Skip to main content

Sidechain Compression

Lesson 22 from: Music Production in Logic Pro X: Vocal Mixing Essentials

Tomas George

new-class music & audio

buy this class


Sale Ends Soon!

starting under


Unlock this classplus 2200+ more >

Lesson Info

22. Sidechain Compression

<b>In this lesson, you will learn how and why we might use Sidechain Compression in Logic Pro.</b>


Class Trailer

Introduction and Welcome to this Class


Project Organization


Faders and Panning


Flex Pitch - Vocals


Flex Time - Vocals


Editing Studio Drums


Song Mix Deconstruct - Mixing Drum Kit Designer


Mixing Files


Lesson Info

Sidechain Compression

Hi. In this video, I'm gonna show you how you can use side chain compression when mixing vocals. In not a pro 10 to side chain means to take another signal as input. But in the context of compression is to take another signals input and use that as the detector for the compression, but the compression is still applying to the sound itself. So for example, in EDM tracks, it could be quite popular to compress like a pad sound but use a kick drum as side chain to trigger the compression that's happening on the pad sound. So it creates this sort of pump effect on the pad sound that has been triggered by the compressor. That's not what we're gonna do here, but we could do that. But the way that we're gonna use side chain today is actually to use a filtered version of the vocal as the trigger for the compressor, but we're still compressing the full band of the vocal. So we're not compressing only a certain frequency band. We're just using a certain frequency band as a trigger for the compres...

sion on the entire focal frequency range. So I'm just gonna go to the side chain menu here and I'll come back to the detection menu. But what I'm gonna focus on first is the filter menu. OK? So if I select listen and then I choose from mode here, I'm gonna choose a low pass filter. What this refers to is the EQ or the filter of the side chain version of this vocal. So let's have a listen to that where you f So right now we are auditioning the filter that we're using as side chain. I selected low pass. I dialed the frequency down until I could only hear the kind of not muddiness but the warmth, the bottom end of his vocal. The reason that I did that is I want the detector of the compressor to ignore the bottom end of his voice because I don't want that triggering the needle. I actually want the higher register content of his voice to trigger the needle instead because some of the highest amplitude values of his voice are going to be some of the bottom end. So what I'm doing now is I've isolated the frequencies that I kind of want the compressor to ignore and I'm gonna switch it now to high pass. So let's have a listen to that. And through you can see now that the needle is behaving very differently. It's not reacting to the lower frequency content of his voice. It's now just reacting to the top end stuff. So the compression should be a bit smoother and a bit more of what I'm expecting now, I'm gonna turn off the filter and it's important to note that the filter is still on, but it's on in the side chain. I'm just not listening to the filter now. OK. So the compression, even though I've turned off the filter here, it's still only reacting to that high pass filtered version of the vocal even though we're hearing now a full band version through my. Do I feel like I'm great where this differs to, for example, a multiband compressor or a dynamic EQ is that in either of those plugins, you are compressing the band. So not only are you detecting a, a particular band like the lows or the mids or the highs, but you're also compressing only that band. In this example, I'm detecting a band but I'm affecting the entire frequency spectrum of the vocal. Now let's move on to the other features that are available to us in the side chain. So right now the detection is on Macs, what that means is between the left channel and the right channel of this compressor, it's reacting to the highest value of the two. So if the left channel is three DB higher than the right channel at a given moment, it's reacting to the highest value which is the, the left channel. However, if I switch it to some, it's now reacting to the sum of those two. So if the left channel was six db louder than the right channel at a given moment, it would react to three db louder than the right, it would react to the sum of the two. We also have options for peak detection or R MS detection. Peak detection is going to be a much faster response because it's tracking the peak values of the waveform. Whereas R MS is generally going to be a slower response because it's reacting to a moving average of those amplitude values for this. I'm going to actually switch over to R MS and keep it on some now because I want it to respond in the way that I think is audible when the vocal comes in. I want it to respond more to the body of the note and not the peak transience of the note through my, the moments. I feel like I'm flying great. So as you can hear, we have a really smooth compression happening now, even though I've got the attack and release times all the way down because it's responding only to a certain band of the vocal frequencies. But it's also responding to the R MS or the moving average of the signal. It's just a much slower detection, it's much slower response and thus a much smoother compression, which can really, really help when you're just trying to balance or smooth out the overall loudness of the signal. Thanks for watching and I'll see you in the next video.

Class Materials

Bonus Materials

Ratings and Reviews

Student Work