LUFS Meter Online — Private True Peak Check

For one streaming master, aim for −13.5…−13.7 LUFS integrated and −1.2…−1.1 dBTP True Peak.

That is the loud end of the streaming-safe zone: slightly hotter than the common −14 LUFS reference, with a little more true-peak margin than −1 dBTP for lossy codecs. The page demo track is more conservative on peak, and that is fine too. If you want the simplest public checkpoint, use −14 LUFS / −1 dBTP.

Do not master "for a distributor". Master the audio file. A distributor delivers that file to streaming services. The streaming services play it to listeners and apply loudness normalization.

CD and Bandcamp are different: they do not loudness-normalize playback. For those, LUFS is an artistic choice; the practical safety check is True Peak at or below 0 dBTP, ideally at or below −0.3 dBTP.

The chain is simple:

Distributor → streaming service → listener.

Distributors such as DistroKid, CD Baby, TuneCore, Amuse, Feiyr, RouteNote, Horus Music and Ditto Music do not play music to listeners. They deliver your file and metadata to streaming services.

Streaming services such as Spotify, Apple Music, YouTube, Amazon Music, Tidal, Deezer and SoundCloud host, transcode, play and normalize the track. When people talk about Spotify LUFS or YouTube LUFS, they are talking about playback behavior inside the streaming service, not something the distributor does.

The distributor-safe target for all of the distributors above converges on −14 LUFS integrated / −1 dBTP True Peak. Treat that as a compatibility check, not as a creative rule.

Integrated LUFS is the gated, K-weighted whole-track loudness calculated by ITU-R BS.1770. It is the headline number for how loud the master is, and the number streaming services compare against their loudness references.

True Peak (dBTP) is the highest reconstructed peak after oversampling, including peaks that can occur between digital samples. Use it to set the limiter ceiling and avoid clipping after playback conversion or lossy encoding.

Loudness Range (LRA) shows how much the loudness moves across the track, in LU. Low LRA feels dense and constant; high LRA gives more contrast between quieter and louder sections.

Short-term max is the loudest 3-second window. It helps you find the section that feels loudest over a phrase.

Momentary max is the loudest 400-ms window. It catches short bursts, hits and transitions.

PLR means Peak-to-Loudness Ratio: True Peak − Integrated LUFS. A higher PLR usually means more transient space; a lower PLR usually means harder limiting.

Most streaming services normalize playback to a loudness reference so tracks do not jump wildly in level. A master louder than the reference is turned down. A quieter master is handled differently depending on the service.

YouTube is down-only: it turns loud tracks down and leaves quiet tracks quiet. Spotify can raise quieter tracks, but only as far as peak headroom allows; its limiter is mainly relevant in Loud mode. So "only turns down" is too simple. The practical lesson is clearer: do not crush a master far above the reference, and do not leave it far below the reference unless the music truly wants that.

Once you have Integrated LUFS, do not use dBFS or RMS as a substitute. They answer different questions.

dBFS sample peak tells you how close the loudest stored sample is to the digital ceiling, 0 dBFS. It is a clipping warning, not a loudness reading. A track can peak near 0 dBFS and still sound quiet, or peak lower and sound loud.

True Peak is the safer peak reading because it checks reconstructed peaks between samples. Use it for delivery ceilings. It does not tell you whether the song feels loud enough; it tells you whether the file has enough peak headroom.

RMS is older average energy. It can be useful when comparing processing, but it does not include the BS.1770 loudness model that streaming normalization uses.

The reason −1 dBTP matters is codec headroom. AAC, MP3 and Ogg can push intersample peaks upward by roughly 0.3–1 dB during encoding. A master that looks safe near 0 dBFS can clip after conversion. For streaming, set the limiter ceiling around −1 dBTP; for Spotify Loud or Amazon-style stricter targets, −2 dBTP may be safer.

No. −14 LUFS is a playback-normalization reference, not an artistic rule.

Master for the song first. Dense EDM, metal, pop and hip-hop often want more density before normalization. Indie, jazz, classical and acoustic music often need more space and can sit quieter. The goal is not to force every song to the same number; the goal is to understand what the streaming service will do to it.

A very loud −8 LUFS master will usually be turned down, so the extra limiting no longer buys playback loudness. You keep the reduced dynamics and lose the level advantage. A very quiet master may be right for the music, but on YouTube it will stay quiet, and on Spotify it may not be raised fully if there is not enough peak headroom.

If you need to move a finished file toward a target, use the browser mastering tool.

This meter uses a Web Audio AudioWorklet implementation of ITU-R BS.1770-5, with EBU Tech 3341/3342 behavior for loudness metering and LRA. It stays close to reference meters such as ffmpeg ebur128. Browser decoding can still create small differences, especially with lossy files, so use a dedicated offline meter for formal QC, label delivery or broadcast sign-off.

Your audio stays on your device. The file is decoded and measured locally in the browser and is never uploaded. There is no account and no server-side audio processing. The page sends only anonymous, cookieless usage stats, never the audio. After the page loads, the meter can keep working offline.

This free tool is made by Darwin’s Cat, a band. If it saved you time, listen to our music or buy us a beer.