Alignment Data base for a Sign Language Concordancer

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 6069–6072 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 6069 Alignment Data base for a Sign Language Concordancer Marion Kaczmarek, Michael Filhol Université Paris-Saclay, CNRS, LIMSI Orsay, France kaczmarek@limsi.fr, michael.filhol@limsi.fr Abstract This article deals with elaborating a data base of alignments of parallel French-LSF segments. This data base is meant to be searched using a concordancer which we are also designing. We wish to equip Sign Language translators with tools similar to those used in text- to-text translation. To do so, we need language resources to feed them. Already existing Sign Language corpora can be found, but do not match our needs: working around a Sign Language concordancer, the corpus must be a parallel one and provide various examples of vocabulary and grammatical construction. We started with a parallel corpus of 40 short news and 120 SL videos , which we aligned manually by segments of various length. We described the methodology we used, how we define our segments and alignments. The last part concerns how we hope to allow the data base to keep growing in a near future. Keywords: Sign Language, Concordancer, Computer-assisted translation 1. Introduction Sign Languages (SL) convey meaning with hands and body gestures, facial expressions, and eye gaze. Each of these is a part of the language grammar. SLs are languages in their own right, even though they still are considered minority languages with very low visibility, if not none. In 2008, the CRPD (Convention on the Rights of Persons with Disabilities) adopted by the United Nations emphasised the right of people with disabilities to fully access any type of information or communication. This has led to an increasing need for translated content in SL, whether for display in public spaces, public transportation, healthcare or broadcasting. Sign Language interpretation and Sign Language translation are not to be confused. The classical way to distinguish the two is to say interpretation concerns oral languages while translation concerns written texts. Concerning SL, as they do not have an editable written form, the major difference is the possibility to post-edit the translation. Translators can rework their translations many times, until they reach what they believe to be the best account of the meaning in the target language. Interpreters however can only provide one version of their translation and once it is delivered, it cannot be edited (Filhol & Tannier, 2014). The number of professional SL translators in France is very low despite the recent creation of a diploma. And the few working SL translators are not equipped with Computer Assisted Translation (CAT) software tools like those used by text-to-text translators, because none of them support SL. CAT software eases the translator’s tasks (Koehn, 2009), by providing him with an integrated working environment along with dedicated tools (translation memories, concordancers, terminology management tools for example). We are currently working on designing CAT software dedicated to SL. This paper presents the steps for …

Content truncated to 500 words