Unless you’re an instrumental progressive metal band, vocals are going to be the most crucial part of your mix. Of course, every genre has its own “rules” on whether the vocals should be upfront or back in the mix.
If you’re aiming for a dirty-lo-fi sound maybe you won’t be interested in most of these points, but it’s still good to know how to treat your vocal track. Your mix could sound absolutely killer, but if something’s wrong with the vocals, that’s going to immediately stand out.
Here are a few tips every engineer should keep in mind when it comes to mixing vocals.
Many engineers could probably argue that point 1 and point 2 should be inverted. It simply depends on what your usual workflow is, what you’re looking for and what’s required for the specific genre you’re mixing. Before throwing every possible compressor on your vocals, listen to the dynamics of the track and how these fit in the mix.
Some people prefer to record vocals with some compression and eq already on it: if you have enough experience to avoid “messing things up” in the early stages and you feel confident with it, that’s fine.
It’s actually better for the singer itself, because if his monitor mix is already sounding “tight and bright” he’s going to deliver a better performance. It’s always good to compress your vocals through different compressors rather than just using a single one.
Start off with a “gentle” ratio on your first compressor, usually 2:1 or 4:1. Set a rather slow attack and a medium release. Start pushing the threshold until you start hearing the compressor working.
The second compressor can be inserted after an eq stage, so you can filter out some of the harsh frequencies that compression tends to enhance (5-7 kHz). This second compressor can be used to “squash” the parts in which the singer occasionally shouts and to give a general tone to the whole track: set a medium attack and a fast release, a high ratio (even 7:1) but keep quite e high threshold.
Some of the great vocal outboard (or plugin emulated) compressors used for vocals are the UA 1176 or the Empirical Labs Distressor for more aggressive sounds; the UA LA-2A and similar tools can be used for more controlled and “tube-sounding” vocals.
There are two types of eq that can and are usually applied on vocals.
The first one is usually a digital eq used to correct and notch the annoying frequencies that keep your vocals “muddy” and lost in the mix.
It’s a common thing to apply a high pass filter to frequencies below 100 Hz (but it’s not a rule!). To make your vocals more bright and less muddy, try cutting a range of frequencies between 250 Hz and 800 Hz. If your vocals are sounding too honky and nasal, cut a few dB around 600 Hz and 1.1k Hz.
The second type of eq is used for enhancing certain frequencies that really define the sound of the singer and make him fit in the mix. It’s usually an analog outboard eq or a plugin emulation, depending on the amount of harmonic distortion and “color” you’re looking for. Pultec (and its plugin emulation) is a terrific vintage sounding eq for this.
If your vocal need more air you’ll have to boost a few db around 10 kHz and 20 kHz.
The presence and intelligibility hides around 1 kHz and 3 kHz, and the real presence is usually between 3 kHz and 6 kHz. Beware of the sibilance you might excessively enhance through the previous compression and eq! (that’s the purpose of point 3)
Even if you’ve recorded the singer with a thick pop-filter in front of the microphone and you’ve correctly applied all of the compression and equalization stages, the letters “s”, “t” and “z” are usually going to stick out right in your face.
There’s only one solution: de-esser is a compressor that attenuates a single annoying frequency when its volume goes above a certain threshold. A useful plugin that does this job amazingly is the Waves DeEsser: you can choose the exact annoying frequency and listen to it (by clicking on the “Monitor S Chain” button on the bottom left corner) and then set the threshold of the compressor.
Make sure this tool is set correctly so you can filter certain frequencies without creating a “pumping effect” on the track. You can apply as many de-esser as you want and need.
Hypothetically speaking, the only limit on the number and type of effects you can apply on the vocals is your imagination. Joking aside, in this step we’re going to consider some of the basic vocal effects, such as reverbs and delays.
Be aware, it’s very easy to mess things up with effects. As an engineer, make sure you’re focus is the overall sound and emotion of the song rather than your personal “ego”. It’s music, not an over-production show off.
Reverbs are useful to create a 3D space for the vocals in the mix. “Room” reverbs have shorter decay times and can be used during the verses or for more tight-sounding mixes. “Hall” reverbs have a much longer decay and are useful to place the vocals more in the mix. “Plate” reverbs carries a lot of early reflections that have a shorter decay time and a thicker sound compared to the previous ones.
Lexicon has made some iconic hardwares that emulate different spaces; if can’t afford one, you can always rely on some cheaper but still amazing Waves plugins, for example the TrueVerb or the Reinassance Reverb.
If reverbs are usually used to give more space to the vocals, delays are used to bring the vocals forwards and backwards in the mix. If you set short delay times you can achieve a good doubling effect that adds thickness to the vocals.
If you increase the delay time and the feedback you can add a long smooth tail to every word, but don’t exaggerate or you’ll end up with a muddy and confused vocal line. If you’re using it as a send/return effect make sure it’s 100% wet, otherwise it’s usually better to keep a maximum dry/wet ratio around 30%.
One of the most iconic tape delays is the Roland RE-201, commonly known as the “Space Echo”, which adds some very interesting tape saturation when it’s needed. A good plugin emulation is certainly the Waves H-Delay.
What’s the point of automating the volume of a super compressed vocal? Well, because it still has to blend with the dynamics of the whole song, so it has to be occasionally turned up and down to match the other instrument’s volume.
Of course, with modern DAWs you can automate every possible parameter of every possible plugin, so be smart: try turning the reverb down during the verses and maybe increase it during the choruses. Or bring the delay up just in some key points during the singer’s last word so you can achieve a nice musical tail to what he’s saying.
Pan automations on the vocals can give the listener quite a shock, so be careful with them. It’s usually good to keep the backing vocals (not the doubling vocals) slightly on the sides of the mix, but not hard panned.
Well, that's basically it. These tips aren't always enough to get a good vocal track in your mix, but it's good to know how to process it correctly.
I’m a young musician and sound engineer based in Italy.
I’m also the Artists & Professionals Assistant at Bantamu and I’m here to get the best music out there, whether it’s playing, engineering or simply writing about it.
Connect with the right Recording, Mixing, Mastering Engineer for your music