What can I use to test the difference between the sounds in different particles it's traveling through?I need .Sound Detector help for a project in...

This would be a great science project. What you will need
is:

two small microphones (you could harvest them from any
number of sources; a trip to your local thrift store is likely to net you discarded
electronics with microphones. It would be best if they were identical, but this is not
absolutely necessary.)

A noisemaker. This could be, for
example, a small speaker that plugs into a
laptop.

Multiplex the two microphones so that they both
feed into a single stereo jack. Make sure that the wire lengths are the same for both
Set up the experiment to that both microphones are in your environment and are separated
by a known distance L.

Create your environment. I would use
some sort of rubber tube where you can place the microphones/speaker inside (isolate
them from external noise), and seal it off to contain the gas. If you use something hard
like plastic or glass, remember that the sound will travel faster along the length of
your environment than it does in the gas
inside.

To make your measurement, simply use an
audio recording software like Audacity and record what happens after you emit a beep
(make it as short as possible, and at a single frequency). You should see two
corresponding peeks on the audio recording. The distance between these two peeks is your
time measurement.

now you know a = L/t, the speed of
sound.

some
tips:

1) don't use hydrogen. You can't get it, and it
explodes. You could use helium instead; it's plentiful and
fun.

2) don't use oxygen. You might be able to get it, but
it can burn violently. use air instead; it's plentiful and lets you breath. Or, nitrogen
if you can find some.

3) Don't rule out liquids or solids.
You can set up a tube filled with water, jello, or even a solid glass
rod.

4) Be prepared to interpret your results. Remember
that there are many pathways for sound to get from from the buzzer to the microphones.
You have to deal with echos, also. Do the math, know what to expect, and use that to
help interpret your results.

5) Set up an audio filter. If
your beep is at a single frequency, you can set up your audio software to exclude
everything but that frequency. This would help you demonstrate your experiment in a
noisy environment, say at the science fair.

6) make your
beep as short as you can.

7) make L long. The two peeks
will blend together if your beep is not short enough or L is too short. If your tube is
coil-able, you can get really long lengths into a small
volume.

8) there are easy ways to get the audio data into
excel. If you look into it, you can make some nice charts. But the audio signal will
probably be sufficient for your needs.

Good
Luck!