Artificial Vocal Tract for Robots

Posted 4 May 2002 at 14:21 UTC by steve Share This

Hideyuki Sawada of Kagawa University in Japan is developing an artificial vocal tract for use in humanoid robots. This is not an electronic voice synthesizer but a mechanical assembly that simulates the human vocal tract by pumping air through vocal chords, a windpipe, and lips. The system is controlled by a neural network. A NewScientist.com article provides a nice summary of the project.

Heimleich, posted 4 May 2002 at 22:22 UTC by The Swirling Brain » (Master)

For crying out loud, wouldn't it be easier to get a cheapo 50 cent speaker and do the sound stuff virtually with software? When the difference is a 50 cent device or a 50 dollars device with comparible output, I'll go with the 50 cent device. It's like the difference between an old $100k pipe organ and a $2k Clavinova, why have the expensive one when you can have it all with the inexpensive one. But of course the expensive one looks nice and all. Really, though, I've thought it would be neat to do this too.

See more of the latest robot news!

Recent blogs

2 Oct 2015 mwaibel (Master)
22 Sep 2015 shimniok (Journeyer)
10 Sep 2015 svo (Master)
9 Aug 2015 Petar.Kormushev (Master)
6 May 2015 spirit (Journeyer)
28 Feb 2015 steve (Master)
19 Feb 2015 Flanneltron (Journeyer)
14 Nov 2014 Sergey Popov (Apprentice)
14 Nov 2014 wedesoft (Master)
3 Jul 2014 jmhenry (Journeyer)
Share this page