Since this group is composed of experts in computers and music (most
of the computers involved are of wood and zephyr skin, but many are
silicon) I figured I might ask for advice.

This is an ambitious project which likely lies at the very limits of my
skill.  I gave up on it a long time ago because it involves extracting
data files from a sound card or other audio program, and I was never
able to figure out how to do that.

What I want to do is demonstrate digital audio and, by extension, sound
recording in general, for the science museum I work with.  Here's one
possible procedure:

First, we demonstrate the relationship between mechanical vibration and
sound, and between the frequency of mechanical vibration and the pitch
of a sound.  I have already done this with the Savart-wheel 'piano,'
(dubbed "The MechHammond" by Principal Constructor Tom Newsome of
Castlewood Organs) which rotates a nest of wooden toothed wheels on
a single spindle.  You hold rolled up pieces of paper against the teeth,
and it makes different notes.  You can play tunes.

This in itself is quite a revelation to many of the museum guests.
I wouldn't have thought so myself, because the relationship is so
obvious to me: constant frequency --> musical note, higher frequency
--> higher note.  But it is not at all obvious to someone who never
learned it or thought about it, including a good many of our museum
staff.

 [ Attach playing cards to be struck by the spokes of bicycle wheels,
 [ and put energetic youths on the seats, and you have a "Savart
 [ bicycle chorus!"  ;-)  -- Robbie

Second, demonstrate that the position of a mechanical lever (probably
the actuator of a digital encoder) can be expressed as numbers, and
that these numbers can be stored and sent back to a stepper motor.
In other words, I wiggle a Lever 1, and when I do so a list of numbers,
each one representing the instantaneous position of Lever 1, shows up
in a list on a computer screen.  Then, we 'play back' the list of numbers
into a stepper motor.  This stepper motor is connected to a Lever 2,
which thus repeats the motions of Lever 1.

I think I can do this second step if I can figure out how to hack into,
say, the parallel port of a PC.  I could do it in hardware; matter of
fact, I think I once did something of the sort with a microprocessor,
but a PC would seem to be vastly easier.  But, I have never mastered
the art of working with computers connected to external circuitry.

The third step is to repeat the lever trick, but at very high speed,
and using a microphone diagram and a speaker diaphragm instead of the
levers.  This, I imagine, would involve the use of the sound card.
We speak into the microphone and watch the analog-to digital [A-D]
device in the sound card write maybe 10,000 samples per second into
the memory and onto the screen.  Slow everything down if possible.

Then, play back the stored voice signal to the speaker, but very, very
slowly, like one sample per second, then five samples per second, then
10, then 20, 100, 500, 1000, 5000, 10,000 samples per second, until the
little bumpy excursions of the speaker cone -- which could be either
felt or observed -- blend into something that sounds like the original
sound signal.  Samples of music could also be downloaded into the
memory and treated the same way.

A possible fourth step would involve the graphing of the signal, or
perhaps every 100th sample of the signal, and then some mathematical
processing of the sample to make it sound weird.

Does any of this sound remotely possible, and what new skills would
I need to enable a PC to do it?  I have access to an infinite number of
scrapped PC's, including a few of my own, and there is a budget for the
purchase of other components.

Any advice, including observations concerning the geological status of
my head, will be humbly and gratefully received.

Mark Kinsler
Lancaster, Ohio, USA
http://www.mkinsler.com/

 [ Your head, fortunately, has been found preserved in the sands of
 [ the NeoBernardShaw geological era and the movie, "My Fair Lady",
 [ with Professor Higgens' laboratory full of early sound machines,
 [ and that's just the sort of thing to capture a kid's imagination
 [ -- and mine, too!  :-)  -- Robbie