Hi there,

I'm currently working on a project with a lot of voice over.
Naturally, there is much stuff recorded that is a nuisance, like swallowing, breathing, etc.
Right now, I edit every recording manually but it's tiresome to edit hours of voice over manually so I'm interested in gating best practices.
Although I'm experienced in recording I'm not experienced in gating. I always said that if you're recording right and leveling properly, there is no need to gate. What a fool I was
I fail to achieve satisfying results right now with ableton live's built in gate. It's either guillotine or noise.
What am I doing wrong?

 

Unfortunately, manual editing is the way to go (cut, cut, delete). At least for me.
I never really got good results with gates, not even with those who have frequency detection key-input (like the free Floorfish). You can try it though:
http://www.digitalfishphones.com/main.php?item=2&subItem=5

Edit: I usually cut, cut and move breaths to another channel and turn that level down (individually or just pulling that fader down). Taking away breaths can sound really strange. Deleting smacks, throat-clearing, noises, etc is totally fine by me.

 

Last edited: Aug 21, 2015

I feared there was no way around editing. I'll try floorfish though.
Thanks bax!

 

If the loudest constant peaks are from the voice it self, and when the gate only opens when it's triggered above the treshold, then you would end up with hearing the voice over only. So you could try to use the voice as a sidechain input for the gate. Results depend on the recording, I guess.

 

@ fraifikmushi

BIAS Master Perfection Suite offers good tools.
With it I recently restored an audio document from WW2, with surprising good results.

Code:

http://www.bias-inc.com/downloads/videos/masterPerfectionSuite/movie.php

But it always depends on the audio material. In many cases you cannot avoid manual work!
Using the right tools wisely you can skip some fuzzy work though.

 

not that familiar with gating, I do usually use Isotope nectar breath control which comes as a separate plug-in in that bundle, combined with manual cutting

 

Breathe sound is a natural part of speaking. If they get in your way and are too loud for your taste any of the above advice will work. It is tedious and repetitive stuff. One time I did a quick fade out after each pause (never lower than the noise floor because it sounds unnatural to do so) but it was only about 3 1/2 minutes of a talk. Fortitude and your own relaxing exhalation are the only medicine. You could get someone to help and bare some of the load and make it go faster, that would be a bonus.

 

would be informative to hear some before and after tracks, to get an idea how the edits sound

 

Agreed gating sounds kind of weird.
You could try downward expansion? Disclaimer: I've not actually tried this

In my experience these sort of things produce weird results at first. You lose the end of an 'S' on a word to the gate for example. Then by the time you get done fiddling with the automation, you may as well have just edited it.

 

I assume you are running parallel compression, so whenever the voice comes in the apparent level in the background goes down in volume behind it? Between that and a sensitive gate with the threshold set at the right level, you should be able to get a fairly good voice over set up. Then, of course, maybe you need to go back to the source. Good speech skills, the right mic with the right placement and wind screen/pop filter, multi -band compressor. Just like any other track, you can only redeem a turd so much.

 

Strip silence?