As an engineer who works at Studio 11 here in Chicago, a studio that has specialized in hip-hop and rap for more than 20 years (having worked with a young Kanye West, Lupe Fiasco, Crucial Conflict, Lil Durk, and more), we pretty much have it down to a science.
Here are just some general rules of thumb when mixing rap vocals that you can try for yourself—coming straight from the horse’s mouth.
Make Room in the Beat
You may ask, "But what does the beat have to do with mixing rap vocals?" Everything.
If you are working with vocals on top of an instrumental, there are two main ways the beat will look: either with tracked-out stems or as a single stereo file (like a wav or mp3 file). Of course, having the stems—separated kick, 808, bass, clap, hi-hats, etc.—will allow you to have greater control over the music. But either way, there is one thing that remains the same no matter which format your working with: Make room for your vocals.
If I do have the stems, I generally like to mix the entire beat before adding the vocals. Some people like to start with drums, bass, and vocals—that's cool too. But for me, I always end up changing the mix of these instruments once I add the other elements of the beat anyway. While I’m doing this, I keep in mind that the vocal is going to need some of the space that’s occupied by the other instruments. The goal is to keep space open in the low-mids and high frequencies so that the voice has room to sit within the beat, not on top.
If all you have is a stereo instrumental track (what used to be called a "two-track"), it will be harder to make room for the vocals without affecting a lot of instruments within the beat itself that maybe you don’t want to touch. But you can still use an EQ to carve out some spots that interfere with the vocal’s body and articulation.
The body of a vocal sound, where the warmth and weight resides, will be found in the low- to upper-mids (about 300Hz to 600Hz). Some rappers will have a deep voice, some have a higher voice, so where exactly you should dip the beat’s frequencies will change. A plugin like the FabFilter Pro-Q works great for hearing and visualizing these areas.
A rapper’s bite and the clarity of syllables will come from transients and other high-frequency information. Again, this will depend on the beat and the vocalist, but generally speaking, a dip somewhere around 2.5kHz to 7kHz should be a good starting point. Use your ears—if loud hi-hats at 7k or the claps at 2.5kHz are stepping on the words, scoop around those frequencies. Sometimes, you may need to go even lower, around 1kHz to 1.5kHz, if there is a piano or other instrument that is mixed too loud.
Bring Compression to the Vocals
I like to compress hard, but fool the ear into thinking I’m not. You’ll see people on YouTube compressing very lightly on vocals, because their college professor told them that was the correct way. While over-compressing can be an issue, don’t be afraid to see the meters moving—rap vocals are very percussive and dynamic, so you need to get those peaks under control.
Listen to the vocals without any compression first. Find any words or lines that are noticeably louder than the rest. Spend some time just moving the gain levels down during those spikes.
Once you even out the vocals, bring in the compressor. There are many compressors that can do the job, but my personal go-to compressor for controlling vocal dynamics is the Waves Renaissance Compressor. It’s not super fancy, but it has a unique color, especially once you start turning up the makeup gain.
Start with a low compression ratio, about 1:5 to 2:1. Dial back the threshold. On the input meter of the compressor, you’ll generally want the needle to hover around the peaks of the transients—sometimes right on the line or sometimes a little below, if you're trying to compress a little harder (which is good for really thin/harsh voices that need less high-end and more warmth). Adjust the ratio to taste.
If a rapper has a very percussive delivery, with lots of hard consonants, lower the threshold and use a slightly higher ratio (say, 4:1) to clamp down those harsh transients. If an artist’s volume fluctuates gently up and down, a lower ratio and lower threshold will smooth out the delivery.
Next, play with the attack if the vocals are sounding a little dull after compression—try to see if you can get those initial transients to poke through, then have the compressor clamp down on the decay of the words.
Since rap is fairly quick, you generally want the compressor to release before the next word hits, so keep the release time pretty short. Otherwise, the transients of the next syllable won’t come through, resulting in a duller/softer sound. I like to measure out the time of a 32nd note in Pro Tools and enter that in for a starting point, then move to taste. As you shorten the release even further, you'll start to hear more high-end information come through, and a vocal will begin to have more teeth.
"De-Ess" to Eliminate Hisses
Don’t be afraid to de-ess fairly aggressively. Getting out those sibilant frequencies, or the hard hissing sounds that hurt our ears, is imperative to making a vocal sound smooth. Sometimes I’ll put two or even three de-essers in the chain. Most standard de-essers work here, but I use the stock Digidesign de-esser, the Waves DeEsser, or the Waves Renaissance DeEsser.
The reason you want to de-ess hard, aside from getting rid of those sibilant spikes, is because nine times out of ten, you’re going to be adding some sort of high-shelf with the EQ to give the vocals more presence. If you don’t de-ess, it’ll be harder to get the vocals upfront in the mix without hurting the ears.
Get Rid of Unwanted Frequency Buildups
There is a whole debate of whether to EQ first or compress first, but it’s honestly whichever sounds better to you. I like to do a little before and a little after, as the compressor will change the frequency response of the vocal anyway.
Start off with wider scoops to get rid of any glaring frequency buildups. I use quite a few EQs for this: Waves Renaissance EQ, Console 1 by Softube, and FabFilter Pro-Q, to name a few, but even stock plug EQs that come with your DAW will work just fine.
Generally, I start by rolling off some low-end (about 130Hz to 275Hz), but not so much that I lose the voice’s warmth and fullness. Dip some low-mid "boxiness" around 500Hz and some harsh sibilance around 3kHz to 5kHz. Then add a high shelf to taste. (Sometimes, frequencies will build up around 200Hz to 300Hz and also around 1600Hz to 2900hz, so pay attention to these areas as well.)
After creating a general EQ curve, listen for resonations in the voice, which will typically sound like a whistle or a single constant frequency poking through the speakers, which will often remain constant throughout a vocal take.
To find them, boost a band of EQ with a rather small Q (not too small, as everything will sound like it's resonating) and a moderate amount of gain. When you hit the area where it sounds like a frequency is significantly louder than the rest around it, dip that down to taste. I like to use the Waves Q series, as the Q can get pretty tight and, to my ears, the EQ is very precise—perfect for small notches.
To finish, listen to the vocal against the music and see if it needs to be boosted anywhere. If the vocal seems a bit buried, a boost around 1kHz to 1500Hz will often help it come through. While I prefer accurate, uncolored EQs for taking away frequencies, I like to have a bit more color (like with an analog unit or analog-emulating plugin) when I boost to add some harmonic excitement.
Add Body and Bite with Saturation
Having a dedicated unit or plugin for saturation works great on vocals (especially really thin-sounding ones). Depending on how you set it, saturation can add thickness and body, as well as a bit more bite in the top-end. Sometimes, saturation can smooth out the top-end, depending on how you set it. .
I'm forever in search of the perfect saturation tool, but I typically will use the Digidesign Lo-Fi, Waves Rennasiance Axx (a compressor, but with a lot of grit), Waves Kramer Tape, and my current favorite for vocals—the overdrive on the Softube Console 1.
Rap vocals don't always need saturation, as sometimes an MC will be coming in hot on the mic and the preamp will naturally overdrive a bit, but if you have a recording with a dull/weak vocal that just needs a little dirty analog vibe, look no further than saturation. You’ll be surprised how far it will take you.
Use a Small Amount of Reverb
Reverb in rap music is actually a little controversial. Historically speaking (and with some exceptions), rap never really used it until fairly recently. Rap vocals have always been an expression of rhythm, and a reverb's decay tends to step on the natural percussiveness and immediacy of an artist’s performance, negating the "in your face" intensity that rap music often aims to achieve.
Having said this, on a lot modern rap records, because of the accessibility and desire to create new types of vocal sounds, reverb is often used at some point. When I use reverb on a lead verse, it's usually a very small amount—just enough so the listener can feel it more than they can hear it. Sometimes the Wet/Dry is on just 1%.
I hear it more as a "glue" tool than an intentional effect, as a small amount of reverb does tend to make the vocal sit in the mix a little better to my ears. It also takes me out of the zone of hearing a purely dry vocal that was recorded in a dead-sounding booth, which can seem a little unnatural to our ears. Play around with different reverbs, and see which ones work the best for you.
Add Excitement with Delay
Once, when one of my favorite engineers, Andrew Scheps, was asked what his favorite reverb was, he quickly responded, "Delay." While the effect can be used in different ways (and can be tricky to master), what Andrew is saying here is that using delay can be a way to make a lead vocal sound wet or pushed back in a mix, without actually using reverb. (As I mentioned above, reverb can murk up a rapper's performance.)
Experiment with tempo-synced delays. My favorite at the moment is Waves H-Delay, but stock plugins work well too. If set tastefully, a subtle 1/8 note or even 1/16 note delay on a lead vocal can really help fill in those empty spaces, without having the delay step on the unaffected vocals.
When using delay, I like to create an auxiliary channel for a couple reasons. The first reason is so that any delay I add isn’t affecting the original lead. The second is so that I can send multiple channels to the same delay without hogging up CPU. Lastly, using an aux channel lets me process the actual echo differently. After sending a vocal to the aux channel, process it with delay and other plugins to create a more musical echo.
By adding an EQ here and rolling off low and high frequencies, the echo won’t interfere with the body or articulation of the vocals. (One reason I love H-Delay is because it has these EQ filters built right in.)
Another way to make the delay duck out of the way of the vocals is to add a compressor after it, and side-chain the lead vocal to the compressor’s input. This way, whenever the main vocal is active in the mix, the delay is compressed and lower in volume, but as soon as a vocal cuts out, the echo from the delay creeps back in.
Process Additional Vocals
In-and-outs—also known as dubs or stacks—are recorded after the lead is already laid down, when the artist adds a second vocal to certain parts of the verse. These will add emphasis to punch lines and key phrases (usually at the ends of each bar) or help to bring clarity to syllables that may not have been fully pronounced the first time around. Because a rapper will be able to take breaths in between lines, they can really make sure that they nail the parts.
The doubled lines can also help give the verse some sort of movement. Instead of a single vocal track, which can be relatively stagnant in volume (especially after compression), it adds a sort of back-and-forth dynamic. To give them their own character and space in the mix, you can compress these second vocals a little harder than the lead and pan them to add a chorus-type effect or stereo depth.
Ad-libs, or additional words thrown in between phrases of the verse, will be treated differently, depending on the rapper. A lot of trap artists I record prefer a lot of effects here, but a lot of old-school hip-hop artists don't like any effects at all. But unless the artist specifies that they want the ad-libs dry, I typically put them in some sort of telephone filter created with a bandpass EQ—that is, strip away the lows and highs and isolate the upper-mids.
One of my favorite things to do is to experiment with different effects on top of the telephone filter. Phasers, flangers, distortion, delays, and reverbs all create a unique sound to the ad-libs, and can put them even more in a specific frequency pocket. I personally like the sound of a drastic reverb with a long tail, because it will give the ad-lib its own 3-dimensional space. This creates a cool overall effect—with the dryer lead vocal up front, quieter in-and-outs on the sides, and wet ad-libs in the back. Experiment and see what kind of new effect chains you can come up with.
It's important to remember that if there was a formula that worked every time on mixing rap vocals, everyone would be able to do it. Truth is it really depends on how the vocals sound to begin with. If they sound great without anything on them, chances are you won't have to do as much. And if not, at least you now have a guide of how to fix them.
Follow these steps and you’ll have rap vocals that sound like they were done in a professional studio.