Talking Brains: A robotic model of the human vocal tract?

Monday, August 2, 2010

A robotic model of the human vocal tract?

It's not pretty, but this anthropomorphic bot developed at Waseda University in the Takanishi lab can talk -- well, produce some vowels anyway. Have a look here.

1 comment:

CEJ said...: It looks like something that was chasing Mrs. Peel in an episode of the Avengers. If it makes human-like vowel sounds, I guess it shows one thing: it isn't that simple to model speech production.; September 7, 2010 at 3:07 AM

Post a Comment

Greg Hickok is Professor of Cognitive Sciences at UC Irvine, Editor-in-Chief of Psychonomic Bulletin & Review, and author of The Myth of Mirror Neurons. David Poeppel, after several years as Professor of Linguistics and Biology at the University of Maryland, College Park, is now Professor of Psychology at NYU. Hickok and Poeppel first crossed paths in 1991 at MIT in the McDonnell-Pew Center for Cognitive Neuroscience where Hickok was a post doc, and Poeppel a grad student. Meeting up again a few years later at a Cognitive Neuroscience Society Meeting in San Francisco, they began a collaboration aimed at developing an integrated model of the functional anatomy of language. Research in both the Hickok and Poeppel labs is supported by NIDCD.