Ducking is an effect often used by DJs and in all kinds of broadcasting. It’s often achieved by the use of side-chain compression. Side-chaining is when you use the dynamic levels of one source to control the compression of your signal. It’s best described in example, for instance, it’s common to use this for the bass and kick drum. Compressing (thus lowering the levels) the bass when the kick drum enters to have them both fit together.
There are other ways to use side-chaining, but in this post we’re going to explore the above principle, usually called “ducking”. We won’t talk about kick drums and basses however. Ducking can be used to lower the music whenever your voice is present on another track. Recognize that? Used in countless radio shows.
Setting it up
First of all, if you’re on Mac, try GarageBand. it actually has a built-in ducking engine. Just choose “ducking” from the menus and you can set which track you want to control the levels of the others. It’s that simple. If you use GarageBand you don’t have to read the rest of this post.
In other applications you might have to set this up manually, and depending on which application you’re using it might be achieved differently. You will have to consult the manuals for this. I will walk you through the basics though.
Assume that we have two tracks, one voice-over track (let’s call it track 1) and one music track (track 2). Insert a compressor on track 2 (the music track). Now, almost every audio application will give you a separate GUI for the compressor (usually in a pop-up window). Set it to do some heavy compression. Now you’ll want to look for something called “side-chain” or “side chain” (not all compressors support this). Connect the side-chain to track 1 (the voice-over track). Depending on the application this is done in certain ways. In some (such as Pro Tools) you will have to route via busses. If you know your application you can probably figure out how to do it (if not RTFM ;)).
Now you can adjust the threshold on the compressor to decide the amount of dynamics need for the compressor to kick in. If you have connected it right, the compressor will start to work every time the voice track enters. Nice huh?
Why don’t just lower the volume manually?
Good question. I know a lot of people doing this still, and I would actually prefer this myself if I were to do a podcast or a radio show. To me, it’s more flexible, and I feel I’m still in charge of what happens. Lowering a volume fader is quick and easy. For music it’s a whole other thing though (for instance the above mentioned example with the kick and the bass). Bottom line, do whatever you like, try ducking and see if you like it.