You can stay in Vegas for this. Align the beginnings of the audio, then CTL drag the end until the audio again is aligned. The keyword in the help file is "stretch event."

Well, first of all, you should always shoot at the standard DV sample rate of 48k. 32k has limited frequency response and potentially non-linear dynamic range. Why cameras come with 32k set as the default sample rate is just wrong. There is virtually no reason that one would ever want or need to use 32k.

Now about your sync issues, even though your two cameras and your audio recorder all have crystal-based clocks, there is nothing to guarantee that all three devices will operate at exactly the same sample rate. In fact, it is guaranteed that they won't. It seems like your audio recorder is way off. Fortunately, Vegas does audio time-stretching quite well, you can change the length of a track without changing the pitch. Just line up your audio at the beginning and go to the end and while holding the CTRL key, left-click and drag the end of the audio track to get it to line up with the audio from the cameras. This is relatively easy to do, just look at the audio waveforms. You might want to turn off "Quantize to Frames" when you're doing this, just don't accidentally move any of the video tracks while you have this turned off.

John

hear hear John. out of the box 32K is a bad joke.
Just to add some detail about the audio stretching: I would say no pitch "compensation" should be made.
You would *want* the pitch to change with the 'speed.' This is no different than two analog tape machines recording the same thing at different rates. To get tape recorded on one machine to play properly on another you would change the speed (AND the pitch) - and the result would be correct.
While this might not make a big difference on an interview, on a long concert recording it's significant
Vegas of course, gives you both options.

Jesse

Jesse, you're absolutely correct.

When I sync my Sharp Minidisc recorders with my Sony HDV camcorders, I usually have to shorten the Minidisc audio by 1 frame for every 4.5 minutes. I think the last time I calculated just how much a pitch change this made, I was astonished by how little the amount was (I think it was a fraction of 1 cent). My assumption, then, is that it makes no difference whether you lock the pitch or allow it to vary with the minimal length adjustment needed for most tasks.

EDIT:
Okay, being somewhat anal and also in a mellow state of mind this evening (induced by a shot of my favorite libation), I did a couple of calculations:

(1) a pitch change of 1 semitone = 5.9% change of frequency
So 1 cent of pitch change = 0.059% change of frequency

(2) adjusting my Minidisc recorder playback length by 1 frame every 4.5 minutes = 1 / 8,092 = 0.012% pitch change.

(3) So my Minidisc pitch will rise by 1/5 of one cent!

For pure tones, you can hear a beat when two nearly-identical frequencies are mixed. For example, if you have a tuning fork for "A" (440Hz) and a second tuning fork of 441Hz, you can easily hear the 1Hz "beat" between the two of them. For music, it's a much different matter. I can barely hear a 5 cent pitch change, so the pitch change given in (3) above is way below my ability to discern.


SECOND EDIT:
Surprise! Thanks to the power of the Internet, I just discovered that my assumption in step (1) was not quite correct. One cent is NOT 1/100 * 5.9%. It turns out that you have to use a bit more math to get the correct value of 0.05778%. But the original assumption is close enough for our purposes.

I agree the amount of pitch change we're discussing is small, however the processing happening with a pitch change 'process' is large.
In Vegas (and SF) pitch changes aren't linear - they mess with time.

An example:
Take a kick drum recording. Duplicate the track. Add a pitch adjust process to the new second track. Play both tracks together.
You'll notice there are timing problems. This is not so much a short coming of Sony but the necessary artifacts to achieve a pitch change.

Jesse