Year: 2015
Type: conference-paper
Abstract: We present ongoing work in the DUEL project, which focuses on the study of disfluencies, exclamations, and laughter in dialogue.Here we focus on the multimodal aspects of disfluent vocalizations, namely laughter and laughed speech, filled pauses, and breathing noises.We exemplify these phenomena in the rich multimodal Dream Apartment Corpus, a natural dialogue corpus, which, in addition to comprehensive disfluency and laughter annotation, comprises tracking data for the body and head.We discuss possible directions for developing models that can perceive as well as generate such multimodal behaviour.
Institution: Bielefeld University
Language: English
Topic: Speech and dialogue systems
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W303133792