Audio Compressor

What it does

An audio compressor reduces the dynamic range of a signal: loud peaks are attenuated while quieter parts can be raised with makeup gain. That keeps levels more consistent for podcasts, demos, and recognition clips that were recorded too unevenly.

Music Identifier’s compressor runs in the browser with the Web Audio API. You set threshold, ratio, knee, attack, release, and makeup gain, preview the result, and download a WAV. Nothing is uploaded to a server.

This is a practical dynamics tool, not a full mastering chain — use moderate settings first and listen for pumping or dull transients.

Reading the controls

Threshold sets the level where compression begins; anything above it is turned down by the amount your ratio specifies (4:1 means 4 dB over the threshold becomes 1 dB over it). Knee softens the transition around the threshold instead of applying it abruptly. Attack decides how quickly the compressor reacts once a peak crosses the threshold, and release decides how quickly it lets go afterward — fast settings catch transients tightly but can sound choppy, while slow settings are smoother but let more through.

A reasonable starting point

For general leveling, a threshold around −20 to −24 dB, a ratio near 3:1 or 4:1, a medium knee, and a release around 150–300 ms is a common, forgiving starting combination. Adjust makeup gain last, once the dynamics feel right, so the final loudness matches what you expect.

Even out level differences afterward with the Audio Normalizer, or shape tone with the Audio Equalizer.

How to use it

Tip: Change one control at a time and preview between changes — threshold and ratio interact, and it is easy to over-compress if you push both aggressively at once.