Reliability of Lexical and Prosodic Cues in Two Real-life Spoken Dialog Corpora

Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC 2004)

Abstract

The present research focuses on analyzing and detecting emotions in speech as re vealed by task-dependent spoken dialogs corpora. Previously, we have conducted s everal experiments on a real-life corpus in order to develop a reliable annotati on method and to detect lexical and prosodic cues correlated to the main emotion class. In this paper we evaluate both the robustness of the annotation scheme a nd of the lexical and prosodic cues by testing them on a new corpus. This work i s carried out in the context of the Amities project in which spoken dialog syste ms for call center services are being developed.