Audio Compressor
Even out loud and quiet passages with a real dynamic-range compressor, then normalise the peak for consistent playback.
Convert with this tool
or choose a file from your device — it is processed in this tab and never uploaded
Accepted: WAV, MP3, FLAC, M4A/AAC, OGG, Opus
Files stay on your device. There is no upload, no queue and no watermark, so a conversion works even with the network switched off after the page has loaded.
Supported formats & processing
| Accepted input | WAV, MP3, FLAC, M4A/AAC, OGG, Opus |
|---|---|
| Produced output | MP3, WAV, FLAC or M4A with the compression applied |
| Processing | Web Audio API + canvas + native JavaScript encoders |
| Upload required | No — the file is read locally and never transmitted |
How to use this tool
- Add the recordingSpeech, interviews, live music and field recordings all work.
- Set the thresholdStart at −24 dBFS for speech, −18 dBFS for music. Anything above this level gets turned down.
- Choose a ratio2:1 is gentle and transparent, 4:1 is the standard for speech, 8:1 and above approaches limiting.
- Leave peak normalisation onIt guarantees the loudest moment sits just below full scale.
- Convert and read the levelsCompare the peak and RMS values in the result card with the source values.
About this tool
Compression in audio has nothing to do with file size — it is dynamic-range control. A compressor turns down anything above a threshold, then makeup gain brings the overall level back up, so quiet parts become audible without the loud parts clipping. Podcasts, voice-overs, lecture recordings and phone interviews all benefit enormously.
This tool uses the browser's DynamicsCompressorNode with the threshold, ratio, attack and release you select, followed by an optional peak normalisation to −0.5 dBFS. Both the peak and the RMS level before and after are reported, so you can see exactly how much the dynamic range was reduced.
Features
- Real DynamicsCompressorNode with adjustable threshold and ratio
- Attack and release tuned for speech rather than music
- Optional peak normalisation to −0.5 dBFS
- Peak and RMS levels measured before and after compression
- No clipping introduced — the normaliser prevents it
- Output as MP3, WAV, FLAC or M4A without leaving the page
Frequently asked questions
Does this reduce the file size?
No. This is dynamic-range compression, not data compression. To shrink the file, lower the bitrate with the bitrate converter instead.
What is the difference between compression and normalisation?
Normalisation scales the whole file by one factor so the loudest peak hits a target level. Compression changes the relationship between loud and quiet parts, reducing the dynamic range itself. Most spoken-word work needs both, which is why this tool does them together.
Will a high ratio make my audio sound bad?
Very high ratios with a low threshold squash the life out of music and can make breathing audible in speech. Start at 4:1, listen, and only push further if you still have to ride the volume.
Can I undo the compression later?
Not exactly — expansion cannot perfectly restore the original dynamics. Keep the uncompressed source if the file matters.