Measure your mix, then apply the gain to hit −16 LUFS
Drop an episode or a master, read integrated LUFS and true peak, then apply one clean gain change to the target — the file exports as WAV and the true peak never goes over the ceiling. Nothing uploaded.
Free · no account · runs in your browser — nothing is uploaded; your audio never leaves your device.
Every platform at once
| Destination | Target | What happens to this file |
|---|
Loudness over time
Short-term loudness (3-second window). The marker is the loudest moment; click the graph to jump the player there.
A/B against a reference
Drop a reference track (a record you want to sit next to). Both are measured the same way and compared.
| Integrated | True peak | |
|---|---|---|
| Yours | – | – |
| Reference | – | – |
Levels are the checklist. The music bed is the craft.
PodDJ in the studio builds the whole sound of your show — beat-aligned music sections, ducking under your voice, broadcast loudness on export. Sign up: 3 full packs free.
Why every platform measures LUFS, not peaks
Loudness normalization ended the loudness war: platforms measure how loud your audio feels (LUFS, gated per ITU BS.1770) and turn everything to the same shelf. Master hotter than the shelf and they simply turn you down — you paid in crushed dynamics and got nothing. Deliver on target with true peak under −1 dBTP and your episode sounds exactly as you mixed it, everywhere.
The meter only prints a number it can defend. Integrated loudness is built from gated 400 ms blocks, and three of those need only about 0.6 s, so a file under one second cannot support a reading at all — we keep a deliberate margin above that minimum rather than measuring on the edge of it. Silence has nothing above the −70 LUFS absolute gate, and a source that is already clipped still gets measured — but its true peak is reported as a floor, because the tops of those waveforms are gone. Changing the level is one gain multiplication: no limiter, no compression, no fades, and never a sample count change.
How to check the LUFS of a song or podcast
- Load the fileDrop a master or an episode on the meter. It stays on your device.
- Read the numbersSee integrated loudness (LUFS), true peak and the loudest moment for the whole file.
- Pick a platformChoose Spotify, Apple Music, YouTube, a podcast target or another destination. The verdict says what that platform does to your file; the table below shows all of them. Targets move, so check your platform's current spec.
- Normalize or comparePress Normalize to target for a WAV at that loudness, with true peak kept under the ceiling. Or drop a reference track to see how far apart you are.
Frequently asked questions
What LUFS should I master to for Spotify?
Spotify's reference is about −14 LUFS integrated with true peak at or below −1 dBTP. Spotify turns louder files down to that level, so a −8 LUFS master plays at the same loudness as a −14 one, with less dynamic range. The meter shows exactly how many dB Spotify will turn your file down or up.
How do I check the LUFS of a song?
Drop the file on the meter. It reads integrated loudness the way platforms do (ITU BS.1770-4, K-weighted and gated), plus true peak and the loudest moment. Then pick a platform from the list to see the verdict, or read the table that shows what every platform does to your file.
What LUFS should my podcast be?
−16 LUFS integrated is the widely used spoken-word target (Apple's guidance sits there), and −14 is common for music-heavy shows and streaming platforms. These are conventions, not laws — the targets here are editable, and your distributor's current spec wins.
What's the difference between LUFS and peak level?
Peak is the tallest single sample; LUFS is perceived loudness over time, gated the way ITU BS.1770 specifies — which is how every platform actually measures. A file can peak at −1 dB and still be whisper-quiet in LUFS. You need both numbers: LUFS on target, true peak under −1 dBTP.
What does the normalize button do exactly?
It applies one clean gain change so your integrated LUFS lands on the target — no compression, no limiting, character untouched. If that gain would push the true peak over −1 dBTP, the tool tells you instead of quietly clipping.
Why does true peak matter if I'm under 0 dB?
Lossy encoders (MP3, AAC) can overshoot between samples — a file that peaks at −0.1 dBFS can clip after encoding. That's why platforms ask for −1 dBTP headroom, measured with oversampling, which is what this meter does.