AVLaughterCycle: An audiovisual laughing machine

Jérôme Urbain, Elisabetta Bevacqua, Thierry Dutoit, Alexis Moinet, Radoslaw Niewiadomski, Catherine Pelachaud, Benjamin Picart, Joëlle Tilemanne and Johannes Wagner

erschienen 2009 "QPSR of the numediart research program", Volume 2, Number 3, Pages 97 - 104


The AVLaugh­ter­Cy­cle project aims at devel­op­ing an audio­vi­sual laugh­ing machine, capa­ble of record­ing the laugh­ter of a user and to respond to it with a machine-generated laugh­ter linked with the input laugh­ter. Dur­ing the project, an audio­vi­sual laugh­ter data­base was recorded, includ­ing facial points track­ing, thanks to the Smart Sen­sor Inte­gra­tion (SSI) soft­ware devel­oped by the Uni­ver­sity of Augs­burg. This tool is also used to extract audio fea­tures, which are sent to a mod­ule called Medi­a­Cy­cle, eval­u­at­ing sim­i­lar­i­ties between a query input and the files in a given data­base. Medi­a­Cy­cle out­puts a link to the most sim­i­lar laugh­ter, sent to GRETA, an Embod­ied Con­ver­sa­tional Agent, who dis­plays the facial ani­ma­tion cor­re­spond­ing to the laugh­ter simul­ta­ne­ously with the audio laugh­ter playing.