1st International Conference “From Scientific Computing to Computational Engineering” 1st IC-SCCE Athens, 8-10 September, 2004 © IC-SCCE MULTIPURPOSE DESIGN OF GREEK SIGN LANGUAGE RESOURCES: A FACTOR TOWARDS UNIVERSAL ACCESS Eleni E. Efthimiou, Stavroula-Evita F. Fotinea Institute for Language and Speech Processing (ILSP) Artemidos 6 & Epidavrou str., Maroussi 151 25, Athens, Greece Keywords: Universal Access, Assistive Technology, educational platform, Greek Sign Language, virtual signer. Abstract. In this paper we present the methodology of data collection and implementation of databases for the creation of extensive lexical and terminological resources for the Greek Sign Language (GSL) in order to introduce the major issue of dynamic sign representation. In respect to electronic linguistic resources of GSL, the focus is on issues of validation of linguistic content, multipurpose design and reusability of created resources, exemplified by the multimedia dictionary products of the projects NOEMA (1999-2001) and PROKLISI (2002-2004). In relation to sign representation, we discuss the use of video as a restricted means of reproducing linguistic content uttered in the three-dimensional space and present an attempt to use avatar and animation technologies, along with exploitation of electronic linguistic resources of the Greek Sign Language (GSL), in order to provide a dynamic device of generating unrestricted sign sequences. The aim is to set up a sign generation mechanism for the Greek Sign Language by means of a virtual human signer (avatar) that can be adaptable to various platforms. The overall emphasis is placed on exploitation of Universal Access and Design for All principles that allow proactive system architectures instead of a-posteriori adaptations. 1 INTRODUCTION The Greek Sign Language (GSL) is a natural visual language used by the members of the Greek Deaf Community with several thousands of native or non-native signers. Research on the grammar of GSL per se is limited; till mid ’90s mostly fragmentary work had been done, whereas recent studies mainly focus on individual aspects of its syntax (negation, morphology, auxiliary verbs), as well as on applied and educational linguistics (GSL acquisition and onomatopoeia). Systematic analysis and creation of linguistic resources for the language has been strongly connected with the maturing of technologies, which enabled the 3D representation of its linguistic content. As a consequence, in the recent past there have also been some serious attempts in the domain of lexicography (NOEMA: a Multimedia Dictionary of GSL Basic Vocabulary and a Children Dictionary of GSL) mainly with educational scope. The major gain of this enterprise has been, however, development of knowledge as to generative mechanisms of the language, along with creation of reusable linguistic resources. 2 GSL LINGUISTIC ANALYSIS – THE BACKGROUND In contrast to other sign language systems, i.e. the ASL [26,29] , systematic lexicographical work in respect to GSL has started only recently, within the framework of the NOEMA project (1999-2001). This was the first attempt to create multipurpose reusable linguistic resources for GSL. Part of the project description was the creation of a digital sign stream narration corpus and an electronic dictionary of basic GSL vocabulary. The spin-off products of that project, among which are a 3,000 entry multimedia bilingual dictionary (GSL-Greek) of basic vocabulary and a multimedia children’s dictionary of GSL [11] , reflect the methodology for creating linguistic resources followed, the content and interface evaluation mechanism adopted, as well as the basic linguistic research carried out to support the lexicographical work [14] . The knowledge acquired with respect to morpho-phonological operations involved in the formation of simple and complex signs allowed for: a) the construction of rules for creating new valid signs, b) the denomination of relevant terms and c) the classification of GSL linguistic resources into terminological lists. All these have significant impact on the development of both communication and educational tools using technologies that allow the 3D representation of linguistic content. Our data consist of digitised language productions of Deaf native GSL signers, triangulated with the participation of Deaf GSL signers in focus group discussions. The followed methodological principles on data collection and analysis are suitable to the minority status of GSL, while standardization is crucially based on