| Summary: | This paper establishes a methodological panorama for the study of speech in a multimodal perspective. Therefore, the aim is not to give results but to uncover what lies behind the scenes of a multimodal study by revealing methodological questions that any researcher should raise before undertaking such a study. Thus, this paper addresses the issue of the collection of a multimodal corpus, coding techniques (including the development of an annotation scheme) and finally the methodology to set up an inter-coder agreement to validate the annotations. Therefore, this paper presents several studies and relates several methodological experiences. The focus will be placed on studies about co-speech gestures but considering the transfer of these methodological approaches to other modalities such as facial expressions, gaze or postures.
|
|---|
