An integrative platform to capture the orchestration of gesture and speech

Christelle Dodane (1) , Dominique Boutet (2) , Ivana Didirkova (1) , Fabrice Hirsch (3) , Slim Ouni (4) , Aliyah Morgenstern (5)

1 Praxiling - Praxiling
2 DYLIS - Dynamique du Langage In Situ
3 Équipe 3 : Parole et discours : fonctionnement/dysfonctionnement et appropriation
4 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
5 PRISMES - PRISMES - Langues, Textes, Arts et Cultures du Monde Anglophone - EA 4398

A number of studies have highlighted the coordination of gesture and intonation (Bolinger, 1983; Darwin, 1872; Kendon, 1980) but the technological setups have been insufficient to couple the acoustic and gestural data with sufficient detail. In this paper, we present the MODALISA platform which enables language specialists to integrate gesture, intonation, speech production and content. The methods of data acquisition, annotation and analysis are detailed. The preliminary results of our pilot study illustrate strong correlations between gestures and intonation when they are simultaneously performed by the speaker. The correlations are particularly strong for proximal segments. Our aim is to expand those results and analyse typical and atypical populations across the lifespan.

Domaines: Interface homme-machine [cs.HC]; Sciences cognitives; Traitement du signal et de l'image [eess.SP]; Informatique; Robotique [cs.RO]; Sciences de l'Homme et Société

Christelle Dodane, Dominique Boutet, Ivana Didirkova, Fabrice Hirsch, Slim Ouni, et al.. An integrative platform to capture the orchestration of gesture and speech. GeSpIn 2019 - Gesture and Speech in Interaction, Sep 2019, Paderborn, Germany. ⟨hal-02278345⟩