I've got some video clips I'd like to use that were recorded with the on-camera mic (XL1S). The camera was close enough to get a decent level, but not close enough to eliminate that hollow sound so often heard when good audio procedures are not followed. The room acoustics were lousy and didn't help either.

Can anyone suggest some tweaks that might work to improve the audio? Boosting the low frequencies seems to help some, but was hoping someone might be able to shed a little more light on the subject. Premiere Pro and Audition are the tools I have available.

Thanks

Try to determine which track, left or right from the camera mic, sounds the best when used individually. Then duplicate or pan center that track, (however your NLE works to make one channel come out both left and right). By eliminating the track that sounds worse, you'll help the sound. Even if both tracks sound equal, eliminating one will reduce the pickup of reverb and spaciousness that comes with a stereo mic and a fairly centered subject. You will probably have to experiment with your levels.
This will probably not totally solve your problem, but it will help.
As for EQ, generally rolling off the lows helps with reverb, but it depends on the location.
Try adding a small wide bump up (about 3db) at 2 to 2.5 khz to help bring up the voice.

Get your hands on Acoustic Mirror from Sony.
Then take a sample of the silent/no transient sections, create an impulse.
Select the entire project's audio, and apply the impulse to one channel of audio.
Then invert that channel. 'emptiness' will be gone. Of course, some hollowness will still be there, because the audio was processed with it. But it's much easier to manage from this point.

Thank you both. I'll give your suggestions a try.

Regarding Acoustic Mirror, can the Noise Reduction filter/effect in Audition do the same thing? The procedure you describe sounds similar to one used in Audition to filter background noise.

Thanks

No, it's not the same at all. Using Acoustic Mirror is using a convolution reverb, which Audition also has, but it's not as robust. Still, i'd give it a shot if you already have Audition. It's got a good convolution of it's own too.

<<<-- Originally posted by Douglas Spotted Eagle : Get your hands on Acoustic Mirror from Sony.
Then take a sample of the silent/no transient sections, create an impulse. (snip) -->>>

According to the Help files on AM, one must create an impulse in the physical space and record it. I'm guessing that if one could isolate a short sound event in the audio, the response could be found and that used to characterize the physical response of the environment in which the audio was recorded.

But I don't know how to create an impulse from silence. Would you elaborate please?

With Acoustic mirror, you can select an area where no desired audio is present, and create an impulse from there. I've been able to do this reasonably well with dialog and find places where I can get a sharp attack with the reverb of the room present, without having to go back to the room. I honestly don't know how to do it with Audition. I should, because I just finished tech editing a book on Audition....now I feel dumb.
I'll dig into it and ask the engineers in Seattle if it can be done with Audition's convolution verb. I'll be shocked if the answer is no....Just gotta work out how.
This is another great argument for always getting a few moments of room tone, and clapping your hands for capturing a little room sound....habit for me, but I'm kinda weird about sound.

I've got Sound Forge.

What I don't understand is how to create an impluse in AM.