How to Mix Vocals, Step by Step: The Vocal Chain Order and Settings

To mix vocals, gain stage the vocal so it averages around −18 dBFS, tune it, high-pass below 80 to 120 Hz, de-ess the 5 to 10 kHz range, compress 3 to 6 dB at 3:1 to 4:1, cut mud and add presence with EQ, add light saturation, put delay and reverb on sends, then ride the level through the song.

That is the whole process in one sentence. The rest of this guide takes each step in order, explains why it sits where it does in the vocal chain, and gives you a concrete starting setting. Treat every number as a place to start listening, not a rule: every voice, microphone and room is different.

Vocal chain order for mixing vocals

The order matters because every plugin reacts to what the one before it did. A compressor in front of the high-pass filter reacts to rumble you're about to remove; reverb in front of pitch correction gives the tuner several pitches to follow. This order avoids both problems:

How to mix vocals, step by step

1. Gain stage the vocal

Before you add a plugin, use clip gain (region gain) so the vocal averages around −18 dBFS with peaks between −10 and −6 dBFS. Many plugins, especially analog-modelled compressors and saturators, are designed around that level, and it leaves plenty of headroom on the mix bus.

Clip gain is also the first and cleanest leveling tool. If one verse was sung 5 dB quieter than the chorus, or a single word is far louder than its neighbours, fix it here by adjusting that region directly. Every plugin after this point then works on a more even signal, and the compressor can shape the sound instead of rescuing the level.

2. Edit and clean the take

• Comp the best take. Pick the best line from each pass. The mix can't fix a flat performance.
• Turn breaths down, don't delete them. Lower loud breaths by 6 to 10 dB. Removing them completely makes the vocal sound cut together and unnatural.
• Cut noise between phrases. Silence the gaps where there's only headphone bleed, chair creaks or room noise.
• Fade every edit. A 5 to 10 ms fade at each region boundary stops clicks.
• Check timing on doubles. Stacked vocals should start and end consonants together, especially T, K and S sounds.

3. Tune the vocal

Pitch correction goes early because it tracks pitch most reliably on a clean, dry signal. Set the song's key and scale first (or let the plugin detect it), then choose a retune speed. The retune speed is the time it takes to pull a note to pitch, and it decides whether the correction is invisible or obvious.

Here's Tune on its Natural Pop preset: Retune at 70 ms, Humanize at 45% and full correction. Humanize relaxes the correction on held notes, so vibrato survives while short, sour notes still land in key. Try the hard-tune presets afterwards to hear how much the retune speed alone changes the character.

4. High-pass and cut the mud

Put a high-pass filter at 80 Hz with a 12 to 18 dB per octave slope and raise it while the track plays until the voice starts to thin out, then back off a little. Deep male voices usually sit around 80 to 100 Hz, lighter voices around 100 to 150 Hz, and backing vocals can go to 150 to 250 Hz.

Then cut what clouds the voice: mud at 200 to 500 Hz and boxiness at 400 to 900 Hz, 2 to 4 dB each, only where you actually hear it. Find the exact spot by boosting a narrow band 6 to 8 dB, sweeping until the problem jumps out, and flipping the boost into a cut. The full frequency chart is in our guide on how to EQ vocals.

5. De-ess the sibilance

Sibilance (harsh S, T and SH sounds) usually sits between 5 and 10 kHz: lower in deeper male voices (around 4 to 7 kHz) and higher in bright female voices (around 6 to 10 kHz). A de-esser turns that band down only while an S is happening. Aim for 3 to 6 dB of reduction on the loudest S sounds. If the singer starts to sound like they have a lisp, you've gone too far.

6. Compress the vocal

A vocal can vary by 10 dB or more between a whispered line and a belted one. Compression narrows that gap so quiet words stay audible and loud ones don't jump out. These settings work for most lead vocals:

For dense pop and rap vocals, use two compressors in a row: a fast one (4:1, fast attack) that catches only the peaks with 2 to 3 dB of reduction, then a slower, smoother one (2:1 to 3:1) that takes another 2 to 3 dB. Each does a little, so neither sounds obvious. For an even more forward rap vocal, add parallel compression: a send to a heavily compressed copy (8:1 or more, 10 dB or more of reduction) blended 20 to 30% under the lead. Our list of the best compressors for vocals has options for each role.

7. EQ for presence and air

With the level steady, shape the tone. A broad boost of 1 to 3 dB around 2 to 5 kHz brings consonants forward so the vocal cuts through, and a high shelf of 2 to 4 dB from 10 to 12 kHz adds breath and sheen. Check the S sounds after every top-end boost.

Often the better move is on the instruments rather than the vocal. Dipping 1 to 3 dB around 2 to 4 kHz on guitars, synths or pads makes room for the voice without making it brighter. A dynamic EQ on the instrument bus, keyed from the vocal, does the same dip only while the singer is singing.

8. Add saturation

Saturation adds harmonics, which makes a vocal sound thicker and more forward and helps it read on phone speakers that can't reproduce its fundamentals. Raise the drive until you clearly hear distortion, back it off until you don't, then blend 10 to 30% wet. Tape and tube styles are the usual choice for vocals; see what tape saturation does for the difference.

Steps 4 to 8 are exactly the chain inside Velvet: Low Cut, De-ess, Compress, Presence, Air and Warmth, then Double and Space, each on one knob. The Radio Ready preset sets a 110 Hz low cut, De-ess at 45%, Compress at 65%, +3 dB of presence, +7 dB of air, a little warmth and Polish at 80%. Try it on the sample vocal or your own microphone, then bypass it to hear how much the chain changes.

9. Add space with delay and reverb

Put delay and reverb on send (aux) tracks rather than on the vocal itself. You can then EQ the effects separately, share them with backing vocals, and change the amount without touching the dry sound.

• Delay time: sync to the song. A quarter note in milliseconds is 60,000 ÷ BPM, so 500 ms at 120 BPM and 667 ms at 90 BPM; an eighth note is half that. A slapback of 80 to 120 ms suits rap and rock.
• Delay feedback: 15 to 30% for a few clean repeats.
• Reverb: a plate with 1.2 to 2 s of decay for pop and R&B, a short room of 0.4 to 0.8 s for rap. Add 20 to 40 ms of pre-delay so the words stay clear before the tail starts.
• EQ the returns: high-pass the effect returns at 200 to 300 Hz and low-pass them at 6 to 8 kHz. The space stays, the mud and hiss don't.
• Duck the effects: sidechain the delay return to the dry vocal so the repeats only bloom in the gaps. Our sidechain guide shows the routing.

If you're using both, the order on a single send matters; see delay or reverb first. For the classic vocal reverb sound, read what plate reverb is.

10. Ride the level and automate

Even after compression, some words get lost under a snare hit or a synth swell. Engineers ride the vocal fader through the whole song, usually by 1 to 3 dB at a time, so every word lands at the same perceived level. Typical moves: lift the last word of a phrase that trails off, raise the vocal 1 dB in the chorus, and push a single word into a delay send (a delay throw) at the end of a line.

A vocal rider does the first half of that job automatically. Placed before the compressor, it raises quiet phrases and lowers loud ones toward a target, slowly enough that it doesn't sound compressed, so the compressor has far less work to do. Rider's Pop Lead Smooth preset steers toward −16 with a 6 dB range, a moderate 45% speed and breath control at 40%, so breaths between phrases stay down instead of being lifted with the quiet words.

Keep manual automation for the musical decisions: a vocal that steps forward in the last chorus, a reverb that opens up in the bridge, a dry, close verse. Our comparison of a vocal rider vs a compressor explains why the two work best together.

How to mix backing vocals and ad-libs

• High-pass higher: 150 to 250 Hz on backing vocals, so they support the lead without adding low-mid weight.
• Less presence: cut 1 to 2 dB around 3 kHz, so the lead keeps the front of the mix.
• Compress harder: 4:1 to 6:1, so a stack behaves like one even pad.
• Pan doubles wide: hard left and right for doubles, and keep the lead in the centre.
• Ad-libs: a narrower band (high-pass at 200 Hz or more), more saturation and more delay than the lead, panned off-centre.

If you need harmonies you didn't record, a harmonizer can build them from the lead; our vocal harmonizer guide covers how.

How to mix vocals in FL Studio

The method is the same in every DAW; only the plugin names change. In FL Studio, route the vocal to its own mixer insert and load the stock plugins in this order:

1. Pitcher or NewTone for pitch correction (Pitcher in real time, NewTone for note-by-note editing).
2. Fruity Parametric EQ 2 for the high-pass filter and the mud cuts.
3. Maximus with a de-esser preset for sibilance.
4. Fruity Limiter in COMP mode, or Fruity Compressor, for 3 to 6 dB of compression.
5. A second Fruity Parametric EQ 2 for presence and air.
6. Fruity Delay 3 and Fruity Reeverb 2 on separate inserts used as sends, with the vocal insert routed to them.

Third-party plugins load into the same slots. A one-window chain like Velvet replaces slots 2 to 5 plus a short reverb; Tune replaces Pitcher.

How to make your voice sound better on a recording

Mixing starts at the microphone. These five habits improve a vocal more than any plugin:

• Distance: sing 15 to 20 cm (6 to 8 inches) from the mic, with a pop filter in between. Closer adds bass and plosives, further adds room.
• Level: record at 24-bit with peaks around −12 to −6 dBFS. There's no benefit to recording hotter, and a clipped take can't be fixed.
• Room: hang a duvet or blankets behind you and to the sides. The mic hears what's behind the singer.
• Headphone mix: give the singer enough of their own voice, and a touch of reverb, to sing in tune and with confidence.
• Takes: record three to five full passes and comp the best lines.

Vocal mixing mistakes to avoid

• Mixing in solo. Make final decisions with the beat playing.
• Judging louder as better. Match levels when you bypass a plugin, or the louder version always wins.
• Too much air before de-essing. The top-end boost turns every S into glass.
• Reverb on the insert. It pushes the vocal back and can't be EQ'd separately. Use a send.
• One compressor doing 10 dB. Split the work between clip gain, a rider and two gentle compressors.
• Copying a preset without adjusting it. Presets are starting points; adjust the low cut, de-essing and compression for each voice.

Once the vocal sits, the last stage is the master. Our guide to the mastering chain order picks up from there, and if you're still writing the song, start with how to write a hit song.