Measure
The tool applies the BS.1770 K-weighting filter, which models how the ear responds to frequencies, and averages the loudness over 400 ms blocks. Quiet passages below the gate are ignored, as in EBU R128.
Bring a song or podcast to a LUFS target and download it — measured and processed in your browser.
Drop a song or podcast here
or click to choose from your computer
MP3 · WAV · FLAC · AAC · M4A · OGG
Working…
Loudness Normalizer sets a track to a target loudness measured in LUFS. Pick a target, check the result on the readouts, then listen to Original and Normalized one after the other. The gain is applied to the whole file in one step, so the sound stays the same and only the level changes.
Peak meters are not enough for this job. Two tracks can hit the same peak and still sound very different in volume, because loud, dense mixes feel louder than sparse ones. LUFS follows perceived loudness, so it is the number streaming platforms and broadcasters use.
Everything runs in your browser. The file is decoded, measured and rendered on your device, and the export is created locally, so there is no upload queue and no server step.
The tool applies the BS.1770 K-weighting filter, which models how the ear responds to frequencies, and averages the loudness over 400 ms blocks. Quiet passages below the gate are ignored, as in EBU R128.
The gain is the target minus the measured loudness. A track at -18 LUFS with a -14 target gets +4 dB. The gain is applied to every sample with the same factor.
If the gain would push the highest peak above -1 dBFS, the gain is reduced to that ceiling. The readout tells you when this happens, so you know the result is below the target.
The target is the loudness you want the whole file to end up at. Pick the one closest to where the file will be played, then compare the Original and Normalized readouts before you export.
If a clip is very quiet, a large gain can raise the noise floor too. Clean hiss first with the AI Noise Reducer, then normalize.
LUFS (Loudness Units relative to Full Scale) is the measure of perceived loudness defined in ITU-R BS.1770. Unlike peak level, it follows how loud a track sounds to a listener, so two tracks at the same LUFS value sound about equally loud.
-14 LUFS is the level most streaming services use as a reference. -16 LUFS is the Apple Music Sound Check level and a common choice for podcasts. -23 LUFS is the EBU R128 broadcast standard for TV and radio.
Either. A track quieter than the target gets a positive gain and becomes louder. A track louder than the target gets a negative gain and becomes quieter. The whole file gets one gain change, with no compression or limiting.
It stops the gain once the highest sample peak reaches -1 dBFS, so the result never distorts. Dynamic tracks can still hit the target exactly. Heavily compressed masters may end up below the target, which is the safe outcome. Turn it off only if you accept some clipping.
The gain is one multiplication on the decoded audio at the file's own sample rate. WAV is exported at 16 bit, so the only loss is rounding to 16 bit. MP3 adds the usual lossy encoding. Choose WAV if you plan to edit the file again.
No. Decoding, measuring and applying the gain all run on your device. Nothing is sent to Sound Booster servers. The MP3 encoder is loaded once from a public CDN, the first time you export an MP3.
Some players send a mono file to both speakers, which makes it sound louder than the measurement. This tool measures the file as it is, using the BS.1770 weighting for a single channel.
Check, boost or clean up the track after you normalize it.