The numbers streaming services actually use
Spotify, Apple Music, YouTube and broadcasters don't judge loudness by peak level. They measure LUFS, a loudness unit weighted to match how we hear, and turn every track up or down to their target. Mastering far louder than the target doesn't make your track louder on those platforms; it just gets turned down, with less punch left.
WAVR's report gives you the same measurements a mastering engineer reads off a meter, plus the exact gain change needed to land on each common target.
What the report shows
- Integrated loudness (LUFS): the whole-file average, gated so silence doesn't drag it down.
- Short-term and momentary max: the loudest 3-second and 400 ms windows.
- Loudness range (LRA): how much the loudness moves, in LU. Pop masters are often 4–8 LU; film and classical much wider.
- True peak (dBTP): the highest point of the reconstructed waveform, measured with 4× oversampling.
- Crest factor, RMS, clipping and DC offset: quick signs of over-limiting, clipped samples or a recording fault.
- Phase correlation and mono loss: whether the mix falls apart on a mono phone speaker or club system.
Common targets
| Destination | Integrated | True peak |
|---|---|---|
| Spotify (Normal) | −14 LUFS | −1 dBTP, or −2 if louder than −14 |
| Apple Podcasts | −16 LUFS ±1 | −1 dBTP |
| EBU R128 broadcast | −23 LUFS | −1 dBTP |
| US TV (ATSC A/85) | −24 LKFS | −2 dBTP |
More detail, including what to do when your track measures louder or quieter than the target, is in What is LUFS and which target should I use?
Compare before and after
Made a change in the editor? Switch the preview between A (original) and B (edited) and turn on Match loudness. WAVR measures both and plays them at the same LUFS, so you're judging the edit, not the volume.
Loudness meter questions
How is loudness measured?
WAVR applies the K-weighting filter from ITU-R BS.1770, measures 400 ms blocks every 100 ms, and gates them at −70 LUFS and −10 LU as EBU R128 specifies. True peak uses 4× oversampling.
Why is true peak higher than sample peak?
The waveform between samples can overshoot the samples themselves. Converters and lossy encoders reconstruct that curve, so a file peaking at 0 dBFS can clip on playback. Keep true peak at or below −1 dBTP for streaming.
Will WAVR's numbers match my plug-in meter?
They should agree closely on integrated loudness, since both follow the same standard. True peak and LRA can differ by a few tenths between meters because the standards allow different oversampling filters and rounding.
Is my audio uploaded?
No. The measurement runs in a background thread in your browser. Unreleased masters stay on your device.