EQ is where most home producers either spend too much time or not enough — boosting frequencies that don't need boosting, cutting the ones that do, or treating every vocal exactly the same way when every vocal is different.
This guide gives you a complete vocal EQ workflow. Not a list of settings to copy — a system for understanding what each frequency does, how to find problems, and how to make decisions that serve the song. Follow it in order, and you'll have a repeatable process that works on any vocal in any genre.
Why EQ Matters for Vocals
A raw vocal recording contains a lot of information that fights for space in the mix. Low-end rumble from the room, harsh consonants, boxy resonances, dull top end — all of these compete with the actual character of the voice you want to preserve. EQ is the tool that separates what you want to keep from what needs to go.
But EQ is also the easiest way to make a vocal sound worse. Too much high-pass, and the vocal loses weight and warmth. Too much boost at 3 kHz, and it becomes aggressive and fatiguing. EQ done wrong turns a great performance into something that's only technically correct.
The goal is surgical cuts — remove what's bad — combined with gentle additions that enhance what's already there. Never use EQ as a fix for a bad performance or a bad recording. If the vocal sounds fundamentally wrong before EQ, fix the source first.
Step 1: The High-Pass Filter — Your Most Important Tool
Every vocal track needs a high-pass filter. Not optional. Not situational. Every time. It removes low-frequency content below a certain point — room rumble, mic handling noise, proximity effect, and any other low-end junk that muddies the mix without contributing anything to the vocal's character.
Set it somewhere between 80 Hz and 120 Hz to start. Roll off at 12 dB/octave or 24 dB/octave — the steeper the better, because you want a clean transition, not a gradual shelf that takes out useful low-end along with the junk.
Set your high-pass to 100 Hz, then listen. If the vocal sounds thin or hollow, back it off to 80 Hz. If you can still hear rumble competing with the kick and bass, push it up to 120 Hz or even 150 Hz. Use your ears — every vocal and every room is different.
Female vocals tend to need higher high-pass settings — try 120 Hz to 150 Hz for higher voices. Male vocals can often go lower, but rarely need anything below 70 Hz. If you can't hear the difference when you bypass the HPF, the HPF isn't doing anything — which means it was set too conservatively, not too aggressively.
Step 2: Finding and Cutting Problem Frequencies
After the high-pass, the next step is finding the specific frequencies that are causing issues in your particular recording. This is where most people go wrong — they apply generic presets instead of listening to what's actually there.
The Sweep Method
The most reliable technique: insert an EQ with a narrow bell curve. Boost it by 6 to 10 dB. Sweep slowly through the frequency range from low to high while the vocal plays. When you hit a frequency that sounds harsh, honky, boxy, or just wrong — that's your problem frequency.
Once you've found it, flip the boost to a cut. A cut of 2 to 5 dB with a narrow Q (1.5 to 4.0) is usually enough. The narrower the Q, the more surgical the cut and the less you affect surrounding frequencies.
| Frequency Range | Common Problem | What It Sounds Like |
|---|---|---|
| 100–250 Hz | Boominess, mud | Uncontrolled low-end, muddys the mix |
| 250–500 Hz | Boxiness | Hollow, cardboard-like quality |
| 500 Hz–1 kHz | Honky, nasal | Character shift, like the voice is coming from a box |
| 1–3 kHz | Harshness | Aggessive, gritty, can cause listening fatigue |
| 3–6 kHz | Sibilance issues | Sharp S and T sounds become painful |
| 6–10 kHz | Presence mask | Vocal disappears in the mix, lacks clarity |
Don't try to fix everything at once. Do one or two surgical cuts, then listen. If the vocal sounds better, move on. If it sounds worse, undo. The goal is to remove just enough to let the vocal sit right — not to reshape it into something it's not.
What to Cut and What to Keep
Not every vocal needs the same cuts. A well-recorded vocal in a treated room may need nothing beyond the high-pass. A vocal recorded in a untreated bedroom with a cheap mic might need cuts at multiple ranges. Listen first, decide second.
Here's a useful principle: cut to solve a problem, not to achieve a target sound. If you can't hear a specific issue, don't fix it. Better to have a slightly raw vocal that sounds natural than a heavily processed one that sounds like a preset.
Step 3: Deessing — Taming Sibilance
Sibilance is the excessive emphasis of S and T sounds. Every vocalist has them — some more than others — and they become much more prominent after EQ boosts in the high-mids and highs. If you've ever heard a vocal that sounded fine until you added high-end, you know exactly what this is.
Use a de-esser, not a static EQ cut. A de-esser listens to the signal and applies gain reduction only when sibilant frequencies exceed a threshold. This means the rest of the vocal is unaffected — you're only controlling the problem moments, not thinning out the entire track.
Settings to start with:
| Parameter | Starting Point | Notes |
|---|---|---|
| Frequency | 6–8 kHz | Where harsh S and T sounds live |
| Threshold | Start high, lower until the de-esser activates on peaks | You want it to grab only the worst sibilance |
| Reduction | 3–6 dB | More than 8 dB usually sounds unnatural |
| Attack | 0.5–2 ms | Fast enough to catch the transient of the S |
| Release | 20–50 ms | Fast enough to reset between words |
If you don't have a de-esser plugin, a narrow cut around 6–8 kHz with a Q of 4–8 will reduce sibilance, but it will also reduce presence and air across the whole vocal — not ideal. A dynamic EQ approach (a cut that only applies when the signal exceeds a threshold) is a good alternative if your EQ has that capability.
Step 4: Adding Presence and Air
Once you've cleaned up the low-end and tamed the harsh frequencies, you can think about adding. This is where most people go wrong — they boost before they cut, and they boost too much. Additive EQ should be gentle. If you can't hear the difference between a 2 dB boost and a 4 dB boost, use 2 dB. Less is almost always more.
Presence Range (3–6 kHz)
This is where vocals cut through the mix. If a vocal sounds dull or gets lost, a gentle boost here adds clarity and definition. A wide bell curve with +2 to +3 dB at 4 kHz can make a vocal feel more present without adding harshness. Don't boost below 3 kHz — you'll make the vocal sound nasal or boxy.
Air Range (8–16 kHz)
A gentle shelf boost above 8 kHz adds air and sparkle to a vocal that sounds dark or closed-in. This is especially useful for vocals recorded with darker-sounding microphones. Use a wide shelf, and keep the boost under +3 dB. Too much air boost sounds brittle and artificial — exactly the opposite of what you're going for.
Add air and presence last, after all your cuts. If you add presence first and then start cutting, you'll end up boosting even more to compensate — and that's how you get a vocal that's over-processed and unnatural.
Step 5: EQ Automation — One Size Does Not Fit All
The biggest mistake home producers make with vocal EQ: set it once and leave it. A verse vocal might need different EQ than a chorus vocal. A quiet phrase might need more presence than a loud one. Static EQ cannot fix dynamic problems — automation can.
If you're mixing in a DAW with automation capabilities (all of them), automate your EQ parameters for different sections. A gentle boost at 4 kHz in the chorus that adds presence and energy might make the verse feel overwhelming. Automate the blend so it's there when you need it and gone when you don't.
For problem frequencies that only appear on certain words or phrases, use automation to apply a temporary cut. A resonant word that sounds harsh in the bridge but nowhere else doesn't need a global cut — it needs a temporary one.
Common Mistakes to Avoid
1. Boosting before cutting. Always clean up first, enhance second. If you start with boosts, you'll end up cutting the same areas to compensate, achieving nothing useful.
2. Using the same EQ on every vocal. No two recordings are the same. The mic, the room, the vocalist, the song — all of these change what the vocal needs. Use your ears every time, not a preset from a YouTube tutorial.
3. Over-EQing. If your EQ is doing more than removing a few specific problems and adding a little air, you've gone too far. A vocal that's been surgically cleaned and gently enhanced should sound like the original performance, just cleaner. If it sounds processed, you've lost the plot.
4. Not checking in mono. Phase issues and frequency conflicts that show up in mono will ruin a mix faster than anything. Check your vocal in mono after EQ — if something changes negatively, you've introduced a problem.
5. Ignoring the mix context. A vocal EQed in isolation might sound perfect — but in the mix, it might be fighting with the guitars, the snare, or the keyboards. Always EQ in the context of the full mix, not the vocal track solo.
A Practical Workflow
Here's the order to do this in every time:
- Gain stage. Make sure your vocal is at a healthy level before any processing.
- High-pass filter. Set between 80–120 Hz. Listen and adjust.
- Sweep for problem frequencies. Use a narrow boost to find them, then flip to a cut.
- Deess. Set threshold to catch only the worst sibilance peaks.
- Add presence. Gentle boost at 3–5 kHz if the vocal needs it.
- Add air. Wide shelf above 8 kHz if the vocal sounds dark.
- Automate. Adjust EQ for different sections and dynamic moments.
- Check in mono. Verify nothing has become worse in the full mix context.
That's it. Seven steps, applied with your ears every time. No presets, no guesswork — just a clear process that works.
Once you've EQed your vocal, learn how to complete the vocal mixing chain — compression, reverb, delay, and automation that makes the difference between a rough mix and a record.