This article presents an original method for Text-to-Sign Translation. It compensates data scarcity using a domain-specific parallel corpus of alignments between text and hierarchical formal descriptions of Sign Language videos. Based on the detection of similarities present in the source text, the proposed algorithm recursively exploits matches and substitutions of aligned segments to build multiple candidate translations for a novel statement. This helps preserving Sign Language structures as much as possible before falling back on literal translations too quickly, in a generative way. The resulting translations are in the form of AZee expressions, designed to be used as input to avatar synthesis systems. We present a test set tailored to showcase its potential for expressiveness and generation of idiomatic target language, and observed limitations. This work finally opens prospects on how to evaluate this kind of translation.

Cet article propose de premières réflexions quant à l’élaboration d’une grammaire formelle pour les langues des signes, basée sur l’approche AZee. Nous avons mené une étude statistique sur un corpus d’expressions AZee, qui décrivent des discours en langue des signes française. Cela nous permet d’entrevoir des contraintes sur ces expressions, qui reflètent plus généralement les contraintes de la langue des signes française. Nous présentons quelques contraintes et positionnons théoriquement notre ébauche de grammaire au sein des différentes grammaires formelles existantes.

Cet article présente une expérimentation de traduction automatique de texte vers la langue des signes (LS). Comme nous ne disposons pas de corpus aligné de grande taille, nous avons exploré une approche à base d’exemples, utilisant AZee, une représentation intermédiaire du discours en LS sous la forme d’expressions hiérarchisées

This paper presents a corpus of AZee discourse expressions, i.e. expressions which formally describe Sign Language utterances of any length using the AZee approach and language. The construction of this corpus had two main goals: a first reference corpus for AZee, and a test of its coverage on a significant sample of real-life utterances. We worked on productions from an existing corpus, namely the “40 breves”, containing an hour of French Sign Language. We wrote the corresponding AZee discourse expressions for the entire video content, i.e. expressions capturing the forms produced by the signers and their associated meaning by combining known production rules, a basic building block for these expressions. These are made available as a version 2 extension of the “40 breves”. We explain the way in which these expressions can be built, present the resulting corpus and set of production rules used, and perform first measurements on it. We also propose an evaluation of our corpus: for one hour of discourse, AZee allows to describe 94% of it, while ongoing studies are increasing this coverage. This corpus offers a lot of future prospects, for instance concerning synthesis with virtual signers, machine translation or formal grammars for Sign Language.

This article presents a new French Sign Language (LSF) corpus called “Rosetta-LSF”. It was created to support future studies on the automatic translation of written French into LSF, rendered through the animation of a virtual signer. An overview of the field highlights the importance of a quality representation of LSF. In order to obtain quality animations understandable by signers, it must surpass the simple “gloss transcription” of the LSF lexical units to use in the discourse. To achieve this, we designed a corpus composed of four types of aligned data, and evaluated its usability. These are: news headlines in French, translations of these headlines into LSF in the form of videos showing animations of a virtual signer, gloss annotations of the “traditional” type—although including additional information on the context in which each gestural unit is performed as well as their potential for adaptation to another context—and AZee representations of the videos, i.e. formal expressions capturing the necessary and sufficient linguistic information. This article describes this data, exhibiting an example from the corpus. It is available online for public research.