Which band should I move to make a voice clearer?

The middle one, up by three or four decibels, and the bass down by a similar amount. That is what the Speech preset does. Most of the muddiness in a voice recording is low-frequency room noise that carries no words at all.

Why did the whole track get quieter?

Because a boost pushed the peaks past the maximum. Rather than clip them, everything was scaled down to fit, which preserves the shape of the sound. You can restore the level afterwards with the volume tool if there is room.

Can this remove background hiss or hum?

Only crudely. Cutting the treble reduces hiss along with everything else up there, and cutting the bass reduces mains hum along with the warmth. A real noise reduction learns the noise and subtracts it, which three tone controls cannot do.

Is this the same as the equaliser in my music player?

It uses the same kind of filter, but with one important difference: your player changes what you hear while leaving the file alone, whereas this writes a new file with the change baked in. Keep the original if you might want it back.

Equalizer

Adjust bass, middle and treble, or pick a ready-made curve.

Everything runs inside your browser. Your files never leave your device.

Drop your file here

.mp3, .wav, .m4a, .aac, .ogg, .oga, .opus, .flac, .webm, .weba · they stay on this device

Ask about Equalizer

Questions about what this tool does, which option to pick, or what it can and cannot handle.

This is the one part of the site that uses a server. The question you type here is sent to an AI provider to be answered — your files and whatever you put in the tool above are not, and the assistant cannot see them. Answers are generated and can be wrong; the tool itself is not guessing.

About the Equalizer

An audio equaliser turns some frequencies up and others down. Three controls — bass, middle and treble — cover most of what anyone actually needs, and they map onto the three things people complain about: too boomy, too muffled, too harsh.

Bass is a shelf below 200 Hz, which means everything under that gets moved together rather than one narrow band being picked out. Treble is the same shape above 4 kHz. The middle control is a peak around 1 kHz, which is where the intelligibility of speech lives — consonants, the difference between "cat" and "cap".

The filters are the standard cookbook biquads, the same formulas the browser's own equaliser node uses. That is deliberate: an equaliser built out of the same maths as everybody else's sounds like everybody else's, and its behaviour can be checked with a test rather than argued about. Boosting the peaking filter by six decibels really does produce six decibels at its centre frequency, and that figure is measured by an automated check rather than trusted.

The presets are starting points rather than magic. "Speech" lifts the middle and thins the low end, which is what makes a recorded voice easier to follow on a phone speaker. "Telephone" cuts hard at both ends for the effect. "Bass boost" does the obvious thing, and the obvious warning applies: boosting bass on a small speaker mostly produces distortion, because the speaker cannot move enough air to reproduce it.

Boosting anything makes the file louder, and a file that was already close to maximum will run out of headroom. Rather than let the peaks clip, the whole track is turned down slightly when that happens, and the result panel tells you it did. That is the right trade — a quiet correct file beats a loud broken one — but if you see that message often, cut the bands you dislike instead of boosting the ones you want.

How to use it

1Drop in the recording.
2Pick a preset, or choose the custom option and set the three bands yourself.
3Boost sparingly and cut freely — cutting what you dislike causes fewer problems than boosting what you want.
4Run it, listen on the speakers you actually use, and adjust.

Questions

Which band should I move to make a voice clearer?: The middle one, up by three or four decibels, and the bass down by a similar amount. That is what the Speech preset does. Most of the muddiness in a voice recording is low-frequency room noise that carries no words at all.
Why did the whole track get quieter?: Because a boost pushed the peaks past the maximum. Rather than clip them, everything was scaled down to fit, which preserves the shape of the sound. You can restore the level afterwards with the volume tool if there is room.
Can this remove background hiss or hum?: Only crudely. Cutting the treble reduces hiss along with everything else up there, and cutting the bass reduces mains hum along with the warmth. A real noise reduction learns the noise and subtracts it, which three tone controls cannot do.
Is this the same as the equaliser in my music player?: It uses the same kind of filter, but with one important difference: your player changes what you hear while leaving the file alone, whereas this writes a new file with the change baked in. Keep the original if you might want it back.