🚀 go-pugleaf

In article <C05JAM.MJL@ais.org> tim@ais.org (Tim Tyler) writes:
+---------------
|      For those of you not familiar with speech inversion, it simply
| flip-flops the voice spectrum so that high pitched sounds are low,
| & vice versa.  It sounds a lot like Single Side Band (SSB)
| transmissions, although an SSB receiver will not decode speech-
| inversion scrambling.
+---------------

As a general statement, incorrect. If the sending radio uses AM, DSB, or SSB
modulation, it's trivial. All you have to do is listen to the upper sideband
of the speech-inverted signal as if it were a *lower*-sideband signal! Or
vice-versa, it doesn't matter. (You will of course have to manually search
for the inversion frequency.)

What you *may* have meant is that inverted speech sent via *FM* (such as
49 MHz cordless phones) cannot (easily) be received with an unassisted SSB
receiver. This is more nearly true, but still not a great protection:

1. Many cheap FM transmitters often have a significant AM component,
   sometimes even enough for an SSB receiver to be used as above
   (though with considerable other "garbage" leaking through).

2. If the FM "deviation" is large, there will be significant AM-like sidebands
   that might be listened to with an SBB receiver with sufficiently narrow
   filters. [O.k., my math's weak on this one. Maybe Phil Karn or some other
   radio-knowledgable type could comment further on this case.]

3. There also may be other "spurious emissions" from the FM set that a good
   AM or SSB set could snoop [such as amplitude variations of an oscillator
   due to poor power-supply filtering between the audio & RF sections of the
   phone]. Just because something meets FCC "Part 15" doesn't mean that it's
   of "TEMPEST" quality...

4. Finally, one simple way to decode speech-inverted FM is to send the audio
   output of an FM receiver (scanner) into an ordinary AM (or SSB) transmitter
   (such as an AM "wireless mike"), and then listen to the output of *that*
   with your SSB receiver on the "wrong" sideband. [This is in fact sort of
   how speech-inversion receivers work internally, using a 10's to 100's of
   KHz internal "AM" channel, and "SSB" reception of the "wrong" sideband by
   filtering (then "carrier-injection", then "AM reception").]

In any case, we all agree that speech-inversion is hardly secure.


-Rob

-----
Rob Warnock, MS-9U/510		rpw3@sgi.com
Silicon Graphics, Inc.		(415)390-1673
2011 N. Shoreline Blvd.
Mountain View, CA  94043