>Thanks guys,
>
>Bob just defined my problem. The task is to recognize a busy signal from any
>PBX. It's rather about recognizing a signal with some stable frequency. I
>don't believe someone would be able to whistle into a mike a busy signal
>accurate enough :-).
>
>Igor Pokorny

IMHO if someone tries to imitate a busy signal, and the thing hangs
up on him, so be it!  Seems to me that the criteria for a busy signal
could be pretty loose.  There are only a few things you would normally
get during call progress.  So, sample the amplitude and see if it fits the
following:

Goes on and off (on= > 60% amplitude, off = < 20% amplitude, and
  yes I made those numbers up, use your own)

The repeat rate of the sound is between 3Hz and 1Hz.

The duty cycle on the 'pulses' is the "same" each time you
sample it.  (You decide how close "same" is).

Pretty loose definition, but it ought to work.  The only reason
I'm brave enough to suggest such a thing is that I've owned a
lot of modems over the years and been surprised at how many
different kinds of busy signals they recognize.  Seemed to me
they weren't being too critical about the whole thing.

Think of this: if you had a light instead of a speaker, could
you "spot" the busy signal?

Barry

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu