https://forum.audacityteam.org/t/rms-normalize/45334/1
Koz
There’s a difference between Chris and RMS Normalize. RMS Normalize sets the overall volume of, say, an hour show. It doesn’t change anything note by note.
Chris is an active processor constantly changing volume over the hour as needed.
Koz
I see, then I think it’s the latter I would want. I guess I could just run Chris over the whole thing after deleting applause. I’ll just leave the final applause until he has run through it all
Just when you think it’s safe to go back…
Broadcast is no safe harbor, either. A very common complaint about broadcast processing is it squeezes the life out of theatrical productions. Two fiddles in the South Forty come out the same volume as the whole orchestra pounding away.
A National Public Radio station in Chicago used to quietly remove the compressors during symphonies and run them by hand.
Koz
I just got an idea, but could this even work? What I was thinking of doing was using the spectral selection to make a notch filter at the very bottom of that silence where it’s red, even deleting if that is better. After that bass is reduced, get a new noise profile with the noise reduction plugin with now just the higher frequencies I’m worried about, then go back and apply this again and I won’t have to worry about those frequencies being affected as much.
You can always tailor the Noise Profile step to make it do what you want. That’s the step where you train Noise Reduction what to attack.
This is a good time to remember Fletcher-Munson. The ear sensitivity isn’t flat over the spectrum. It’s most sensitive to tones around 3000Hz. I call it “Baby Screaming on a Jet” and “Fingernails on Blackboard.” You get the most bang for the buck by processing at these tones.
Koz
Thank you very much, I hadn’t actually known about that, and will use this knowledge going forward!