Okay, here’s the deal – I’m an editor and engineer who also happens to do voice over and on camera acting occasionally so I’m especially suited to talk about this topic.
Voice over is recorded in a sound booth or sometimes a sound stage (if you need room tone applied to the track already) and it does not include any extraneous sounds. That’s the point really. You get clean audio from the start and you can apply it to anything you want or in any situation you want.
Now keep in mind in order to have a clean sounding voice over while your mileage may vary you want to have a noise floor below -60 decibels. The lower, the better for voice over. What’s a noise floor, it’s similar to room tone in that it’s the sound of the booth or area where you’re doing your voice over without any noise.
Dialog is used as part of a scene that was recorded on location or a sound stage. There’s usually more noise present than just the voice and that includes room tone. You’re just going to have more to deal with because it is meant to sound true to life or to at least fit the scene in the scheme of the story.
So it’s usually easier to just deal with voice over but that’s not how movies work. Also keep in mind that voice over, depending on the actor, can sound omnipresent. That means, the sound will sound like it comes from everywhere because it was recorded in front of a microphone and not on set. Smart voice actors know how to work the mic but not everyone does.
So how do you treat the audio in a documentary vs a film that deals with dialog?
In a documentary the voice over is supposed to omnipresent. It doesn’t interact with reality. You’re going to treat every track the same. It was recorded in a sound booth so in some ways it’s easier but what you want is a consistent sound all 30, 60 or 120 minutes.
Dialog based films are a bit trickier. You’ll have people far away, you’ll have people behind barriers, the dialog in a film was created or is motivated by activity and as such you need it to sound live within the circumstances that are taking place in the now. If you’ve had an actor do ADR and you’re getting the lines, the lines need to be cut back into the scene with the room tone of the room the actor was in during that scene laid under it and the actor may need to have a low pass filter laid over their voice to replicate them being in a closet while talking to someone outside of the closet. You’ll also be dropping in SFX in some cases behind this and then music, of course and yes, it’s a bit more complex. You’re not going to have everything sound the exact same outside of a scene either.
So there are differences between the two but ultimately no matter what you’re working on you may encounter both and need to always listen to everything first.