What LUFS level should my audio be?

It depends on where it goes. Spotify plays music at -14 LUFS, Apple Podcasts recommends about -16 LKFS for episodes, EBU R 128 broadcast uses -23 LUFS and US television -24 LKFS. Pick the target of the place your file will be heard, and keep the true peak under that target's limit, usually -1 dBTP.

What is the difference between LUFS and dBFS?

dBFS measures signal level against digital full scale, sample by sample, and says nothing about how loud a file sounds. LUFS measures perceived loudness over time: the signal is K-weighted, averaged and gated. A peak meter reads dBFS; a loudness meter reads LUFS.

Why is the true peak higher than the sample peak?

The analog waveform rebuilt from the samples can swing above the highest sample, most often in loud, bright material. The meter estimates those in-between peaks by oversampling four times. If the true peak sits above -1 dBTP, a lossy encoder or a converter may clip it.

Does this meter follow EBU R 128?

It measures the way EBU R 128 asks: BS.1770 gating for integrated loudness, EBU Tech 3342 for loudness range and an oversampled true peak, and it reads the EBU Tech 3341 and 3342 test tones within tolerance. It is a file meter, not a certified broadcast instrument, and it does not apply the dialogue gating used for US long-form TV.

Is my audio file uploaded?

No. The file is decoded and measured in your browser and is not uploaded; a file you save with gain applied is written on your device. The page records ordinary analytics events, such as that a measurement finished, not the file, its name or its results.

LUFS Meter

A loudness meter for audio files. Drop a song, podcast episode or voice recording to read its integrated loudness in LUFS, its true peak and loudness range, and see how far it sits from the figures Spotify, Apple Podcasts and broadcasters publish.

Drop an audio file here or click to browse

MP3, M4A, AAC, WAV, OGG, OPUS, FLAC, WEBM or MP4, up to 100 MB and 30 minutes.

The file is measured in your browser and is not uploaded.

How to measure LUFS

The whole file is measured in your browser.

1
Drop the file
MP3, M4A, WAV, FLAC, AAC, OGG, OPUS, WEBM or MP4, up to 100 MB and 30 minutes. Measuring takes a few seconds for a song and longer for a long episode.
2
Read the numbers
Integrated loudness is the headline figure. True peak, sample peak, loudness range and the momentary and short-term maxima sit below it.
3
Compare with the targets
A table shows the gain each published target would need and whether that gain would push the true peak over the limit. Services that publish no figure are marked as such.
4
Optional: apply the gain
Pick a target, keep the true-peak limit on or off, and save a WAV or MP3 with one gain applied. Nothing else about the audio changes.

Need the highest peak at a set level instead of a loudness target? That is the Audio Normalizer. For a level boost past the peak, with a limiter, use the Volume Booster.

What each number means

Integrated loudness (LUFS)
The average loudness of the whole file after K-weighting, a filter that roughly follows how the ear hears, with very quiet stretches gated out. LUFS and LKFS are the same unit; one LU is one decibel of loudness.
True peak (dBTP)
The highest level of the waveform between samples, estimated by oversampling four times. It can be higher than the sample peak, and it is what decides whether a lossy encoder or a converter will clip.
Loudness range (LRA)
How much the loudness moves over time, in LU: the spread between the quieter and louder parts of the file, measured on 3-second windows with the extremes left out. A dense, heavily limited master reads only a few LU; a film with quiet dialogue and loud action reads far more.
Momentary and short-term max
The loudest 400-millisecond and 3-second windows. They show the loudest moment, which can sit well above the integrated figure in a quiet recording with one loud burst.

Published loudness targets

Each figure comes from the service's or standards body's own document. Where none is published, the row says so.

Who	Integrated loudness	True peak	Applies to
Spotify	-14 LUFS	Below -1 dBTP; below -2 dBTP if the master is louder than -14 LUFS	Music, adjusted during playback
Apple Podcasts	About -16 LKFS, ±1 dB	Not above -1 dBFS	Podcast episodes (a recommendation)
EBU R 128	-23 LUFS, ±0.5 LU	At most -1 dBTP	European broadcast programs
ATSC A/85	-24 LKFS	-2 dBTP	US television content delivered without loudness metadata
YouTube	Not published	Not published	The help pages describe Stable volume without a figure
Apple Music	Not published	At least 1 dB of headroom	Sound Check is described without a LUFS figure
Amazon Music	Not published	Not published	A Loudness Normalization setting, no figure

As published on November 10, 2026. Figures that circulate for YouTube, Apple Music and Amazon Music come from third-party measurements, not from the services.

Why platforms normalize loudness

For years, louder masters won attention in a playlist or on the radio, so music and ads were pushed ever closer to full scale with heavy limiting. Loudness normalization ends that race: the player or the broadcaster measures each program and turns it up or down so they all play at about the same loudness.

Once a service normalizes, a master that is louder than the target is simply turned down. It keeps the squashed dynamics it was given but loses the loudness it was squashed for. A master close to the target plays as it was mixed.

Peak level cannot do this job. Two files that both peak at -1 dBFS can differ by many decibels in how loud they sound: a dense song sits near its peak all the time, while a quiet interview has one laugh near the top and everything else far below. That is why the standards use LUFS for loudness and true peak only as a ceiling.

True peak matters because the waveform between samples can rise above the highest sample. When a file is converted to AAC, Opus or MP3, or played through a digital-to-analog converter, those in-between peaks can clip. The -1 and -2 dBTP limits leave room for that.

How this meter measures, and its limits

The method
ITU-R BS.1770-5: K-weighting derived for the file's sample rate, 400 ms blocks with 75% overlap, an absolute gate at -70 LUFS and a relative gate 10 LU below. LRA follows EBU Tech 3342; true peak uses the 4x interpolation filter printed in BS.1770 Annex 2.
Checked against the EBU test signals
The code is tested with the EBU Tech 3341 and 3342 minimum-requirement signals: a 1 kHz tone at -23 dBFS on both stereo channels reads -23.0 LUFS at 44.1, 48 and 96 kHz, and the gated and loudness-range cases land inside their tolerances.
Mono and surround
A mono file counts as one channel, as BS.1770 specifies, so it reads about 3 LU lower than the same audio on both stereo channels. In a 5.1 file the surround channels get +1.5 dB and the LFE channel is left out.
What it does not do
It measures the full mix; the dialogue gating that ATSC A/85 asks for on long-form TV programs is not part of it. The browser decodes the file at the device's sample rate, which leaves loudness unchanged but can shift a true-peak reading by a few tenths of a dB. It is not a live meter for a microphone.

For the sound level in a room through your microphone, use the Decibel Meter.

Levels checked. Now the words.

If the recording is a meeting, interview or voice note, ParrotNotes records it on your phone and gives you the transcript and a summary, free for 100 minutes of recording a month. Ask before you record anyone.

Get ParrotNotes free

Related tools

LUFS Meter FAQ

Sources

LUFS Meter

How to measure LUFS

Drop the file

Read the numbers

Compare with the targets

Optional: apply the gain

What each number means

Integrated loudness (LUFS)

True peak (dBTP)

Loudness range (LRA)

Momentary and short-term max