Mediterranean Arabic Language and speech technology

HANSEN, DORTE HALTRUP
Tel: +45-35329070
Fax: +45-35329089

UNIVERSITE MOHAMMED V-SOUISSI     MOROCCO
SAKHR SOFTWARE CO.     EGYPT
THE ENGINEERING COMPANY FOR DIGITAL SYSTEMS DEVELOPEMENT     EGYPT
IBM WORLD TRADE CORPORATION JOINT STOCK COMPANY     EGYPT
UNIVERSITY OF BALAMAND     LEBANON
BIRZEIT UNIVERSITY     WEST BANK AND GAZA STRIP
THE OPEN UNIVERSITY     UNITED KINGDOM
UNIVERSITEIT UTRECHT     NETHERLANDS
CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE     FRANCE
COMMISSARIAT A L'ENERGIE ATOMIQUE     FRANCE
INSTITUTE FOR LANGUAGE AND SPEECH PROCESSING - "ATHENA" RESEARCH CENTER     GREECE
UNIVERSITE LYON 2 LOUIS LUMIERE     FRANCE
AL-AHLYYA AMMAN UNIVERSITY     JORDAN
EVALUATIONS AND LANGUAGE RESOURCES DISTRIBUTION AGENCY     FRANCE

KOBENHAVNS UNIVERSITET
CENTER FOR SPROGTEKNOLOGI
NOERREGADE 10 PO Box 2177
1017   COPENHAGEN
DENMARK

MEDAR addresses International Cooperation with the Arabic region on Speech and Language Technologies. MEDAR is structured around 3 pillars, 4 main objectives, and a number of instruments. The 3 pillars are: 1) producing a knowledge base on Human Language Technology (HLT) players, existing language resources (LRs) and processing tools, activities and products for Arabic, 2) designing a strong cooperation roadmap between EU and Arabic countries, within the Arabic countries, and between academia and industry and 3) focusing on Machine Translation (MT) and Multilingual Information Retrieval (MLIR) for which required technology components, LRs, benchmarking methodologies will be identified.

The 4 objectives are: 1) consolidating a network of players in all areas of HLT, 2) developing the Cooperation Roadmap based on a clear picture of the foreseeable technological trends, market potentials, and cooperation possibilities, 3) Updating the Basic Language Resource Kit: the minimum set of resources and tools necessary for carrying out research and training on LRs and HLT, with a focus on MT and MLIR and 4) Supporting the development of tools and resources, in particular MT and MLIR on the basis of partners' technologies and open source code (e.g. Statistical MT, MLIR, and speech recognition) and the framework for their benchmarking. A variety of instruments will be used to achieve the objectives: Surveys will be conducted as well as market analysis. Partners' background will be exploited to identify the best multilingual open-source tools to customize to Arabic.

A variety of instruments will be used to achieve the objectives: Surveys will be conducted as well as market analysis.

Partners' background will be exploited to identify the best multilingual open-source tools to customize to Arabic. Expertise on evaluation within the consortium will create the right benchmarking framework. Key achievements will be highlighted for each strategic partnership in addition to the actions to be taken. Human resources will be at the heart of our recommendations. Information will be disseminated through e.g. website and e-newsletter, and a major conference on LRs, tools, and Evaluation for Arabic.