How to Mix Vocals: The Complete Step-by-Step Guide

11 min read

Vocals are the loudest, most important element in 90% of modern music. When they're wrong, the whole track feels wrong — muddy, distant, harsh, or buried. When they're right, the listener doesn't think about them at all; the vocal just sits there, present and emotional, riding the beat effortlessly.

The good news: mixing vocals is a repeatable process, not a mystery. There is a sensible order of operations, and once you know it, you can apply it to any vocal in any genre. This guide walks through every step — from raw takes to a finished, release-ready vocal chain — in a way that works in any DAW and with any plugin set.

The goal before the tools

Before you touch a single plugin, decide what the vocal should do in the track. A lead pop vocal sits upfront, dry-ish, and intimate. A rap lead is punchy and present with aggressive dynamics control. A backing vocal sits behind, wide, and softer. An indie rock vocal can be more natural and dynamic. Knowing the target keeps you from over-processing.

The three things every vocal needs:

• Clarity — the listener hears every word.
• Consistency — level and tone stay even across phrases.
• Space — the vocal occupies its own slot, not fighting the instruments.

Every plugin below serves one of these three goals. If a move isn't doing that, skip it.

Step 1: Clean up and edit the takes

Skip this and you'll spend the rest of the mix fighting noise and unevenness. A clean edit is worth more than any expensive plugin.

Step 2: Gain staging

Get the vocal hitting the mix bus at a sensible level — around -18 dBFS average, peaking no higher than -6 dBFS. If you've already read our gain staging guide, this is the same principle applied to the vocal channel. Proper staging means every plugin downstream behaves predictably.

If your recorded vocal is quiet with a noise floor, use a clip gain or trim plugin to bring it up before compression. If it's already loud, trim it down. Don't let the compressor do level matching — that's not its job.

Step 3: Corrective EQ (subtractive first)

Before boosting anything, remove what doesn't belong. This is the most important EQ work you'll do on a vocal.

1. High-pass filter at 80–100Hz. Vocals carry nothing useful below this. Cut it and your low end cleans up instantly. Use 12dB/octave to start; steeper if you have audible rumble.
2. Find and cut the box frequency. Sweep a narrow band around 200–500Hz with a 6dB boost. When it gets worse, cut 2–4dB there. Most vocals have a build-up around 250–350Hz.
3. Check for nasal resonance around 500Hz–1kHz. A small 1–2dB cut can remove the "honky" character.
4. Address muddiness — if the vocal sounds thick and unclear, a gentle 1–2dB cut around 300Hz opens it up. (Same fix as our muddy mix guide, applied to the vocal channel.)

After corrective EQ, the vocal should sound cleaner — not necessarily brighter or louder. That's the point. You've removed the masks.

Step 4: Compression (the consistency maker)

Compression evens out the vocal so quiet words and loud words sit together. If you're new to this, our compression basics article covers the controls; here's the vocal-specific application.

First compressor: gentle leveling

• Ratio: 2:1 to 3:1
• Threshold: set so you get 3–5dB of gain reduction on average phrases
• Attack: 10–30ms (let transients through, catch the body)
• Release: auto, or 100–200ms (recover before the next word)
• Make-up gain: match the input level so you A/B fairly

This pass tames the loudest peaks and lifts the quietest words. After it, the vocal fader should sit in roughly the same spot for the whole song.

Second compressor (optional): character and glue

If the vocal still jumps around, add a second, faster compressor — an 1176-style or FET emulation with a 4:1 ratio, fast attack (3–10ms), and fast release. Just 1–3dB of gain reduction here. This catches anything the first one missed and adds punch.

Don't over-compress. If you see gain reduction bouncing 8–12dB constantly, the vocal will sound flat and lifeless. Two gentle compressors always beat one crushing one.

Step 5: De-essing

Compression makes sibilance worse — "s", "t", and "sh" sounds jump out and stab the listener's ear. A de-esser is a frequency-aware compressor that ducks only the sibilant range (usually 5–9kHz).

• Set the de-esser to trigger on 5–8kHz for male vocals, 7–10kHz for female.
• Reduce only 2–4dB on sibilants. More than that and the vocal sounds lispy.
• Place it after compression, since compression creates the sibilance you're fixing.

Step 6: Enhancement EQ (additive, sparingly)

Now you can add character. Keep moves small:

• Presence (2–5kHz): +1–2dB to bring the vocal forward and improve intelligibility. This is where the vocal "face" lives.
• Air (8–15kHz): +1–2dB with a high shelf for openness and detail. Use a wide, gentle band.
• Body (100–200Hz): +1dB only if the vocal sounds thin — usually not needed.

If you find yourself boosting +4dB or more, something earlier in the chain is wrong. Go back to corrective EQ or compression before reaching for more boost.

Step 7: Saturation (optional, for character)

Light tape or tube saturation adds harmonics that help a vocal cut through on small speakers without raising the actual level. Use it subtly — you should hear "warmth," not "distortion." Saturation is especially useful on rap and rock vocals where you want edge.

Step 8: Reverb and delay (creating space)

Effects place the vocal in a room. The wrong reverb makes the vocal distant and muddy; the right one makes it feel like it belongs to the track.

Reverb

• Use a plate or hall for pop, a room for intimate vocals, a short spring for vintage.
• Keep the decay short: 0.8–1.5s for most modern mixes. Long reverbs push vocals back.
• Pre-delay of 20–60ms keeps the vocal upfront while the reverb tail sits behind.
• High-pass the reverb return at 200Hz and low-pass at 8–10kHz so the reverb doesn't muddy the low end or add harshness.
• Send, don't insert. Keep the dry signal clean.

Delay

Slap delay (a single echo, 80–150ms, no feedback) is often more effective than reverb for keeping vocals upfront while adding space. Throwaway echoes on the end of phrases add interest without cluttering. Use a tempo-synced 1/8 or 1/4 note delay with a low-pass filter to keep it tasteful.

Step 9: Automation

This is the step that separates amateur vocal mixes from professional ones. A static vocal fader will never sound finished — some words are too loud, some get buried by the beat, and the chorus needs to feel bigger than the verse.

• Volume automation: ride words that jump out and lift words that disappear. Do this before compression for natural results, or after for surgical control.
• Send automation: increase reverb on the chorus, dry it down in the verse.
• Filter or tone automation: add a slight phone-filter effect on an intro, or open up the high end when the hook hits.

Automate first; reach for a plugin second. Most "vocal won't sit" problems are automation problems, not plugin problems.

Step 10: Mix in context and verify

Solo is for cleaning and editing. Mixing decisions — level, EQ, reverb — happen with the full beat playing. The vocal that sounds great soloed might be completely buried against a full mix, or it might mask the snare.

Once it sits, verify with tools:

• Use a spectral analyzer to confirm the vocal's energy sits above the instruments in the 1–4kHz range.
• Check your LUFS — heavy vocal processing changes integrated loudness.
• Listen on phone speakers. If the vocal vanishes, you have a mono compatibility problem (see our mono compatibility guide).
• Upload your mix to MixDiagnose and let the AI flag remaining frequency conflicts between the vocal and instruments.

Common vocal mixing mistakes

• Boosting presence before cutting mud. A dull vocal often just has too much 250Hz. Cut first.
• Too much reverb. The vocal sounds "professional" soloed but disappears in the mix. Less reverb, more pre-delay.
• One heavy compressor. Two gentle compressors sound more natural and catch more.
• Skipping automation. No plugin fixes a vocal that's 2dB too quiet in the second verse. Automate.
• Ignoring the rest of the mix. Carve space in guitars and keys around 2–4kHz so the vocal doesn't have to fight. See our EQ guide for frequency carving.
• Processing a bad take. No amount of plugins saves a weak performance. Comp better first.

The vocal chain, summarized

1. Edit and comp the takes.
2. Gain stage to -18 dBFS.
3. Corrective EQ: high-pass, cut box and mud frequencies.
4. First compressor: gentle leveling, 3–5dB GR.
5. De-esser: tame 5–9kHz sibilance.
6. Enhancement EQ: small presence and air boosts.
7. Second compressor (optional): fast, character, 1–3dB GR.
8. Saturation (optional): subtle warmth.
9. Reverb and delay sends: short decay, pre-delay, filtered.
10. Automation: ride volume and effects.
11. Mix in context, verify with analysis.

Follow that order on every vocal and your mixes will get more consistent, more professional, and faster to finish. The chain isn't a preset — adjust to the song — but the order rarely changes.

Once your vocal sits, don't forget the rest of the mix. Check out our guides on stereo width, taming harshness, and knowing when your mix is ready for mastering.

← Back to blog