How to Remove Plosives and Pops

Flat illustration of a studio microphone behind a round pop filter beside a clean waveform, showing how to remove plosives and pops

To remove plosives from a recording, cut the low-frequency thump they leave behind: pull the 80 Hz low shelf down in the free Equalizer, a 6-band audio equalizer that runs in your browser, or run speech through the Audio Enhancer in Voice mode, which applies an 80 Hz high-pass filter (a low cut) for you. Pops that are still loud after that, usually because they clipped, need a local fix or a retake, and a pop filter stops most of them before they happen. Everything runs in your browser, and your audio never leaves your device.

What a plosive is

A plosive is the puff of air you push out on certain consonants. "P" and "b" are the worst offenders; "t", "d", "k" and "g" do it too, usually more gently. Hold your hand in front of your mouth and say "paper": you'll feel it. When that puff hits a microphone capsule head-on, the diaphragm gets shoved by moving air rather than by sound, and the result is a dull thump or pop on the recording.

Most of that energy sits low, often below about 100–150 Hz, well under the range that makes speech intelligible. That's why a low cut works so well on plosives. A hard one can also overload the mic or recorder for a split second, flattening the waveform, and that kind of damage no filter can undo.

Plosives get confused with two other vocal problems that need different fixes:

  • Mouth clicks are short, sharp ticks from saliva and lip noise. They live in the mids and highs, so a low cut does nothing for them.
  • Sibilance is the harsh hiss on "s" and "sh" sounds, roughly 4–10 kHz depending on the voice. That's a job for a de-esser or a narrow EQ cut up top, not a low cut.

If what you're hearing is a soft thud on "p" and "b", it's a plosive.

Spotting plosives on a spectrogram

Load the file into the Spectrogram and look along the bottom edge. Each plosive shows up as a short, bright burst in the low frequencies, lined up with the start of a word, unlike the steady line of hum or the drifting blotches of wind. A burst that climbs well up into the mids was a strong one and may have clipped. How to Read a Spectrogram explains what you're looking at.

What you'll need

  • A recording you already have: a voice memo, podcast episode, vocal take or voiceover, in a common format like WAV, MP3, M4A or FLAC.
  • A browser. No account, no upload, no install.
  • Headphones. Laptop speakers barely reproduce the low end, so you won't hear a plosive's thump, or whether your low cut fixed it, without them.

Step 1: remove plosives with a low cut

A low cut (or high-pass filter) turns down everything below a chosen frequency. A speaking voice has very little useful content under about 80 Hz, so you can cut there firmly and the voice still sounds like itself. Vocal Cut gives you two ways to do it.

Option A: the Audio Enhancer in Voice mode. It's the quickest route for spoken word. Voice mode applies a fixed 80 Hz high-pass, a clarity EQ, compression, and loudness normalization to −16 LUFS with true peaks kept under −1 dBTP. A Strength slider sets how hard the EQ and compression work. Voice mode also switches on its noise-reduction step by default; leave it on for a noisy room or turn it off if the voice starts to sound processed. The 80 Hz cutoff isn't adjustable, but for most voices it's a sensible place to start, and you get a finished, levelled file in one pass.

Option B: the Equalizer. This one gives you more control. It's a 6-band graphic EQ with fixed bands: an 80 Hz low shelf, peaking bands at 240 Hz, 750 Hz, 2.2 kHz and 6 kHz, and a 12 kHz high shelf. Each band moves from −12 to +12 dB in 1 dB steps.

  1. Open the Equalizer and add your file. Drag it onto the drop zone or click to browse.
  2. Pull the 80 Hz band down 6–12 dB. Start around −6 dB (deeper than the 3–6 dB you'd use for general vocal cleanup), play a passage with a few hard "p" words, and keep lowering until the thumps sit back.
  3. Touch 240 Hz only if boom remains. If the pops still sound boomy, take a dB or two off the 240 Hz band. Be careful: that band carries the body and warmth of a voice, and too much cut makes it sound thin.
  4. Compare and reset if needed. Use the play button to preview. If you've overdone it, Flatten puts every band back to zero.
  5. Export. Choose WAV, MP3 or Opus (where your browser supports it) in the format selector, then click Export WAV (the label follows the format you picked).

Graphic vs parametric audio equalizer for plosives

The Equalizer's 80 Hz band is a low shelf, not a true high-pass. It lowers the whole low end gently, all the time, rather than removing only the pop. On most speech that's fine. On a deep voice you may hear a little body go with it.

A parametric audio equalizer, or the high-pass filter in a free editor like Audacity, lets you set the cutoff yourself. The usual approach is to start around 80 Hz and raise it slowly, toward 100–120 Hz, only until the pops stop thumping, then back off a touch. If you want to understand the trade-off between the two styles, Parametric vs Graphic EQ: The Difference covers it.

Step 2: fix the pops that survive

A low cut takes care of most plosives. The ones still sticking out afterwards are usually the loud ones: they carried energy higher up the spectrum, or they clipped. Those need attention one at a time.

Vocal Cut doesn't have per-region editing inside a file, so you can't select one pop and filter just that spot. Here's what you can do:

  • Trim a pop at the start or end. In the Audio Cutter, set the Trim Start point and Trim End point sliders to keep the part you want, then click Cut & Download. That's ideal for a pop on the first word of a take.
  • Cut around a pop in the middle. Cut the section before the pop and the section after it as two separate files, then rejoin them with the Audio Joiner. Only do this where a tiny gap, a breath or a pause, won't be noticed.
  • Soften an edge. A short fade with Audio Fade can take the edge off a pop right at the start or end of a clip.

If you have a DAW or audio editor, you can be more surgical: select just the pop, a few tens of milliseconds, and apply a high-pass to that region alone. Because it only touches the pop, you can set it higher than a whole-file cut, somewhere around 150–250 Hz, without thinning the rest of the voice. Or lower its clip gain until it blends in. A volume ride on that syllable does the same job.

If a pop clipped, the top of the waveform is flattened and the information is gone. You can turn it down, but you can't rebuild it. Re-record the line if you can.

Step 3: stop plosives before they happen

Every fix above costs the voice a little. Prevention costs nothing, and it does the most.

  • Use a pop filter. A mesh or foam screen breaks up the burst of air before it reaches the capsule. Place it roughly 2–4 inches (5–10 cm) in front of the mic, and keep your mouth a little further back from it. More on how they work in What Is a Pop Filter? Plosives Explained.
  • Go off-axis. Aim the mic at your mouth from slightly above or to the side rather than straight on. The puff of air travels forward and misses the capsule, while your voice still reaches it. Mic Placement for Clean Voice Recordings shows the common setups.
  • Back off a little. Plosives get much worse up close. An extra few centimetres of distance often tames them without changing the sound much.
  • Test first. Say a few "p" and "b" words into the Mic Test before you record and watch for spikes.

For a complete setup routine, see Record Clean Vocals at Home on an Audio Recorder or, for spoken work, Record a Voiceover With an Online Voice Recorder.

Troubleshooting

The voice sounds thin after the low cut. You cut too much, or you pulled 240 Hz down too far. Give back a few dB on the 80 Hz band and leave 240 Hz flat.

The pops are still there. They probably clipped, or their energy reaches higher than your cut, up into the 200–400 Hz range on some hard plosives. A deeper EQ cut there will hurt the whole voice, so fix those pops locally (Step 2) instead.

It's a vocal over music, and the bass disappeared. A low cut on a full mix takes the kick and bass with it. Work on the vocal track before it's mixed if you have it. If you only have the finished song, you could separate the vocal first with the Vocal Remover, clean that stem, and rebuild the mix, but AI separation takes real processing time on a long track, and the result won't be perfect.

Related cleanup guides

The Noise Reducer handles steady background noise like hiss, fans and room tone. It isn't built for plosives, though: a pop is a short burst inside the voice itself, not steady background noise. Once the pops are under control, How to EQ a Vocal in Your Browser and How to EQ a Podcast Voice for Clear Speech cover shaping the rest of the voice. For outdoor recordings, the low-end problem is often wind rather than breath; How to Reduce Wind Noise in Recordings walks through that fix.

Frequently asked questions

What frequency should I cut to remove plosives?

Start with a high-pass or low cut around 80 Hz for most voices, then raise it slowly toward 100–120 Hz only until the pops stop thumping. Deep male voices may need a lower cutoff to keep their body; lighter voices can usually take a higher one. With a graphic EQ, pull the lowest band down 6–12 dB instead, and listen on headphones while you adjust.

Can you remove plosives after recording?

Mostly, yes. A low cut removes the thump from the majority of plosives, and the stubborn ones can be trimmed, faded or turned down individually. What you can't fix is a pop that clipped the recording: the waveform is flattened and the missing detail is gone. For those, lowering the pop helps, but re-recording the line gives the cleanest result.

Does a pop filter stop all plosives?

No, but it stops most of them. A pop filter breaks up the burst of air before it hits the capsule, which is usually enough for normal speech and singing. Very forceful "p" sounds, or a mouth right up against the screen, can still get through. Combining a pop filter with slightly off-axis mic placement and a bit more distance handles nearly all the rest.

What's the difference between plosives and mouth clicks?

Plosives are low-frequency thumps caused by bursts of air on consonants like "p" and "b". Mouth clicks are short, sharp ticks from saliva and lip movement, and they sit in the mid and high frequencies. A low cut fixes plosives but does nothing for clicks, which need a de-clicker or manual editing, and drinking water while you record helps prevent them.

Will a de-esser fix pops?

No. A de-esser targets sibilance, the harsh "s" and "sh" energy roughly between 4 and 10 kHz. Plosives live at the other end of the spectrum, mostly below 150 Hz, so a de-esser won't touch them. Use a low cut or high-pass filter for pops, and save the de-esser for hissy consonants.