AubiomixVisit Website →
← Back to blog

Stop Shaving Tone: Mix Engineers' A/B Workflow for Vocal High Pass

October 7, 2026
Stop Shaving Tone: Mix Engineers' A/B Workflow for Vocal High Pass

Yes, use a high pass filter on vocals as a surgical tool to clear rumble and free up room for bass and kick, not as a blanket default on every track. Sweep up from 20 Hz while soloing the vocal, find where the tone starts thinning, then back off. Most leads settle somewhere in a moderate low-frequency range suitable for vocals, but slope and phase mode change how that cutoff actually sounds.

TL;DR:

  • Use a high pass filter mainly as a surgical tool to remove rumble, starting around 50 to 100 Hz depending on the vocal type, and always sweep with your ears.
  • Steeper slopes such as 24 or 48 dB per octave create sharper cuts but risk removing vital vocal warmth, especially on chestier male voices.
  • Always verify cutoff points with full mix checks and multiple playback systems to prevent thinning the singer's fundamental frequencies.
  • Minimize phase issues by defaulting to minimum-phase filters unless working on multiple related tracks where phase alignment is critical.
  • Avoid aggressive high-passing on low-voiced singers or when low-end warmth is part of the vocal character, opting for dynamic EQ or subtractive cuts instead.
Aubiomix
aubiomix.com
Validate Your Vocal Mix Decisions
Aubiomix gives producers and mix engineers detailed feedback and actionable steps on mixing and mastering uploaded songs.
Get detailed feedback

Table of Contents

Quick settings and rules of thumb

Every voice sits differently in the frequency spectrum, so starting points are just that: starting points, not presets to copy blindly. We find these ranges hold up across genres, but your ears get the final vote.

  • Lead male vocals: start around 50 to 70 Hz and sweep from there.
  • Lead female vocals: start around 60 to 100 Hz, since the fundamental typically sits higher.
  • Background vocals: push further, 80 to 120 Hz, since they rarely need low-end weight.
  • Spoken word: similarly 80 to 120 Hz, as clarity matters more than warmth here.

Slope matters as much as the cutoff frequency itself. A gentle 6 to 12 dB per octave filter removes rumble without touching much of the perceived body of the voice. Push to 24 or 48 dB per octave and you get a sharper, more surgical cut, but you risk shaving off meat the singer actually needs, especially on chestier male vocals. Steeper isn't automatically better; it is just a different tool for a different problem, like feedback control on a live stage versus a gentle mix bus tidy-up.

One detail that trips people up: the low-cut switch on a microphone or preamp is not the same as a plugin HPF. Hardware low-cuts are usually fixed at a single frequency with a gentle slope, baked in at the source and impossible to undo later. A plugin HPF gives you full control after the fact, which is why most engineers leave the hardware switch off and handle it properly in the mix.

Quick settings and rules of thumb — overview diagram

How to choose the right cutoff: sweep method and listening checks

Finding the correct cutoff isn't about memorising a number, it's about listening while you move the filter. This is the single most useful habit you can build into your vocal chain.

  1. Engage the HPF at 20 Hz on a solo'd vocal track, where it has effectively no audible effect yet.
  2. Slowly raise the cutoff while listening in solo, watching for the point where the voice starts to sound thin or hollow.
  3. Note that frequency, then back off by roughly 10 to 20 Hz, or pull the cutoff down by 3 to 6 dB worth of movement.
  4. Switch out of solo and listen in the full mix, since a cutoff that sounded fine alone can suddenly feel thin against a dense arrangement.
  5. Cross-check with a spectrum analyser to confirm you haven't cut into the singer's actual fundamental.
  6. Repeat the check on a second playback system, phone speaker, car, headphones, since low-end decisions translate unpredictably.

A typical male voice fundamental averages around 115 to 120 Hz, while a typical female voice fundamental averages around 200 to 210 Hz, according to the University of Minnesota Lions Voice Clinic. That spread is exactly why a single universal cutoff setting is unsafe: a filter set confidently for a tenor could quietly eat into a bass singer's actual pitch centre. Always sweep by ear for the performance in front of you, then use the published averages as a sanity check rather than a rule.

Filter type, slope and phase: minimum phase vs linear phase and audible artefacts

Most plugin HPFs default to minimum-phase operation, and for good reason: low latency, simple behaviour and a sound most engineers have trained their ears on for years. The trade-off is some phase shift around the cutoff frequency, which is usually inaudible on a single track but can add up when you stack several filtered signals together.

Linear-phase filters solve that phase-shift problem by keeping all frequencies aligned in time, which matters when you are high-passing multiple correlated sources, like a doubled vocal or a background stack, and want them to sum cleanly. The cost is pre-ringing, a small pre-echo that can smear transients, plus added latency that needs compensating or committing offline.

  • Minimum-phase: low latency, slight phase shift, the standard choice for tracking and most mix work.
  • Linear-phase: preserves phase relationships, introduces pre-ring, better suited to stacked or transient-critical material.
  • Steep slopes on either type can cause ringing right at the cutoff frequency, according to iZotope's guidance on high-pass filters, so a gentler slope is often the safer first move before reaching for a steeper one.

Pro Tip: Default to minimum-phase for everyday vocal work, and reach for linear-phase only when you are correcting multiple related tracks at once, like a five-part harmony stack, where phase alignment actually changes how they sum.

When not to high-pass and better alternatives

A high-pass filter is not automatically correct just because it is common practice. Plenty of vocal performances need their low end left alone.

  • Avoid aggressive high-passing when the low end is part of the singer's desired warmth, particularly on baritone, bass-register or operatic vocalists whose fundamental sits lower than average.
  • Reach for a dynamic EQ instead when the problem is intermittent, only showing up on certain notes or phrases, rather than a constant rumble across the whole take.
  • Use a narrow subtractive cut for a specific resonant frequency rather than a broad HPF that takes a chunk of everything below it.
  • Try a gentle low shelf when you want to tame low-end weight while keeping some tonal colour, since a shelf rolls off gradually rather than cutting hard.
  • Address proximity effect with mic distance or a different polar pattern before reaching for a filter plugin, since fixing it at the source avoids processing altogether.

Shure's guidance for vocal engineers consistently favours this kind of minimalist approach: fix it with technique first, then filter only what remains. Our own piece on safe high-pass cutoffs for mixers walks through specific cutoff mistakes to avoid.

Routing and stacking: track, bus and master considerations

HPFs are cumulative. A gentle 6 dB per octave filter on an individual background vocal track seems harmless, but once you've applied similar filters to eight stacked harmonies, then added another HPF on the vocal bus, the combined slope around that frequency region can be far steeper than any single filter suggests.

  • Audit your chain by listening to the full bus with all filters engaged, not just checking each track in isolation.
  • Prefer modest, track-level HPFs for individual problems, then apply one corrective HPF on the bus only if the combined signal still feels muddy.
  • For a background stack, a light 80 Hz cut per track plus a single 60 Hz bus filter usually tidies things up without hollowing the group out.
  • For a lead vocal chain, keep the track-level filter as the primary tool and skip the bus filter entirely unless the rest of the arrangement demands it.

Practical workflow checklist and presets to try

A repeatable process beats guesswork every session. This is the order we follow, start to finish.

  1. Listen to the raw vocal solo and in context before touching anything.
  2. Engage the HPF at 20 Hz as your neutral starting point.
  3. Sweep upward until you hear thinning, then back off.
  4. Choose minimum-phase or linear-phase based on whether you're stacking related tracks.
  5. Recheck in the full mix and on a second playback system.
  6. Validate with an A/B comparison before committing.

Keep a short note of what you settled on and why for each session, so future mixes on the same artist start from a known position rather than from scratch.

Testing and verifying your HPF choices

The surest way to know whether a cutoff actually helped is to compare both versions objectively rather than trusting memory alone. Export the vocal before and after your HPF change, then run both through AubioMix for a side-by-side mix analysis.

A/B vocal HPF analysis workflow

Review the low-frequency masking score, the spectral balance chart and any translation flags across both reports. A well-judged HPF should lower masking against the bass and kick without measurably reducing the energy around the singer's actual fundamental, confirming the cut solved a problem rather than just thinning the voice.

Author's perspective: principled minimalism and musical context

The best HPF decision is often no HPF decision, because the real fix was a better mic position, a quieter room or a performance re-take. We'd rather solve a masking problem with arrangement, level or panning than reach for an EQ curve first, since those tools often do the job with less collateral damage to tone. When a filter genuinely earns its place, verify it on more than one system and keep a short note of the setting, so the next session starts from judgement rather than guesswork.

— AubioMix

How AubioMix can help you validate your HPF decisions

Once you've made your cutoff and slope choices, a second opinion backed by measurement is worth having before the mix goes out the door. We offer a single Mix Review for one-off checks, alongside Professional and Professional + plans for mixers who want regular, data-driven feedback across every element of a track, high pass decisions included.

Aubiomix

  • A single Mix Review costs £4.99 one-off, with an optional £2.99 Reference Pass if you want a benchmark track compared alongside your own.
  • Reports flag low-frequency masking, spectral balance and translation issues, giving you exact points to revisit rather than a vague "it sounds fine" verdict.
  • Subscription plans are offered with varying monthly fees to accommodate different review volumes.

Upload your pre and post HPF versions and see the difference measured, not guessed at.

FAQ

What is the purpose of a high-pass filter on vocals?

A high-pass filter removes low-frequency rumble, mic handling noise and unwanted bass buildup from a vocal track, clearing room for the kick and bass in the mix. It works best as a corrective tool for a specific problem rather than a default setting applied to every track, according to iZotope's guidance.

What should my HPF be set to?

There's no single correct number, since it depends on the singer's voice and the slope you choose. Sweep from 20 Hz upward while soloing the vocal until the tone thins, then back off; most leads land somewhere between 50 and 100 Hz.

What is the rarest vocal range?

Vocal range classification varies by source and teaching tradition, so there's no single agreed answer to which range is rarest. What's well established is that fundamental frequency varies widely between singers, with male voices averaging around 115 to 120 Hz and female voices around 200 to 210 Hz, per the Lions Voice Clinic.

Why does Billie Eilish sing so quietly?

Quiet, breathy vocal delivery is a stylistic and performance choice rather than a mixing technique, and it isn't something a high-pass filter creates or removes. It's handled through microphone technique, performance dynamics and production choices well before an engineer reaches for an EQ.

Sources