Annotation of LSF subtitled videos without a pre-existing dictionary - Architectures et Modèles pour l'Interaction Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Annotation of LSF subtitled videos without a pre-existing dictionary

Résumé

This paper proposes a method for the automatic annotation of lexical units in LSF videos, using a subtitled corpus without annotation. This method, based on machine learning and involving linguists for added precision and reliability, comprises several stages. The first consists of building a bilingual lexicon (including potential variants of a given lexical unit) in a weakly supervised manner. The resulting lexicon is then refined and cleaned by LSF experts. This data serves next to train a supervised classifier for automatic annotation of lexical units on the Mediapi-RGB corpus. Our Pytorch implementation is publicly available.
Fichier principal
Vignette du fichier
24012 (1).pdf (1.29 Mo) Télécharger le fichier

Dates et versions

hal-04593866 , version 1 (30-05-2024)

Licence

Identifiants

  • HAL Id : hal-04593866 , version 1

Citer

Julie Lascar, Michèle Gouiffès, Annelies Braffort, Claire Danet. Annotation of LSF subtitled videos without a pre-existing dictionary. LREC-COLING 2024 11th Workshop on the Representation and Processing of Sign Languages: Evaluation of Sign Language Resources, May 2024, Turin (IT), Italy. pp.100-108. ⟨hal-04593866⟩
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More