Free LUFS meter
Measure a mix or master the way streaming services and broadcasters do: integrated loudness, true peak, loudness range and a loudness chart. It runs in your browser, so the file never leaves your device.
Your file never leaves your device. Everything is measured in this browser tab.
How to use it
- Choose a file or drop it on the panel. WAV, AIFF, FLAC, MP3, M4A and OGG work, up to 1.5 GB.
- Wait a moment. WAV and AIFF are read straight from your disk in small pieces, so even a long session file works without filling your memory. The bar shows progress.
- Read the numbers. The big one is integrated loudness. The six beside it tell you about peaks and dynamics.
- Listen along. Press play and the red line follows the chart. Click the chart to jump to a loud chorus or a quiet intro.
What the numbers mean
- Integrated loudness (LUFS)
- The loudness of the whole file, start to finish, with silence and very quiet passages left out (gating). Streaming services compare this one number with their reference to decide how much to turn your track down.
- Short-term (3 s)
- Loudness over the last three seconds. Its maximum is close to how loud the loudest chorus feels.
- Momentary (400 ms)
- Loudness over the last 0.4 seconds. Its maximum is the single biggest hit.
- Loudness range (LU)
- How far apart the quiet and loud sections sit, from the 10th to the 95th percentile of short-term loudness. Heavily limited pop is often around 3 to 6 LU; film and classical can exceed 15.
- True peak (dBTP)
- The highest level the waveform reaches between samples, found by oversampling 4x. A DAC or an MP3/AAC encoder can go over 0 dBFS even when no sample does, which is why delivery specs ask for −1 dBTP.
- Sample peak (dBFS)
- The highest sample value in the file. It is never higher than the true peak.
- Peak to loudness ratio (PLR)
- True peak minus integrated loudness. A rough measure of punch: around 8 dB is very dense, 12 dB and above leaves the transients room to breathe.
How loud should a master be?
There is no single target. Streaming services turn loud tracks down to their reference level, so a −8 LUFS master and a −14 LUFS master end up at a similar volume on Spotify. What changes is how the louder one sounds after limiting.
Most commercial pop, hip-hop and electronic releases measure louder than −14 LUFS, often between −10 and −6. Acoustic, jazz and classical records are usually quieter. Spotify suggests −14 LUFS as a mastering target, but it is a suggestion: a louder master is simply turned down, never rejected.
Two numbers matter for delivery: keep the true peak at or below −1 dBTP (−2 dBTP if the master is louder than −14 LUFS, per Spotify), and make sure the loudness you chose still sounds good after the service turns it down. The table shows where each service plays music.
Streaming loudness references
| Service | Plays at | Turns quieter tracks up? | Source |
|---|---|---|---|
| Spotify (Normal)Premium listeners can pick Loud (−11 LUFS, with a limiter) or Quiet (−19 LUFS). Albums are normalised as a whole; the web player does not normalise. | −14 LUFS | Yes, until the true peak reaches −1 dBTP | Spotify for Artists: Loudness normalization, 2026-11-07official |
| YouTubeYouTube publishes no figure; measured from “Stats for nerds”. | −14 LUFS | No | MeterPlugs: YouTube changes loudness reference to −14 LUFS, 2019-09-18measured |
| Apple Music (Sound Check)Apple’s own mastering brief describes Sound Check but gives no level. | −16 LUFS | Not documented | MeterPlugs: Apple switch to LUFS, 2022-03-23measured |
| Amazon Music | −14 LUFS | Not documented | MeterPlugs: Loudness Penalty adds Amazon Music and Deezer, 2019-10-15measured |
| TIDALAlbum normalisation. | −14 LUFS | No | MeterPlugs: TIDAL adds loudness normalization, 2016-10-07measured |
| Deezer | −15 LUFS | Not documented | Deezer Community: reply from the Deezer team, 2019-12-18measured |
| Broadcast TV and radio (EBU R 128) | −23 LUFS | Delivery target: −23 LUFS, true peak no higher than −1 dBTP | EBU R 128 v5.0, 2023-11official |
Checked 7 November 2026. Only Spotify and the EBU publish their figures; the others come from first-hand measurements or statements by the service’s staff, dated as shown. Services change these settings, so treat them as a guide.
How this meter measures
The meter follows ITU-R BS.1770-4 and EBU R 128. Each channel is K-weighted at the file’s own sample rate, 400 ms blocks with 75 % overlap are gated at −70 LUFS and then 10 LU below the gated average, and loudness range follows EBU Tech 3342 (short-term values, −20 LU gate, 10th to 95th percentile). True peak is the largest value after 4x oversampling.
It is tested against the EBU’s own compliance signals (Tech 3341 and 3342, November 2023): a stereo 1 kHz sine at −23 dBFS reads −23.0 LUFS, every loudness range case lands on its expected value, and the true-peak cases fall within 0.03 dB of the reference.
A mono file counts as one channel, as the standard says, so it reads 3 dB lower than the same audio copied to both channels of a stereo file. MP3 and AAC files are decoded by your browser first; the result can differ from the WAV by a tenth of a dB or so.
Want the exact turn-down on each service? Use the loudness penalty checker.
Questions
LUFS (Loudness Units relative to Full Scale) measures how loud audio sounds rather than how high its peaks go. It weights the frequencies our ears are most sensitive to and averages over time. One LU is the same size as one dB.
On its Normal setting Spotify plays music at −14 LUFS and turns louder tracks down to that. Spotify suggests −14 LUFS as a target, but nothing stops a louder master: it is just played quieter, and most released music is louder. Choose the loudness that sounds best for the song and keep the true peak at −1 dBTP or lower, or −2 dBTP if the master is louder than −14 LUFS.
Loudness is measured downwards from digital full scale, where 0 is the ceiling. Real music always sits below it: −6 LUFS is extremely loud, −23 LUFS is the broadcast level.
Sample peak is the highest stored sample. True peak estimates the highest point of the reconstructed waveform between samples, which is what a converter or a lossy encoder actually produces. True peak is always equal to or higher than sample peak.
Common reasons: the other meter treats a mono file as two channels, it was reset partway through, it is measuring a different export (for example with the limiter bypassed), or it measures a lossy copy. Meters that follow BS.1770-4 agree within about 0.1 LU on the same file.
No. The file is opened by your browser and measured on your computer or phone. Nothing is sent to our servers, and closing the tab leaves nothing behind.
Up to 1.5 GB. WAV and AIFF are read from disk in pieces, so long files are fine. Compressed formats (MP3, FLAC, M4A, OGG) have to be decoded in memory first, so they are limited to 300 MB; export a WAV for anything bigger.
Not yet. The meter reads the whole file. The chart shows short-term loudness over time, so you can still see how loud each section is.
More free tools
- Loudness penalty checker: How much Spotify, YouTube, Apple Music, Amazon Music, TIDAL and Deezer turn your track down, and a preview at that level.
- BPM and key finder: Tempo, musical key and Camelot code for any song file. No upload and no daily limit.
- Mastering for Spotify: master to Spotify’s loudness target.
- Mastering for YouTube: master to YouTube’s loudness target.
- Make an MP3 louder: raise loudness without clipping.
Numbers fine, sound not there yet?
Upload the mix and hear it mastered in about a minute. The preview is free and needs no account.