Infrastructure for Multi-dimensional Software Architecture

Introduction

Metamodels play the role of abstract grammars in model-based software description. Accordingly, metamodels share similar purposes as context-free grammars in traditional formal languages. In this project we study metamodel-based approaches for defining a voice control language for a domain defined by a metamodel. Possible target applications are UML editors and various automation systems.

The project aims to create a model based speech control environment, where speech control grammars (consisting of different commands) are generated from the model (called "application model"). Model also defines the information related to the implementation of the commands. In some cases it is possible to generate the speech control grammar and the corresponding implementation using only the application model. Metamodels can also be used to verify the commands before they are executed. Guiding the user to use valid commands on each situation can be done based on the information given in the model.

The project is funded by Academy of Finland in 2006-2008.