Skip to main content
SafeStudio

( Free · in the browser · no sign-up )

Audio compressor onlinefor controlled dynamics

Reduce peaks above a chosen threshold with adjustable ratio, attack, release, and knee, then compare and export locally in your browser.

Or drop an audio or video file anywhere on this page. Compatibility depends on the file codecs and device: MP3, WAV, M4A, AAC, FLAC, OGG, WebM, MP4, MOV, MKV, and more.

The file you choose is processed in this tab; the page still loads site resources and consented analytics. Read the Privacy Policy.

  • 01Real dynamics processing
  • 02Adjustable timing
  • 03Local A/B preview

( What it does )

Dynamic-range compression, not file-size compression

This tool uses the browser's DynamicsCompressorNode to turn down signal portions that cross the threshold. Ratio controls how strongly those portions are reduced, while attack, release, and knee shape how the gain reduction begins and recovers.

  • 01

    Practical starting preset

    The editor opens at −18 dB threshold, 4:1 ratio, 10 ms attack, 250 ms release, and a 12 dB knee.

  • 02

    Preview equals export

    The same compressor settings feed real-time playback and offline WAV or compressed export.

  • 03

    A/B comparison

    Bypass the compressor without losing the current settings, then switch back to judge the dynamics change.

( How it works )

How to set an audio compressor

Begin with a representative passage and change one parameter at a time. Compression is easier to judge when the preview includes both ordinary and loud moments.

  1. 1

    Choose a representative section

    Play a passage containing normal level and the peaks you want to control. Start with the practical preset rather than extreme settings.

  2. 2

    Set amount before timing

    Adjust threshold and ratio to decide how much is reduced, then refine attack, release, and knee for the way compression enters and recovers.

  3. 3

    Use A/B and export deliberately

    Bypass the effect to compare dynamics, not just loudness. This tool has no automatic makeup gain, so adjust the final level separately when needed.

( Limits )

What the compressor does not promise

  • Dynamics compression does not reduce the file size; use the bitrate converter for that task.
  • The tool does not measure or normalize integrated loudness in LUFS.
  • Compression can make transients dull or increase audible noise when settings are too aggressive.
  • There is no automatic makeup gain, limiter, side-chain input, or multiband processing.

( Uses )

Useful starting points for speech and music

  • Control occasional voice peaks before manually adjusting the final level.
  • Tighten the dynamic range of a rough podcast or narration preview.
  • Soften strong musical transients with a moderate ratio while comparing against the original.

( Concepts )

Compressor controls in practical terms

These controls interact. Moderate changes make it easier to hear which parameter solved the original problem.

Threshold and ratio

Where compression starts and how strongly signal above that point is reduced.

Lower thresholds and higher ratios produce more gain reduction.

Attack

How quickly gain reduction responds after the signal crosses the threshold.

A fast attack controls transients more firmly; a slower one lets more initial impact pass.

Release and knee

How quickly reduction recovers and how gradually compression begins around the threshold.

Release that is too short may pump; a softer knee usually makes entry more gradual.

( Other tools )

Choose the tool that matches the task

( Where your audio goes )

Dynamics processing stays local

The decoded samples are compressed, previewed, and rendered in browser memory. The application server does not receive the selected audio for this effect. Site resources and consented analytics may still make network requests.

( Questions? Answered )

Audio-compressor questions

Distinguish dynamics control from normalization, limiting, and file-size compression.

What does a compressor do to audio?

It turns down the parts that are louder than a level you choose, which brings the loud and quiet parts of a recording closer together. The result is a track that stays audible without the listener reaching for the volume — not a track that is simply louder.

How do I compress a voice recording?

Start with a threshold just under the level of normal speech, a ratio around 3:1, a fast attack and a release near a quarter of a second. Watch how much the loudest words are being pulled down; a few decibels is usually enough. Then set the final level separately.

Does an audio compressor make the file smaller?

No, and this is the most common mix-up on this page. Audio compression reduces dynamic range inside the sound; file compression reduces bytes. For a smaller MP3, use the bitrate converter instead.

Does compression automatically make audio louder?

Not here. This tool turns down what is above the threshold and adds no automatic makeup gain, so the output is usually quieter than the input. Raise it afterwards with the volume tool or by normalising.

What do threshold, ratio, attack, release and knee mean?

Threshold is the level above which it acts. Ratio is how hard it pulls down past that point — 4:1 means four decibels in, one out. Attack is how quickly it reacts, release how quickly it stops, and knee is how abruptly the ratio arrives.

What settings work for a podcast?

A gentle ratio of 2:1 to 3:1 with a threshold that only catches the loudest phrases keeps speech natural. Heavy settings on a quiet room also lift the noise floor, which is the usual reason a compressed voice starts sounding like a telephone.

What causes pumping or breathing?

Very strong gain reduction, or a release that does not match the material, makes the background level rise and fall audibly between words. Use a milder ratio, a higher threshold, or a longer release.

Should I normalise before or after compression?

Compress first to shape the dynamics, then normalise to set the delivery level. Doing it the other way means the compressor acts on a level you then change, and leaves less headroom between stages.

Is this a limiter?

No. A limiter is a compressor with a very high ratio used as a ceiling, and this tool does not guarantee one. If you need a hard maximum, normalise afterwards and check the peak.

Can compression repair a recording that is too quiet?

Not on its own. It reduces the distance between loud and quiet, which also brings up room noise and breaths. A quiet recording usually wants gain or normalisation first, and compression only if the dynamics are actually uneven.

Which formats can I open and export?

Common audio and video containers are accepted; the editor tries native decoding first and falls back to a local FFmpeg build. Exports are MP3 from 64 to 320 kbps, 16-bit PCM WAV, M4A/AAC at 192 kbps, and FLAC.

Is my file uploaded to a server?

No. The audio is decoded and compressed in this browser tab and is not sent to the application server. The page itself still loads site resources and consented analytics.

( Field notes )

Learn the technique behind the tool

Guides written and reviewed by the person who builds the editor, including the real limits of each feature.

See all articles