Sign in to follow this  
Janice Yap

How to compress vocals on LogicPro and/or GarageBand?

Rate this topic

Recommended Posts

Hi Everyone! 

When recording a powerful voice on LogicPro and/or GarageBand, what sort of compressors or settings in the compressor should we tweak so that the output of the voice is limited and will not explode into the red volume zone, without losing the volume on the softer singing parts?

I understand logically that singing further away from the mic when singing loudly often does the job, but I've seen producers slap on a compressor of some sort that prevents the voice from "exploding" (and sounding distorted) without the singer having to pull themselves away from the mic - anyone know how to do this? Please share smile.png 

Share this post


Link to post
Share on other sites

Scroll through the presets a bit, to see if anything sounds good. Otherwise, it can get very complicated. I've been a mixing and mastering engineer for more years than I like to admit, and compression on vocals is still a difficult concept to describe.

The main/basic idea: You want enough of a ratio to even out the vocals without it sounding odd. That ratio can be relatively high for vocals, this is the main part of compresssion. I often end up between an 8:1 and 9.5:1, and handle extreme volume changes with an automated volume control. Then set the threshhold low enough to capture the softest sounds but not background noise. I usually set threshold to -30 when recording, but much higher in a mix down). At this point, output gain may need to be lowered a bit to keep it from clipping.

Knee should be round enough to not sound robotic, crank it if you're not sure. Attack will likely be low for vocals so that you'll still catch sibilance, perhaps even as low as 2. Release depends on how fast or drawn out you're singing, but 30 to 40 is a safe bet.

A limiter set to -1 threshold will help even more, but only after you have the compressor under a pretty well balanced control.

Share this post


Link to post
Share on other sites

Great question Janice.

Using voice proximity is not the same thing, not even close. If your trying to even out the vocal levels in your recording, pulling away from the mic is not how you do it. All that will do is thin out your voice color and greatly compromise any of the merits that your microphone is trying to give you.

Here is what I do, but understand, I am NOT a compression expert.

Any DAW has compression plug-ins available. Most are probably going to be fine for you. I happen to use the focusrite plugin that comes with their suite of plugins when you purchase the Scarlett or Clarrett interfaces. I also use the CLA compression setting on the plugin produced by www.waves.com .

VERY basically, I just set the ratio on the Focusrite plugin to about -5.0. on threshold. I also use the medium compression template. Thats it... and it seems to kinda do ok. 

Here is the video that demonstrates these free plugins from the Scarlett.

See 5:15 to get to the compressor tutorial:

 

Share this post


Link to post
Share on other sites

There are several compressor templates on Logic. You should record clean and then apply the templates or plugins, as Robert suggested, and listen to the result. If you are not sure, you could also create different vocal tracks from copying and pasting the original vocal track, and trying different templates. After you choose a template you could also tweek it to your liking, changing the settings, as Draven suggested. Cheers

Share this post


Link to post
Share on other sites

Heres what would be done in most cases if you went to a professional studio to have your vocals mixed, or soke close variation. First thing after editing out the noise on the front and back ends of every take, would be to do a fader automation. This is the slow, tedious task of turning up the lower spots and turning down the louder ones to make everything more consistent. Plugs like Waves Vocal Rider make much quicker work of this, and MVP can work well coming after it. Now that your levels are smoother it wont cause the compressor to attack transients and s’s and such so hard. Now youre likely to be run into an LA-2A (either hardware or a well done emulation plug) which is a very easy compressor to work with. Just turn it up until youre shaving a couple decibles off... listen... dont work by sight. Close your eyes and listen.  If shaving 12 db sounds better than shaving 3, then by all means. But thats unlikely to be the case, go easy in most cases. Next theyll often run you into a faster 1176 style compressor to catch the last of the transients and to add some more color. 1176 is a much trickier and powerful compressor than the LA—2A and you really have to pay attention to what its doing. Youre really not wanting more than 1-3 db coming off at this stage,  just trying to catch the occasional peaks that might screw with your levels later on. Sometimes you can accomplish exactly what you want with some good accurate fader rides/automation, and you wont even need to compress. The LA-2A adds a nice euphonic color though so always try one even if you dont need it and its not compressing any. Same with a pultec EQ afterwards. The color it imparts is often nice even if its used in bypass, as the signal will still pass through the tube (or the tube emulation in plug ins) on the way through. 

Oh, try high passing the low end that you dont need out of the vocals and also de-essing prior to hitting the compressors as well. Itll help get a smoother result. 

Never be afraid to Experiment. You cant break anything. Good luck, and enjoy. 

Share this post


Link to post
Share on other sites

Another neat compressor trick. Either copy and paste the vocals to a new track (or send a feed of the vocals to an aux) and throw an 1176 style compressor on it. Push all the buttons in and just smash the needle. Compress it to death until it starts to get pretty hairy. Now turn the track all the way down and start playback. Slowly turn up the heavily compressed track until you start to hear it thicken up the main/original vocal line and it becomes noticable, and then back it off just a hair. Add to taste.

This is an old engineers trick that works great when it works. Also great for drum submixes as well. If you have a compressor with a mix knob on it, you can do it by just using a small amount into the mix directly on the vocal track without needing to duplicate it or whatever. 

YMMV. 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this