I used to work with the guy that wrote this, you might find it useful.

http://www.toscintl.com/callsoft/

CML makes some nifty caller-id chips that would let you DIY very easily.

In fact I wrote a bunch of PIC assembler code to extract caller-id info
using a software based modem.  It's a Rube Goldberg special, but it
worked, just not 100%.  :-(  Most of the time it would get most of the
information correct, but sometimes it would spit out complete garbage
for some still unknown reason.  In the true spirit of hardware hacking,
the PIC and an LCD were powered directly by the phone line.

I amplified the signal from the line to the point of clipping.  I then
fed that to a comparator to really square things up and simply measured
the width to the pulses.  Narrow pulses = high tone (2200 Hz), wide
pulse = low tone (1200 Hz).  Seems like 2400 Hz would have been a better
choice for the high tone, but who am I.  By setting an output pin to
indicate the width of the last pulse received, you get a crude (untimed)
representation of the serial bit stream.  I didn't like that so I added
another ISR to retime the data.  The output pin was fed back into RB0
for an ISR that bit bang received the serial stream and stuffed the data
into a queue.  The main level code just took data from the queue and
displayed it on an ordinary Hitachi type LCD.  It was truly the busiest
PIC program I ever had going, with multiple ISRs and the CCP going to
time the incoming pulses.  I'm actually surprised that it worked at all,
but it was really frustrating that I couldn't get it to work perfectly.
Debugging stuff that's connected directly to the phone line is tricky
since you can't ground any part of the circuit or it won't work.  I
chalked up my occasional bit errors to 60Hz hum on the line.  Yet
another stupid PIC trick.  ;-)

You should be able to find stuff about it in the piclist archive if
you're interested.

----- Original Message ----- 
From: "John Nall" <jwnall@gmail.com>
To: "Microcontroller discussion list - Public." <piclist@mit.edu>
Sent: Saturday, November 12, 2005 11:03 AM
Subject: **SPAM** [OT] Looking for a phone silencer


> (I wasn't sure what topic this should go under -- probably either PIC
or
> EE could be justified.  But I'll use OT anyway).
>
> I'm looking for a phone silencer, and given all the experience on the
> piclist in designing things and just general knowledge, I feel sure
> someone here can probably point me in the right direction.  I've tried
> using Google to search, but apparently am not phrasing the appropriate
> search terms, because I have not come up with anything even close.
>
> By "phone silencer" I mean a box that sits between the phone and the
> jack, and decides whether or not to let the phone actually ring.
This
> device would have the ability to have a list of numbers in memory, and
> would test the incoming caller-id number against the list.  If the
> number were on the list, then the phone would ring.  If it were not on
> the list, then the phone would not ring but everything else would work
> the same (which is to say that after 5 silent rings, voice-mail would
> pick up and take a message if the caller wanted to leave one).
>
> If anyone can point me to such a device, either commercial or just a
> circuit, I would be grateful.
> -- 
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist