Towards Automatic Addressee Identification in Multi-party Dialogues

Abstract

The paper is about the issue of addressing in multi-party dialogues. Analysis of addressing behavior in face to face meetings results in the identification of several addressing mechanisms. From these we extract several utterance features and features of non-verbal communicative behavior of a speaker, like gaze and gesturing, that are relevant for observers to identify the participants the speaker is talking to. A method for the automatic prediction of the addressee of speech acts is discussed.