Guide · 4 min read · Updated 12 October 2026
Why your voice sounds muddy, boxy, nasal or harsh — and which frequencies to fix
A voice is not one sound but a stack of ranges, from deep to bright, and each range has its own character. When a recording sounds wrong you can usually name the problem in one word — muddy, boxy, nasal, harsh, muffled — and every one of those words points to a particular range. This guide is the map, and the voice equalizer lets you apply it to your own recording.
The voice, range by range
Hz (hertz) is how many times per second the sound vibrates: low numbers are deep, high numbers are bright. This is what each part of the spectrum does to a speaking voice:
| Range | What it is | Too much | Too little |
|---|---|---|---|
| Below 100 Hz | Almost none of the voice; room and desk rumble | Rumble, thumps, a heavy recording | Nothing lost |
| 100–400 Hz | The body: weight and warmth | Boomy, muddy | Thin |
| 400–800 Hz | The “box” | Boxy, like talking into a cardboard box | Hollow |
| 800–1,500 Hz | The nose | Honky, nasal | Usually fine |
| 1.5–5 kHz | Clarity: what makes words understandable | Harsh, tiring | Muffled, dull |
| Above 12 kHz | Air | Hissy | Closed in, as if behind a curtain |
Start with the one word that describes the problem, find its row, and only move that range. Changing everything at once makes it impossible to hear what helped.
Muddy or boomy: look at 100–400 Hz
Too much energy in the body range makes a voice heavy and unclear, especially when you hold the microphone close (the proximity effect boosts the lows). Cut a few decibels around 200–300 Hz with a broad curve. If the voice now sounds thin, you went too far: bring it back halfway.
Rumble below 100 Hz is different: it is not part of the voice at all, so a high-pass filter at about 80 Hz removes it safely. Do this first on almost every voice recording.
Boxy or hollow: 400–800 Hz
A small, hard room or a microphone too close to a wall makes a voice sound like it is inside a box. A narrow cut of 2–4 dB near 500–600 Hz removes it. If the voice sounds hollow instead, there is too little here: a small boost brings back a sense of substance.
Nasal or honky: 800–1,500 Hz
This is where the nasal quality sits. Some voices have a lot of it naturally, and cheap microphones make it worse. Sweep a narrow cut around 1 kHz and stop when the “honk” disappears. Cut less than you think you need: this range also carries the vowels that make a voice sound like a person.
Harsh or muffled: 1.5–5 kHz
This range decides how easy the words are to understand. Too much of it is harsh and tiring to listen to; too little makes the voice sound muffled, as if the speaker is talking through a scarf. A gentle boost of 1–3 dB near 3 kHz helps a muffled voice; a cut of the same size calms a harsh one.
Air above 12 kHz
A touch of boost up high adds openness and presence. It also boosts hiss, so use it sparingly and only on clean recordings. Many phone and video recordings contain almost nothing up there, and boosting nothing only raises noise.
A safe order of work
- Cut rumble with a high-pass filter at 80 Hz.
- Fix the one thing that bothers you most (muddy, boxy, nasal or harsh) with a small cut.
- Only then add a little clarity or air if the voice still feels dull.
- Compare with the original at the same loudness. A louder version always sounds better at first, which makes this the most common EQ mistake.
Cutting a problem usually sounds more natural than boosting everything around it, and moves of 1–3 dB are normally enough. An equalizer cannot fix a very bad room or a clipped recording, so record closer, in a softer space, if you can.
🎙 EQVoice equalizerFind out why your voice sounds muddy, boxy, nasal or harsh and fix it with plain-language sliders and Hz filters. Free, private, runs in your browser.Frequently asked questions
Why does my voice sound muddy on a microphone?
Usually too much energy between 100 and 400 Hz, often from speaking close to a directional microphone or from a small, untreated room. A broad cut around 200–300 Hz and a high-pass filter at 80 Hz are the usual fix.
Which frequency makes a voice sound nasal?
Roughly 800 to 1,500 Hz. A narrow cut around 1 kHz reduces it.
What does 12 kHz and above do for a voice?
It is the “air”: a small boost makes the voice sound open and close to the microphone. It also lifts hiss, so use it only on clean recordings.
Should I boost or cut?
Cut the problem first. A cut sounds more natural than boosting the other ranges, and the recording keeps more of its headroom.
Can I use this on a video?
Yes. The voice equalizer can write the video back with the new sound; the picture is copied as it is.