ORELO Project

The ORELO (Origine des REdacteurs et des LOcuteurs) project aims to develop tools to automatically identify dialectal Arabic from texts written in Arabic or Latin characters as well as from speech.

Two main approaches will be evaluated and compared. The first one is a statistical method using automatic learning techniques. The second one is based on the use of dialect dictionaries including dialect specific words. This second method can be used even if only a few dialect words are used.

Dialects: Algerian, Egyptian, Moroccan and Tunisian.

The project started on March 1st, 2014 and is expected to last 24 months. It is funded in the framework of the RAPID program conducted jointly by DGCIS (Ministry of Industry) and DGA (Ministry of Defense).

Partners: GeolSemantics (dialect identification in text, project leader), LIMSI (data annotation and dialect identification), Vocapia Research (dialect identification in speech)

Last modified: Saturday,11-October-14 04:55:50 CEST

Spoken Language Processing Group (TLP)

ORELO Project