The other day, I showed you how I helped my client Lee fix a noise problem in his recordings. The screen shots of his normalization dialog in Audacity caused a couple of people to point out that his normalization level was too high.
I couldn’t have disagreed more. But now, I’ve changed my mind.
An example of the emails I got – this one from Charles:
Hey David! Always love reading your stuff. I’ll make this quick.
I noticed his “normalization” level is 0.0. Shouldn’t it be set to -3 as that’s what sites such as Voice123 require for submissions?
Have a good one!
I get this a lot. And I’d like to take a moment to make sure that this misconception is cleared up.
Voice123 doesn’t require that, they suggest that. And, I used to think that they were completely and irretrievably incorrect in doing so.
Normalizing a piece of audio to -3.0 dB reduces the maximum volume to only 72% of full volume. Now, I’m all for headroom and avoiding the possibility of distorting sound, but I thought that went too far, that it was too much headroom, and the resulting audio sounded weak.
I wanted my clients and students and readers – you – to sound powerful and professional, full and rich. And there really was no danger of distortion – if you don’t distort your recording to begin with, no amount of normalization will add distortion.
It took putting the finishing touches on the editing system I created years ago for audiobooks that made the difference in what I recommend. That, and the production requirements for ACX audiobook projects.
ACX wants all their submissions to be normalized to -3.0 dB. So, OK. I’ve changed my mind about this – and I’m happy to change my advice as well. This one’s now a no-brainer:
Normalize to -3.0.
Hope that helps.