Handling sign language annotations of the handshapes - Formes et représentations en linguistique et littérature Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Handling sign language annotations of the handshapes

Gestion des annotations des formes manuelles en langage des signes

Gestire le annotazioni delle forme manuali in lingua dei segni

Résumé

Historically, transcriptions in SL have largely focused on relaying the meaning of a sign rather than the sign's physical shape [1,2]. Two main graphical systems of annotation —SignWriting [3] and HamNoSys [4]— are able to annotate the form of the parameters: the first one is not designed to annotate [5]; the second, in most cases, is coupled to a database tool for integrating SL corpus, named iLex [6]. Therefore, except for few corpora [7,8], the parameters are only partially annotated. One of the reasons is the time needed to do so, even if this has not been documented in literature. Another challenge in the transcription of signs is the integration of generic information (features) into the specificities of each parameter of signs. At a phonological level [9,10], the features extracted for handshapes generate multi-million possibilities. Some rules to reduce these potential combinations have been established [11,12], and others might still be found. The aim of this communication is to present Typannot [13,14], a type font for handshapes created to transcribe all the existing SL (142 SL for Ethnologue [15]). This glyphic type font is readable, writable, searchable.It works upon a modular system; the separate features can be assembled into glyphs that enable the representation of every handshape.The features are all searchable through a set of generic (symbolic) glyphs. Those glyphs can be visualized by the annotator at any time thanks to the use of ligatures (Opentype format [16]), and are therefore transparent. At present, this font —freely downloadable on every operating system— can be used with ELAN [17] to transcribe directly by using a virtual keyboard (MacOS) or, for some SL, by using a dedicated template. Apart from the keyboard and templates for ELAN, a third annotation device has been developed. It works with a motion capture (MoCap) device (based on Leap Motion [18]) by which the annotator can transcribe a corpus by directly using his own hand (real-time controlled). We are in the process of measuring the annotation time per minute of corpus for the different levels and devices: with generic glyphs, with specific glyphs and with MoCap. The tendency seems to favour of the use of the MoCap device. Finally, we will present some preliminary results of the transcription made with the so-called Typannot Typefont, at several levels of annotation: from the selection of the fingers, the shapes and the bending angles of the fingers in the palm, their lateral distance to the final glyph of each handshape through the combination shape+angle.   Bibliography: [1] Johnston, T. (2008). Corpus linguistics and signed languages: no lemmata, no corpus. Proceedings of 3rd Workshop on the Representation and Processing of Sign Languages (p. 82-88). O. Crasborn, E. Efthimiou, T.s Hanke, E. D. Thoutenhoofd, I. Zwitserlood eds. [2] Fenlon, J., Schembri, A., Johnston, T., & Cormier, K. (2015). Documentary and corpus approaches to sign language research. Research Methods in Sign Language Studies: a Practical Guide, 156-172. [3] Sutton, V. (1995). Lessons in SignWriting: Textbook. SignWriting. [4] Prillwitz, S., Leven, R., Zienert, H., Hanke, T., & Henning, J. (1989). Hamburg Notation System for sign languages: an introductory guide (Signum Press). Hamburg. [5] Bianchini, C. S. (2012, november 18th). Analyse métalinguistique de l’émergence d’un système d’écriture des Langues des Signes : SignWriting et son application à la Langue des Signes Italienne (LIS). PhD dissertation, Paris 8 & Università di Perugia, Paris. [6] Hanke, T. (2002). iLex - A tool for sign language lexicography and corpus analysis (p. 923-926). Proceedings of the 3rd International Conference on Language Resources and Evaluation, Las Palma de Gran Canaria, Spain. [7] Efthimiou, E., Fontinea, S.E., Vogler, C., Hanke, T., Glauert, J., Bowden, R., Braffort, A., Collet, C., Maragos, P., Ségouat, J. (2010). Dicta-sign–sign language recognition, generation and modelling: a research effort with applications in deaf communication (p. 80-83). Proceedings of the 4th Workshop on the Representation and Processing of Sign Languages: Corpora and Sign Language Technologies, La Valletta (Malta), 22-23/05/2010. [8] Hanke, T., König, S., Konrad, R., & Langer, G. (2012, May). Towards tagging of multi-sign lexemes and other multi-unit structures. LREC 2012, Istanbul, Turkey. [9] Liddell, S. K. (1990). Structures for representing handshape and local movement at the phonemic level. Theoretical Issues in Sign Language Research, 1, 37-65. [10] Brentari, D. (1998). A prosodic model of sign language phonology. The MIT Press. Cambridge. [11] Ann, J. (1996). On the relation between ease of articulation and frequency of occurrence of handshapes in two sign languages. Lingua, 98(1), 19–41. [12] Brentari, D. (2012). Handshape in Sign Language phonology: the word and sub-lexical structure. In Sign language: an international handbook (Vol. 37). Walter de Gruyter. [13] Boutet, D., Bianchini, C. S., Doan, P., Goguely, T., Rébulard, M., &Danet, C. (2016, juillet). Typannot: a glyphic system for the transcription of handshapes. Colloque international présenté à 7th Conference of the International Society of Gestures Studies, Paris. [14] Boutet, D., Doan, P., Danet, C., Bianchini C. S., Goguely, T., Contesse, A., Rébulard, M., (Forhtcoming). Systèmes graphématiques et écritures des langues signées. Signata, 8. Presses universitaires de Liège. [15] Lewis, M.P. (2017). Ethnologue: languages of the world, vers. 20. SIL International. http://www.ethnologue.com [online; accessed: 1/12/2017] [16] Wikipedia (2017). OpenType. https://en.wikipedia.org/wiki/OpenType [online; accessed: 1/12/2017]. [17] Crasborn, O., & Sloetjes, H. (2008). Enhanced ELAN functionality for sign language corpora. Proceedings of 6th International Conference on Language Resources and Evaluation (Marrakesh, Morocco). [18] Wikipedia (2017). Leap Motion. https://en.wikipedia.org/wiki/Leap_Motion [online; accessed: 1/12/2017].
4C C043 2018-presentazione+ AFLiCoJET Nanterre commentata x HAL.pdf (1.97 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02342798 , version 1 (16-05-2020)

Identifiants

  • HAL Id : hal-02342798 , version 1

Citer

Dominique Boutet, Claudia S. Bianchini, Claire Danet, Patrick Doan, Morgane Rébulard, et al.. Handling sign language annotations of the handshapes. Aflico-Jet 2018 Corpora and Representativeness, May 2018, Nanterre, France. pp.15-17. ⟨hal-02342798⟩
136 Consultations
27 Téléchargements

Partager

Gmail Facebook X LinkedIn More