How to Compress Vocals – Part 1 of 2

By | December 11, 2014

Vocals are arguably the most important part of a song. They are so important that they deserve special attention. I’ll be discussing ways to achieve an ideal vocal track, by showing you how to compress vocals in various ways.

Now, when it comes to vocals, they usually have their own special needs. Here are things I find extremely important to take into account when approaching vocal compression:

  1. Low Midrange
  2. Dynamics of the vocal track
  3. High-end “hiss”

In part one, I’ll show you how to get those wild low mids under control in your vocals.

Tackling the low mids

When I start mixing vocals, and encounter problems with the sound of them, the low midrange is causing the problem 90% of the time. Why is this? Well, let me tell you why. The low mids are an area where most of the MUD lies. And mud is bad, mud is boomy and mud goes all over the place. It needs to be set under control. You might be wondering: “If I insert a compressor on my vocal, wouldn’t it affect ALL the frequencies in the vocal, and not just the low midrange?” And you would be absolutely right. This is why we need to use a special technique to achieve this. A few methods come to mind:

  1. A single-band compressor with a side-chain filter

In this method you would simply isolate the frequency range you want to compress with the compressor’s side-chain filter, in this case the low midrange, and the compressor would only affect those frequencies. Very easy and useful. NOTE: you need a compressor with customizable high and lowpass (or bandpass) filters to isolate your frequency area, such as FabFilter Pro-C

Read my review of FabFilter Pro-C here.

The compressor is processing only the frequencies set by the side-chain filter (200-600Hz in this example). NOTE: In Fabfilter Pro-C, you need to click the "Expert" button to open the sidechain feature.

The compressor is processing only the frequencies set by the side-chain filter (200-600Hz in this example). NOTE: In Fabfilter Pro-C, you need to click on the “Expert” button to open the sidechain feature.

  1. A multiband compressor

With multiband compression you will achieve essentially the same end result as in example 1. The only difference is, a multiband compressor has more than one band of filters. Personally, I haven’t used a multiband compressor because I have always accomplished the tasks it does in other ways, but I know lots of producers and engineers use them and are no doubt an excellent tool, especially with a vocal.

Waves C4 Multiband Compressor

Waves C4 Multiband Compressor. You can spot four different filters by their crossover points.

  1. An external EQ feeding into a compressor’s side-chain input

This method is a bit more advanced. It is also similar to the method accomplished in example 1, but is more time consuming, yet more precise. The precision comes from using an EQ and very carefully adjusting the filters that are being sent from it into the compressor. With Fabfilter Pro-Q 2
 you can use filter slopes up to 96 dB/octave, which means the problematic area in the vocal can be pinned down very precisely.

To make it work, duplicate your vocal track and make sure no audio is sent from it to the main outputs. In other words, the duplicate is now playing as a “ghost” track. On the original vocal track, you should have a compressor with external side-chain capabilities, such as Fabfilter Pro-C. On the duplicate, you should have an EQ. Now, here’s the key: create a send on the duplicate track, so it’s audio will be routed into a bus. Then, open the compressor on the original vocal track, enable it’s external side-chain, and have the side-chain input routed to the same bus the duplicate was sent to. Now the compressor is only listening to the “ghost” signal on the duplicate track, which in fact is the same signal as the original, and that’s exactly the point. Lower the compressor’s threshold significantly so it starts working. Create a very narrow band-pass filter in the EQ and sweep it around to hear how the compressor affects only the range of frequencies in the filter.

1. Duplicate the track and insert a compressor with external side-chain abilities on the original, and an EQ on the duplicate track. Make sure no audio is going to the main outputs from the duplicate track (in Pro Tools done from the I/O menu).

1. Duplicate the track and insert a compressor with external side-chain abilities on the original, and an EQ on the duplicate. Make sure no audio is going to the main outputs from the duplicate track (in Pro Tools this is done from the I/O menu).

2. Send the duplicate to a bus. In Pro Tools, I have sent it to bus 128 for example's sake. Open the compressor on the original track, enable its external side-chain and choose the same bus you sent it to as the sidechain input.

2. Send the duplicate to a bus. In Pro Tools, I have sent it to bus 128 for example’s sake. Open the compressor on the original track, enable its external side chain and choose the same bus you sent it to as the side chain input. The method should be working by now.

Now just apply filters in your EQ and see how the compressor reacts to it. In this example, I have created a narrow 48 dB/oct bandpass filter targeting the low-midrange.

3. Now just apply filters in your EQ on the duplicate, and see how the compressor reacts to it. In this example, I have created a narrow 48 dB/oct bandpass filter targeting the low midrange.

The principle for this method is the same for all DAWs. If you don’t use Pro Tools, you should look into how to apply side chaining in your particular DAW. I hope this article has taken you to the next step towards achieving a great vocal sound. If you have any questions, please leave a comment below. Happy mixing!

-JP

Continued in Part 2.

Leave a Reply

Your email address will not be published. Required fields are marked *


*