Unsupervised Learning for Speech Motion Editing

Loading...
Thumbnail Image
Date
2003
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
We present a new method for editing speech related facial motions. Our method uses an unsupervised learning technique, Independent Component Analysis (ICA), to extract a set of meaningful parameters without any annotation of the data. With ICA, we are able to solve a blind source separation problem and describe the original data as a linear combination of two sources. One source captures content (speech) and the other captures style (emotion). By manipulating the independent components we can edit the motions in intuitive ways.
Description

        
@inproceedings{
:10.2312/SCA03/225-231
, booktitle = {
Symposium on Computer Animation
}, editor = {
D. Breen and M. Lin
}, title = {{
Unsupervised Learning for Speech Motion Editing
}}, author = {
Cao, Yong
and
Faloutsos, Petros
and
Pighin, Frédéric
}, year = {
2003
}, publisher = {
The Eurographics Association
}, ISSN = {
1727-5288
}, ISBN = {
1-58113-659-5
}, DOI = {
/10.2312/SCA03/225-231
} }
Citation