I’m still figuring out how best to moderate the noise removal tool.
I’m going in, applying some noise removal, then re-amplifying the track to make up for the audio loss. I’d like to do a few rounds of that.
What gets in the way is each time I go to re-amplify, the clipped peaks keep changing and getting in the way. I know I can use the envelope to go in and reduce them manually, but I imagine there must be a tool to do this automatically (like set max levels to a certain dB).
I’d like to do a few rounds of that.
Some people do find multiple gentle applications of Noise Removal to work better than one massive one.
In that case I would skip the Amplify or Normalize steps in the middle.
Get the presentation in good shape with your editing and one pass of Amplify to make it as big as possible. Get your profile and run Noise Removal once. Then do it again with Effect > Repeat (Control-R This shortcut key may be different for you). You might be able to completely automate this (but not the Profile step) with Chains.
Then, when it’s quiet, do the final amplitude and compression for export.
I can’t stress enough the importance of Effect > Noise Removal > Profile. Audacity will try to remove anything you put in that step from the show. If you get the slightest bit of voice in there by accident, Audacity will try to remove your voice.
If you customize each manual pass of Noise Removal, then you’re dead. You need to do everything manually and there are no shortcuts.
Noise Removal should not cause any part of the show to increase in volume. I’ve never had that happen. If it insists on doing that, then start the process way up at the very first Amplify to a smaller number, like Amplify to -1 or -2 instead of zero. That will give the system a little room to grow without causing clipping or other damage.
You should know that Audacity will not clip while the show is on the timeline even if the blue waves go way up. You can always safely bring the show back down later to normal sizes and volumes for export. If you Export with a clipped show, the clipping damage in the export is permanent.
I don’t understand this Koz. Are you saying to do noise profiles on each pass, or just to use the Ctrl+R repeat function?
What makes you dead if you do what?
You lost me right at the start cybrmarc.
Why do you need to amplify? Normally after noise reduction, the audio will be virtually the same as the audio before noise reduction, just with a bit less noise in the quiet parts.
Are you trying to dig out some sounds from a sea of noise?