“Normalizing” is a way of boosting (or reducing) the audio. By peak values, you can tell the DAW to change the audio to meet a certain level - you’re aware of this. In terms of loudness, there are a number of set industry standards a piece of audio has to meet (per broadcast, radio, etc). Because it’s nearly impossible to dial in the gain and have the talent perform to meet those standards EXACTLY, normalization exists. BUT the LESS you can “normalize”, the better (for most applications). Unless the project itself has a different purpose, generally speaking, you want to represent your most “natural” voice in an audition. —— think about it this way: if you record your voice with such a low gain (let’s say your peaks are hitting -24 dbs) and you normalize to -1dbs peak, the audio is going to sound blown out and super loud. You’re also going to hear all the room tone and other sounds that were recorded, BOOSTED by that amount. Now imagine this - you record your audio so that the peaks are hitting -3 dbs. You normalize to -1dbs, and the AMOUNT of increase is only 2 dbs. This will not change the audio level by much, and you maintain the natural feel of your actual voice. I hope this is making sense. —— In pursuit of the best representation of your voice, you want as LESS normalization as possible. The moment you put an effect on a voice - whether that’s compression, normalization, EQ, etc - you’re essentially changing your voice, further away from what you actually sound like. The greater amount of effects applied, more it changes. Does this make sense?

For AUDITION purposes, typically you’ll see three lines here on CCC. Generally speaking, those would be three different scenes/moments (unless they’re not). It’s appropriate to use three different gain levels to record those. BUT when you’re DELIVERING your audio FOR the project, you’d want to use one gain setting for each scene, depending on the context of the story, so that the audio remains CONSISTENT throughout the scene. —— in addition, making a loud sound quieter is much easier (and natural sounding) than making a small sound louder. So your conversation with a friend across the table can be recorded with -6 dbs peaks, 12 inches from the microphone, and you yelling at the top of your lungs, raging at the waiter for getting your order wrong, 24 inches from the microphone, can also be recorded with -6 dbs peaks. The audio editor will make appropriate adjustments (quieting it down) to the conversation audio to lower it down. This, of course, is providing that you’re recording in a treated space with equipment that can handle it all.