The OTIM Formal Annotation Model: A Preliminary Step before Annotation Scheme

Large annotation projects, typically those addressing the question of multimodal annotation in which many different kinds of informationhave to be encoded, have to elaborate precise and high level annotation schemes. Doing this requires first to define the structure of theinformation: the different objects and their organization. This stage has to be as much independent as possible from the coding languageconstraints. This is the reason why we propose a preliminary formal annotation model, represented with typed feature structures. Thisrepresentation requires a precise definition of the different objects, their properties (or features) and their relations, represented in termsof type hierarchies. This approach has been used to specify the annotation scheme of a large multimodal annotation project (OTIM) andexperimented in the annotation of a multimodal corpus (CID, Corpus of Interactional Data). This project aims at collecting, annotatingand exploiting a dialogue video corpus in a multimodal perspective (including speech and gesture modalities). The corpus itself, is madeof 8 hours of dialogues, fully transcribed and richly annotated (phonetics, syntax, pragmatics, gestures, etc.).