blending audio after feedback reduction
 

I'm sure this has been discussed and I just didn't use the right key words on a search, but here goes:

At a seminar last week the guy in charge of audio (and my feed) left the room just as the main speaker began. Her lapel mic along with her standing close to the podium mic caused feedback at a frequency fairly close to her voice.

I went to the board and [finally] got it stopped, so now I'm trying to reduce the feedback on that portion. All I've done so far is put that section on its own track and isolate the frequency as best I can using the paragraphic EQ. It's much better but I'm all for a nicer solution if someone has one. I'd also like to make it blend a little better with the rest of the audio during the transition. I'm not really an audio person so...any ideas for me to try?

A narrow (have you made it as narrow as you can?) notch in a parametric/paragraphic eq filter is a great first step.

These filters are not keyframeable, so, if you want to ease into it, you might put straight audio on one track and EQed audio on another, and use volume envelopes to cross-fade.

The next step would be "noise reduction", probably the search terms you should use. In NR, a sample of the noise is taken, then, magic algorithims take out some of the noise from the content you want to keep. This more-or-less works, usually an improvement.

Sony's Noise Reduction 2.0 is an excellent product, and comes with Sound Forge. Sound Soap gets good reviews, I've not used it. Audacity has a very basic NR, and is free/open source.

Have a look at NR, see how it works for you. I'd be happy to apply Sony NR to a short sample, if you can post it somewhere I can get it, PM me.

Got the other things you mentioned covered so let me give NR a try. If I'm not happy with my results I'll get in touch with you. Thank you, Seth!