An overview can be found from the following research article:
Markku Turunen, Jaakko Hakulinen, Kari-Jouko Räihä, Esa-Pekka Salonen, Anssi Kainulainen and Perttu Prusi. An architecture and applications for speech-based accessibility systems (PDF, 20 pages). IBM Systems Journal, Vol. 44, No 3, 2005.
Our other publications provide further information on the Jaspis architecture and its applications.
For more informatiom, send e-mail (jaspis@cs.uta.fi).
Several practical applications have been constructed on the Jaspis architecture, for example:
In addition to our internal applications, the Jaspis architecture is in use in various national and international research projects.
Jaspis is based on Java and XML technologies. It is open source software, see release and download information.
Jaspis is developed at the University of Tampere, Department of Computer Sciences, Human-Computer Interaction Unit in the Speech-based and Pervasive Interaction Group.