Nasal vocals are not a problem you fix by dragging down every midrange band. First hear whether the pinched tone is in the dry take, appears only after processing, or shows up when the beat is playing. Then make the smallest change that solves that version of the problem.
If the voice sounds like it is stuck behind the nose while the words are still clear, this guide gives you a repeatable way to find and control that sound in a music mix. It is about an audio result, not a diagnosis of a singer or advice about vocal health.
Rys Up Audio makes RysUpEQ, which appears below. The listening steps work with the stock EQ in your DAW too. We have not tested one universal setting that works for every singer.
Hear the real problem before touching EQ
Loop one phrase where the pinched tone jumps out. Listen to the dry track with the beat muted, then listen with the full arrangement. Finally bypass the vocal plugins one at a time. Write down the first state where you notice the problem. If it is already obvious in the dry recording, changing your reverb or de esser is unlikely to be the best first move.
If the dry voice sounds balanced but the processed version turns honky, bypass any broad mid boost, saturation, and heavy compression. A processor that keeps pushing the same part of the voice forward can create the impression of nasality even when the original take is fine. Level match before comparing so the louder version does not automatically win.
Do not confuse nasal, boxy, and sharp
Nasal is usually a focused midrange character. Boxy is a lower, closed, cardboard like buildup. Sharp consonants are a different event again. Our boxy vocal guide handles room and low mid buildup, while the de essing guide handles consonants that jump out for a moment. A big cut aimed at the wrong problem can remove presence without solving the complaint.
iZotope's current vocal EQ guide uses roughly 800 Hz to 1.5 kHz as a search area for nasal resonance. That is a useful map, not a preset. The center can move with the singer, note, mic, and arrangement. Do not carve out that entire range just because a chart names it.
Fix the recording when another take is possible
Record the same phrase twice before processing anything. Keep the performance similar and change just one variable, such as microphone distance, angle, or position. Shure notes that microphone choice and small off axis moves change vocal tone. That makes a new take a real option when the mic exaggerates a quality you do not want.
Listen in the mix, not only solo. Keep the version that sounds open while the singer still feels like themselves. If a change loses clarity or invites room noise, undo it. The goal is not to make the voice look flat on an analyzer. If the recording is already finished, you can still make a careful mix correction without pretending a plugin rewrote the original performance.
Find the offending spot without guessing
Put a bell band on the vocal and loop the exact word that bothers you. Temporarily make a narrow boost and sweep through the broad middle of the spectrum at a safe monitor level. When the ugly tone gets louder, stop. Return the band to zero before making a cut. Work with the full mix on so you can hear whether that spot truly needs less energy.
Start with a small reduction, about 1 to 3 dB, then bypass the EQ at matched loudness. Widen or narrow the band by ear. A very narrow cut can miss a moving resonance, while a broad cut can take the life out of vowels. The number is only a starting move. If you cannot hear an improvement in the song, reset the band.
iZotope's vocal EQ walkthrough demonstrates a targeted nasal move and explains why the resonant point differs between voices. Our complete vocal EQ guide covers the wider tonal workflow once this one issue is controlled.
Choose a static or dynamic cut
If the same honk is present on every word, a gentle static cut is easy to understand and recall. If it appears only on louder notes or certain vowels, a dynamic EQ band may be a better fit because it pulls that range down only when it becomes excessive. FabFilter's own dynamic EQ documentation describes this level responsive behavior. RysUpEQ's current shipping version also has per band dynamics, so you can make this decision on one EQ surface.
Set a dynamic band to react only when the troublesome phrase arrives. Watch and listen for occasional movement, then compare with the band bypassed. If it clamps on every syllable, reduce the range or change the threshold. A moving band is not automatically more natural than a static one. Pick the simpler method that leaves the rest of the song intact.
Make space around the voice if the beat is the issue
Sometimes the vocal only feels nasal when an instrument occupies the same space. Pull the beat down briefly. If the voice opens up, do not punish the singer with a deep vocal cut. Try a small change to the competing synth, guitar, or sample, or adjust its level in the line that masks the lyric. A half dB of arrangement space can beat an aggressive vocal notch.
Then put all processing back in. Check the loudest chorus and the quietest verse. Any correction that helps one line but makes another disappear is too broad for the whole song. Use automation or a dynamic move for that phrase, and leave the rest of the vocal alone.
Run this two minute listening test
- Loop the worst word and level match the vocal with and without your correction.
- Listen solo long enough to identify the tone, then make the decision in the full beat.
- Try a small static cut first. Use a dynamic band only if the problem changes across the performance.
- Switch between verse and chorus. If diction, warmth, or energy gets worse, back the cut off.
- Take a short break, then compare again at ordinary listening volume.
That sequence protects the voice from the classic mistake of making a dramatic solo improvement that vanishes in the song. If the correction is still clearly useful after a break, keep it. Otherwise revert and check the recording or arrangement again.
Where RysUpEQ fits
RysUpEQ is one way to do the static or dynamic move above. The current 2.7.0 release is available through RysUpHub. You can start with the stock EQ you already have and decide whether you want one surface for both the broad vocal tone and frequency dependent control. If you are building a bigger chain, RysUpSuite gives access to the current paid plugin lineup while active.
Keep the character
The best nasal vocal fix is the smallest change that makes the lyric easier to hear without sanding away the singer. Find the source, test in context, and stop when the song works. If the issue is really lower and more enclosed, go to the boxy vocal workflow. If it is a broad tonal imbalance, use the complete EQ guide instead.
Sources checked September 24, 2026: iZotope vocal EQ guidance, Shure vocal recording guidance, FabFilter dynamic EQ documentation, the live RysUpEQ product page, and the current RysUpEQ update feed. Frequency and gain figures here are listening starting points, not a claim about every voice.