The University of Edinburgh Speaker Personality and MoCap Dataset

Kathrin Haag, Hiroshi Shimodaira

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We announce the release of a dialogue dataset with motion capture for the head and body which includes introverted and extroverted speaker personality styles. The dataset will be used to synthesize personality based non-verbal behaviour from speech. Initial copy synthesis experiments show that the recorded head motion clearly reflects characteristics of introverted and extroverted speakers.
Original languageEnglish
Title of host publicationFAA '15 Proceedings of the Facial Analysis and Animation
Place of PublicationNew York, NY, USA
PublisherACM
Number of pages2
ISBN (Print)978-1-4503-3530-0
DOIs
Publication statusPublished - 2015

Fingerprint

Dive into the research topics of 'The University of Edinburgh Speaker Personality and MoCap Dataset'. Together they form a unique fingerprint.

Cite this