Modeling French Sign Language: a proposal for a semantically compositional system

Modeling French Sign Language: a proposal for a semantically compositional system Mohamed Nassime Hadjadj, Michael Filhol, Annelies Braffort LIMSI–CNRS B.P. 133, 91403 Orsay cedex, France {mohamed.hadjadj, michael.filhol, annelies.braffort}@limsi.fr Abstract The recognition of French Sign Language (LSF) as a natural language in 2005 created an important need for the development of tools to make information accessible to the deaf public. With this prospect, the goal of this article is to propose a linguistic approach aimed at modeling the French sign language. We first present the models proposed in computer science to formalize the sign language (SL). We also show the difficulty of applying the grammars originally designed for spoken languages to model SL. In a second step, we propose an approach allowing to take into account the linguistic properties of the SL while respecting the constraints of a modelisation process. By studying the links between semantic functions and their observed forms in Corpus, we have identified several production rules that govern the functioning of the LSF. We finally present the rule functioning as a system capable of modeling an entire utterance in French sign language. Keywords: French sign language, computer modelling, formal grammar 1. Introduction Since the ‘60s, numerous efforts have proposed linguistic models to describe sign languages (Stokoe, 1960). However, none of the proposed studies has yet succeeded to fully describe the specificities of SL, such as the use of the signing space and the simultaneous articulation of multiple channels. Thus, researchers in Sign Language Processing (SLP) must model languages without fine linguistic descriptions, without a written form and with a limited amount of corpus (Cuxac and Dalle, 2007). In this context, most SL modeling approaches assume that the difference of modality has a minor impact on the organization of the utterance. The modeling of SLs is then carried out primarily using models originally designed for spoken languages. This operation implies linearity constraints that characterize other languages. The aim of this article is to propose a modeling approach to the French Sign Language (LSF), allowing to take into account the specificities relating to their visual-gestural modality. In the section below (section 2), we present the most important approaches for the modeling of SL. Section 3 present our corpus study methodology to identify production rules that govern the functioning of the LSF. Finally, Section 4 proposes the combination of rules as a system capable of modeling an entire utterance. 2. Research in Sign Language Processing Research in SLs modeling can be classified into two categories. In the first category, models were based on approaches initially proposed to model spoken languages. In the second category, models were designed specifically for SLs. In the following, we present some examples of approaches that love been developed with aim to translate from text to SL. 2.1 Models initially dedicated to spoken languages The two projects Team (Zhao et al., 2000) and ASL Workbench (Speers, 2001) present a translation system from English to ASL using a syntactic transfer. To generate ASL, Team Project represents statements with …

Content truncated to 500 words