Les contingences liées au recueil de données multimodales en entreprise

Abstract

Contingencies related to the collection of multimodal data in a company

Nowadays, multi-modality is a necessary dimension for exploitation and analysis of speech-in- interaction since it enriches the classical linguistic descriptions exclusively based on the verbal. However, the constitution of visual and corporal resources by the use of the video remains complex and raises many difficulties in particular those related to the consent of the resource persons to be filmed. This is a classic problem of linguistics but it is rarely discussed in detail. In a resolutely ethico-methodological approach, and based on a field experience conducted in a company, we aim to shed light on this phenomenon which, as we will show, is significantly subject to constraints specifically related to context of the company as well as cultural considerations, particularly those related to the gender issue.