Browse files
or drag and drop it anywhere in this area
No file size limit · minimum 10 seconds · 30+ seconds recommended for analysis
Files stay on your device — no uploads · no account · private processing in this browser tab.
Target
使い方
Loudness here is measured the way streaming services measure it: the ITU-R BS.1770 algorithm from EBU R128. Your audio is filtered with a K-weighting curve that mimics how we perceive loudness, split into overlapping 400 ms blocks, then averaged with two gates — one absolute at -70 LUFS and one 10 LU below the ungated mean — so silence and fade-outs do not drag the reading down. The result is a single integrated LUFS value. Normalising simply applies the constant gain needed to move that value to your target, then either backs the gain off or soft-limits so the oversampled true peak stays under your ceiling. Constant gain means the mix's internal dynamics are preserved: nothing is compressed unless you choose the limiting mode.
- 01
Upload
Pick a file or drop it in — decoding happens on your device.
- 02
Process
The spectral engine renders the result in this browser tab.
- 03
Download
Preview and save the output as a lossless WAV file.
Why -14 LUFS matters
Streaming platforms play everything at a similar perceived level. Master louder than the target and the platform turns your track down anyway — you keep the squashed dynamics and lose nothing but headroom. Master quieter and some services leave it quiet rather than boosting it.
Leave the ceiling at -1 dBTP. Lossy encoders such as MP3 and AAC can overshoot the original sample peaks, and that margin prevents inter-sample clipping on playback.
Podcasts and spoken word usually target -16 LUFS mono/-16 to -19 LUFS stereo, while broadcast delivery in Europe uses -23 LUFS. Pick the target for where the audio will actually be published.
Spectral engine
Frequency-domain processing tuned for voice and instrument content.
Fast
Rendered locally in this tab — no upload waiting or queue.
Free
Unlimited tracks, no account and no usage limits.
Analysis and separation results are estimates produced by signal processing, not definitive verdicts. Quality depends on the source recording.
Loudness normalisation and LUFS explained
Peak normalisation only looks at the single loudest sample, which is why two files that both peak at 0 dBFS can still sound wildly different in level. Loudness normalisation instead measures perceived loudness over time using the ITU-R BS.1770 / EBU R128 method — K-weighting plus gated integration — and reports the result in LUFS.
Streaming platforms play everything back at a common loudness: roughly -14 LUFS on Spotify and YouTube and -16 LUFS on Apple Music, with anything louder simply turned down. Normalising to the target yourself means you hear what listeners will hear, rather than being surprised by a platform's gain reduction.
True peak is measured alongside the integrated value, because raising gain can push inter-sample peaks past 0 dBTP and cause distortion in lossy codecs. You can either protect dynamics by reducing gain until the peak is safe, or apply gentle limiting to keep the loudness target.
How to use the volume normalizer
- 01Load a file to measure its integrated LUFS, loudness range and true peak.
- 02Pick a target — Spotify, Apple Music, YouTube, broadcast, or a custom value.
- 03Choose whether to protect dynamics or limit to hit the target exactly.
- 04Export the normalised file.
Frequently asked questions
-14 LUFS integrated suits Spotify and YouTube, -16 LUFS suits Apple Music and podcasts, and -23 LUFS is the broadcast standard in Europe.
dBFS describes sample amplitude at an instant. LUFS describes perceived loudness across the whole programme, weighted for how human hearing responds to frequency.
Reconstructing the analogue waveform between samples can overshoot the highest sample value. True-peak metering estimates those inter-sample peaks, which is what codecs actually have to reproduce.