How to EQ a Podcast Voice for Clear Speech
To EQ a podcast voice, open the free browser equalizer, load your recording, and make a few small moves: cut the low rumble around 80 Hz, thin out any 240 Hz mud, lift 2.2 kHz for presence, and gently tame 6 kHz if the esses are harsh. The whole thing takes about five minutes, runs entirely on your own device, and needs no account and no upload. This guide maps each podcast-voice goal to the exact band that controls it.
Good spoken-word EQ rarely means drastic changes. Most of the work is clearing away what buries the voice and lifting the frequencies that make consonants easy to follow, so a listener on cheap earbuds still catches every word.
What you'll need
- A recorded voice file in MP3, WAV, FLAC, or M4A.
- Nothing else. The 6-band graphic EQ works in your browser — no software to install, no sign-up, and your audio never leaves your device.
If your recording has steady background noise — a fan, an air conditioner, a faint hiss — deal with that first. EQ shapes tone; it can't cleanly separate a hum from a voice that shares the same frequencies. Run the file through the noise reducer before you touch a single slider, and you'll have far less to fight.
How a graphic EQ works (and its limits)
The equalizer on Vocal Cut is a fixed 6-band graphic EQ. That word "graphic" matters: you control the gain of six preset bands — how much to boost or cut each one, from −12 dB to +12 dB in 1 dB steps. You can't move a band to a different frequency, and you can't narrow or widen it. The six bands sit at fixed points across the spectrum:
- 80 Hz — low shelf
- 240 Hz — peaking
- 750 Hz — peaking
- 2.2 kHz — peaking
- 6 kHz — peaking
- 12 kHz — high shelf
The trade-off is honest. A graphic EQ can't hunt for one exact problem frequency the way a sweepable parametric EQ can. But six sliders are fast, hard to mess up, and cover every broad tonal move a spoken voice actually needs. For clearing mud, adding presence, and warming or brightening a voice, that's the right tool. For surgically notching one resonant ring or true de-essing a heavily sibilant voice, it isn't — and this guide flags where that limit bites.
Step-by-step: EQ a podcast voice
1. Load your file and start previewing
Drop your recording onto the free browser equalizer, or click to select it. The Graphic EQ panel appears with the six band sliders. Press the round Play button to preview — and here's the useful part: while it's previewing, moving any slider updates the sound live, so you hear each change as you make it. EQ by ear, not by numbers.
Every dB value below is a starting point. Every voice, mic, and room differs, so set it, listen, and trust your ears from there.
2. Cut the rumble — 80 Hz, about −3 to −6 dB
Pull the 80 Hz low shelf down a few dB. This is where room rumble, HVAC hum, desk thumps, footsteps, and the low-frequency thud of plosives (hard p and b pops) live. Almost none of a spoken voice's intelligibility sits down here, so clearing it out tightens everything above. One caution: a very deep voice carries real fundamental energy near 80 Hz, so don't over-cut, or it goes thin and lifeless. Back off until it sounds clean but still grounded.
3. Clear the mud — 240 Hz, about −2 to −3 dB (only if needed)
If the voice sounds boomy, boxy, or muddy, the culprit usually sits around 240 Hz. Close-mic recording exaggerates it through the proximity effect — sitting close to a directional mic boosts the bass, and a warm voice tips into muddy. A small cut here opens things up. If you want the full picture on why buildup in this range clouds a recording, see why your mix sounds muddy. Only cut if it's actually muddy — a clean, balanced voice doesn't need it.
4. Leave 750 Hz mostly alone
The 750 Hz band carries the body and warmth of the voice. Usually leave it flat. If the voice sounds boxy, nasal, or "cardboard," a small −1 to −2 dB cut can help — but go gently. Cut too much here and you hollow the voice out, trading a minor problem for a bigger one.
5. Add presence — 2.2 kHz, about +1 to +3 dB
This is the single most useful move for spoken-word clarity. The 2.2 kHz band sits squarely in the presence and intelligibility range, where consonant definition lives. A modest boost lifts the edges of words so speech cuts through — especially on laptop speakers, phone speakers, and earbuds, where the low end barely reproduces and clarity is all the listener has. If you do only one thing, do this.
6. Tame or brighten — 6 kHz, −1 to −3 dB (or +1 to +2 dB)
The 6 kHz band is the sibilance and harshness region — the ess, sh, and t sounds. If the voice is harsh or essy, cut −1 to −3 dB. If it's dull up top, a small +1 to +2 dB boost adds crispness instead.
Be realistic about what this does. Because the band is broad and fixed (its Q is roughly 1), cutting it tames sibilance across a whole range rather than surgically removing it — so it can dull a little general brightness while it softens the esses. That's a fair trade for a mildly essy voice. For genuinely heavy sibilance, a dedicated de-esser will do a cleaner job than any graphic EQ can.
7. Add air — 12 kHz, +1 to +2 dB (optional)
A gentle boost on the 12 kHz high shelf adds "air" and openness — that sense of a polished, expensive recording. Keep it small. Skip it entirely on a recording that's already bright or a little hissy, because a high-shelf boost lifts hiss right along with the air.
8. A/B with Flatten, then export
Press Flatten to reset every band to 0 dB instantly. This is your A/B test: flatten, listen to the raw voice, then bring your settings back and hear the difference. If flat sounds better, your moves were too aggressive — dial them back. Reset File starts over with a new recording.
When it sounds right, choose your export format and press Export. WAV is the lossless default and the best choice if this file feeds into more editing; MP3 and Opus are also available for a smaller finished file.
Where EQ fits in your podcast chain
EQ is one link in a chain, and order matters. A reliable spoken-word sequence runs:
- Clean it up — reduce hum, hiss, and background noise with the noise reducer first.
- EQ — shape the tone, which is this guide.
- Compress — even out the loud and quiet moments so the voice stays present. See how to use a compressor.
- Set loudness — match your target for the platform. Podcasts commonly aim for around −16 LUFS for stereo; what LUFS should I master to covers the targets.
Keep in mind that spoken-word EQ isn't the same as EQ for a sung vocal. Speech prioritizes intelligibility above all; a sung vocal has to sit inside a busy instrumental. If you're working on singing rather than talking, how to EQ a vocal approaches the same six bands with a different goal.
Troubleshooting
The voice sounds thin after cutting the lows. You over-cut 80 Hz, 240 Hz, or both. The low end carries a voice's weight and authority, especially a deep one. Ease the cuts back until it sounds clean and grounded — a thin voice tires listeners faster than a slightly warm one.
It's still muddy no matter how much I cut 240 Hz. The problem may be at the source: heavy proximity effect from sitting too close to the mic pushes more low-mid energy than EQ should have to remove alone. Back off the mic a few inches next time; for this file, a slightly deeper 240 Hz cut plus a touch off 80 Hz usually gets there.
The esses are still sharp after cutting 6 kHz. A graphic EQ can only tame sibilance broadly — it can't isolate one ess frequency. Cut a little more at 6 kHz, but if it's still piercing, you've hit the limit of what this tool does well, and a dedicated de-esser is the honest next step.
It sounds worse than the original. Almost always too much boosting. Boosts stack — a few bands pushed up at once add real level and can clip the output. Press Flatten, rebuild with smaller moves favoring cuts over boosts, and re-check the level before you export.
Frequently asked questions
What EQ settings should I use for a podcast voice? Start here, then adjust by ear: cut 80 Hz by −3 to −6 dB for rumble, cut 240 Hz by −2 to −3 dB only if it sounds muddy, leave 750 Hz flat, boost 2.2 kHz by +1 to +3 dB for presence, cut 6 kHz by −1 to −3 dB if the esses are harsh, and add +1 to +2 dB at 12 kHz for air. Every voice and mic differs, so treat these as a launch point.
Can I EQ a podcast voice for free online? Yes. Vocal Cut's equalizer is a free 6-band graphic EQ that runs entirely in your browser — no account, no software, no upload, and no sign-up limits. Your audio never leaves your device; the processing happens on your own machine. Load an MP3, WAV, FLAC, or M4A, adjust the six sliders while previewing live, and export a WAV, MP3, or Opus file.
Should I EQ or compress my voice first? EQ first, then compress. EQ shapes the tone and removes problem frequencies; compressing afterward evens out the dynamics of a voice that already sounds right. Compressing first would let the compressor react to rumble or mud you were about to remove, so you'd be flattening the dynamics of a signal you hadn't cleaned up yet.
Why does my voice sound muddy on the mic? Usually the proximity effect. Directional mics boost the low end when you're close, so sitting right on the mic adds a boomy, muddy weight around 240 Hz. Cut that band a couple of dB, and back off the mic slightly when you record. Persistent low-frequency buildup below the voice, near 80 Hz, adds to the same muddy impression.