Edinburgh Research Archive >
Centre for Speech Technology Research >
CSTR publications >
Please use this identifier to cite or link to this item:
|Title: ||Analysis and Synthesis of Head Motionfor Life like Conversational Agents|
|Authors: ||Shimodaira, Hiroshi|
Kawamoto, Shin ichi
Hofer, Gregor O
|Issue Date: ||2005|
|Citation: ||Machine Learning for Multimodal Interaction, Second International Workshop, MLMI 2005, Edinburgh, UK, July 11-13, 2005|
|Abstract: ||This study aims to investigate which and what motions of lifelike conversational
agents play essential role to make the agents natural. Some preliminary experimental
results and future plan are shown.
Embodying conversational agents is intended to imitate paralinguistic channel
or back channel which plays important role in human-human conversation.
Although there have been not a few lifelike conversational agents developed, appearance
and behaviour of the agents are far from being natural, and merits of
embodying have not been realized that much. One of the main reasons is the
lack of natural motions of the agents, including the motion of head, mouth, eyes
and eye brows. To improve the quality of these motions, several pioneering works
have been reported [1–3].
To tackle the problem, the present study was started conducting a subjective
test of head motion and analysis of head motion using signal processing
|Keywords: ||head motion|
|Appears in Collections:||CSTR publications|
Items in ERA are protected by copyright, with all rights reserved, unless otherwise indicated.