The name of this post is Talking Heads.

November 28, 2008 3:19 PM

The Waseda Talker has been turning heads (har har) lately. It's a mechanical simulation of the human vocal tract, from the motion of its synthetic lips down to the hypnotic undulation of its rubbery vocal folds (compare the genuine article here). Think this is new? Well, these days we do most of this stuff electronically — but talking simulacra have a long and weird history, starting back when electronic synthesizers were just a pipe dream. Here's a talking pair of bellows from 1791, and a head you can play like a trumpet as recently as 1937. The granddaddy of 'em all are the Kratzenstein resonators (not Frankenstein, Kratzenstein!) from 1779. Make your own with pipe insulation and a duck call.
posted by nebulawindphone (12 comments total)

Your mom blows my head like a trumpet.


(seriously, cool stuff)
posted by Saxon Kane at 5:00 PM on November 28, 2008

Oh yeah, I need one of those. This is cooool!
posted by moonmilk at 5:28 PM on November 28, 2008

That Waseda Talker is cool stuff; I wonder if any of the ENT guys I work with have heard about it. The high speed vidoe of the robo-vocal cords is uncanny; compare it with the real thing.
posted by TedW at 5:44 PM on November 28, 2008

Some day the magic of the Waseda will be re-purposed, and will be experienced in private booths with the addition of enough tokens.
posted by Tube at 7:15 PM on November 28, 2008 [2 favorites]

They are trying to get the robot to reproduce human speech so that they can reduce it to mechanical movements, and build that into a cell phone, which would be capable of an extremely high degree of voice compression.

...followed shortly thereafter by a Flash game -- call it QWOPERA -- in which players must manipulate the basic controls to reproduce a simple melody.

(And then, while we're all moaning and drooling and swallowing our virtual tongues, there will be some inexplicable savant who can belt out a flawless aria.)
posted by Rhaomi at 9:00 PM on November 28, 2008

Interesting that the papipupepo sounds more like fafifufefo—the consonant that would have been romanized "p" in Old Japanese also gradually changed into "f" and eventually the "h" (and occasional "f"-oid) in modern Japanese.

Wait, that isn't interesting at all. Sorry. Carry on.
posted by No-sword at 10:14 PM on November 28, 2008 [1 favorite]

These would make such great instruments. I have this image now of a band where each performer controls some mechanical or electrical vocal simulacrum. Has this been done already? I'd love to see it.
posted by twoleftfeet at 11:38 PM on November 28, 2008

That pipe insulation + duck call resonator is a lot like the top of a krumhorn.
posted by Araucaria at 10:29 AM on November 29, 2008

I apologize, that YTL verges on rickroll. Here's the wikipedia page on it, for more serious readers.
posted by Araucaria at 10:36 AM on November 29, 2008

Wait, that isn't interesting at all. Sorry. Carry on.

Since when is deaffrication disinteresting?
posted by The Great Big Mulp at 2:49 PM on November 29, 2008

twoleftfeet, they're apparently working on it at UBC. You can even apply for a job making speech synth instruments.

DIVAs [DIgital Ventriloquized Actors] will be used in three composed stage works of increasing complexity to be performed in Canada and internationally, starting with one performer initially and culminating in three performers simultaneously using their natural voices as well as the hand-based synthesizer.

And here's a neat-looking speech synthesizer intended as a performance instrument. And another!
posted by moonmilk at 12:00 PM on November 30, 2008

Thanks, moonmilk.
posted by twoleftfeet at 3:36 PM on December 6, 2008