Mixing rap and trap vocals with Auto-Tune: the chain I use
In rap the vocal is the rhythm; in trap, Auto-Tune is an instrument. The mix has to respect both: constant presence, character and no artifacts you did not choose.
- In trap, Auto-Tune is an artistic choice, often already printed on the track. The mix makes it consistent, it does not hide it.
- The chain: clean-up, tuning if it is not printed, firmer compression than in pop, saturation, de-esser, EQ, then tempo-synced delays and short reverbs.
- Ad-libs and doubles play a different role from the lead: lower, filtered, with their own effects.
A rap vocal must always come through: every syllable is rhythm, and if a word gets lost under the 808, the flow gets lost too. A trap vocal adds a layer: automatic tuning is not a hidden correction but the colour of the track. Mixing it well means holding together two needs that look opposite, constant presence and an effect that lives on controlled instability.
Effect or correction?
This is the first question to ask the artist, because it changes everything else.
As an effect, Auto-Tune works with the correction speed (Retune Speed) at or near zero: notes jump from one pitch to the next with no transition, and that is the “robotic” sound behind so much trap. The decisive thing here is the key. If the scale set in the plugin is not the beat’s, it will drag the wrong notes exactly where they should not be. Before touching any parameter, check the key of the beat: the LUFS meter estimates it from the file, and the Camelot wheel shows you the compatible ones.
As a correction, the speed is slower and features such as Humanize let long notes breathe. When I need total transparency I work note by note in a graphical editor: it is slower, but nobody notices.
Many artists record with Auto-Tune in their headphones and deliver the processed track. That is fine, but always ask for the dry version too: if the correction produces a wrong jump somewhere, I can fix just that spot.
The chain, in the order I use it
- Clean-up. Noises, clicks, plosives. I do not remove every breath: in rap they are part of the rhythm, so I turn them down. High-pass between 80 and 120 Hz: higher than usual, because underneath there is an 808 that needs that space.
- Tuning, if it is not already printed, in the key of the beat.
- Two-stage compression, firmer than in pop. A first fast compressor taking 5–8 dB off the peaks, a second slower one working 2–3 dB constantly, often with a parallel bus underneath. A rap vocal has to sit at the same level over the beat all the time. You will find the starting values in the vocal compression tutorial.
- Light saturation. Harmonics that make the vocal come through even on a phone speaker, where the low end of the beat disappears and the vocal has to stand on its own.
- De-esser. Heavily tuned vocals with a lot of presence easily get harsh on “s” and “z” sounds. A firm de-esser is better than an EQ that dulls the whole vocal.
- EQ. Clean-up between 200 and 400 Hz if the vocal is muddy, presence between 2 and 5 kHz with care around the most aggressive area near 3–4 kHz, soft air at the top. On cheap earbuds harsh highs sound twice as harsh.
- Space. Tempo-synced delays rather than long reverbs: a big reverb smears the rhythm, a filtered delay at 1/8 or 1/4 underlines it. The delay and reverb calculator gives you the right milliseconds for the track’s BPM. “Delay throws”, automated on the last word of a phrase, are the simplest way to add movement without clutter.
In rap every syllable is rhythm. A long reverb blurs it; a tempo-synced delay repeats it in time.
Doubles, ad-libs and choruses
Doubles reinforce the lead vocal, they do not replace it. I keep them 6 to 10 dB below the lead, time-aligned, with a firmer de-esser and less top end: two slightly offset “s” sounds come across as a whistle. Depending on the track I spread them to the sides or keep them tight behind the vocal.
Ad-libs are a different character. Lower, often filtered (band-pass, telephone effect, distortion), panned, compressed even harder, with their own effects. I automate them so they do not cover the lead’s words: they should answer, not talk over it.
In the choruses the vocals widen: stacks on the sides, lead in the centre, a little more space than in the verses. The contrast between a dry verse and a wider chorus makes the chorus feel bigger without turning it up.
The vocal and the 808
Vocal and 808 share fewer frequencies than you might think: the 808’s fundamental sits below 100 Hz, the vocal above. The trouble comes from the 808’s harmonics, especially when it is distorted: they fill the range between 200 Hz and 1 kHz, exactly where the vocal has its body. The cleanest solution is a dynamic EQ on the beat, keyed from the vocal, that opens space only when the vocal is there. If you are also working on tuning the low end, start from the kick and 808 tutorial.
The most common mistakes
- The wrong scale in Auto-Tune: “almost right” notes that sound out of tune in a strange way.
- A vocal whose level changes from verse to verse because it was recorded on different days: it needs evening out before compression, not after.
- Too much reverb, which kills the flow.
- Too much air and presence: the vocal sounds clear in the studio and cuts on earbuds.
- Ad-libs at the same level as the lead vocal.
Got a vocal that needs work before the mix? Vocal tuning, timing and de-essing are included in mix and master, from $110.
Start a project