🚀 go-pugleaf

In article <1993Jan6.083201.7026@netcom.com> rcain@netcom.com (Robert Cain) writes:


   Owen.Lewis@purplet.demon.co.uk (Owen Lewis) writes:
   :  -=> Quoting Ted Dunning to All <=-
   : 
   :  TD> ......... simple speech inversion really shouldn't be called a
   :  TD> security device.
   : 
   : 
   : 
   :  Exactly so. I couldn't have put it better.
   : 

   What about the modern analog devices that use split band inversion where
   one of 32 programmable split points in the voice band may be selected
   and changed at 60 Hz or faster at "random?"  Two separate inversions
   are performed, one for the lower sideband and one for the upper
   sideband around the split point.  This seems *much* harder.

seems much harder, but unless noise is injected, the ww2 experiments
showed that it wasn't hard to piece the spectra back together.  in
fact, trained listeners could mostly understand even these agile
systems.


   Now that I think about it, a PRN
   that can spit out a new number every 500 ps sounds rather
   difficult.

as long as you don't have to get the number off chip, it isn't so
terribly hard.  a pulson modulator only has to have one bit per cycle.