12 Nov
2007
12 Nov
'07
9:03 a.m.
On Sun, 2007-11-11 at 20:49 -0600, Jim Leonard wrote:
Chuck Guzis wrote:
It's not surprising if they used the same mikes and mike technique to
record the initial samples as they did to record the demo tape. Listen
to the "Technical Details" samples - *TUUUH*echnical *DDDUUUUUHH*etails
(it's quite hard to describe thermonuclear explosion breath pops in
ASCII). I know that the codec used in TI's stuff was particularly
susceptible to background noise.
Gordon
On 11 Nov 2007 at 17:11, Jason T wrote:
I was going to ask you why it cracked you up until I heard the
http://incolor.inetnebr.com/bill_r/computalker4-text2speech.wav example
-- holy crud, I couldn't understand a single word!
The digitized examples (Kennedy, etc.) weren't that much better
either... I wonder, with modern computers making the FFT analysis, could
the computalker produce intelligible output? I can only assume that the
original analysis was limited by the speed of its day. Has anyone put the audio file online of the demo
tape?
Found it: http://incolor.inetnebr.com/bill_r/computalker.htm