---
title: "Loudness Meter Online: LUFS, True Peak & Loudness Range"
description: "Free online loudness meter: drop a song and read its integrated LUFS, true peak, loudness range and PLR, and what Spotify's normalization does to it. No upload."
url: "https://theaudiostuff.com/tools/loudness-meter/"
type: "website"
author: "Jakub Charkiewicz"
---

> Markdown rendering of https://theaudiostuff.com/tools/loudness-meter/
> Canonical URL to cite: https://theaudiostuff.com/tools/loudness-meter/
> Generated from the published page. Text and links only; see the HTML for layout.

Breadcrumb: [Home](https://theaudiostuff.com/) > [Tools](https://theaudiostuff.com/tools/) > Loudness Meter

# Online Loudness Meter (LUFS)

A free LUFS meter for audio files: drop a song, a mix or a whole album and read its integrated loudness in LUFS, its true peak in dBTP, its loudness range and its peak to loudness ratio, measured to ITU-R BS.1770-4 and EBU Tech 3342. The page draws the loudness over the length of the track, counts clipped samples and shows the loudness penalty: how far Spotify's normalization, or any target you type, turns the track up or down. The file is measured in your browser and never uploaded.

**Before you start:** the page opens on a demo track made in your browser, so every figure below is a real reading. Choose your own files to replace it.

## Loudness {#ldm-tool-h}

Demo track · made in your browser, 44.1 kHz stereo, 24 s

- **Integrated loudness**: −13.5 LUFS
- **Integrated loudness**: The loudness of the whole track, the figure streaming services level by.
- **True peak**: 0.0 dBTP
- **True peak**: The highest point of the waveform between the samples.
- **Loudness range (LRA)**: 11.6 LU
- **Loudness range (LRA)**: How far the loudness moves between quiet and loud passages.
- **Peak to loudness (PLR)**: 13.5 dB
- **Peak to loudness (PLR)**: The true peak above the integrated loudness.

- **Loudest 3 s (short-term)**: −10.2 LUFS
- **Loudest 0.4 s (momentary)**: −10.0 LUFS
- **Sample peak**: 0.0 dBFS
- **Length measured**: 0:24

The peaks stand a moderate way above the average level.

Clipping: 13 runs of samples at full scale (339 samples), the first at 0:18.

Short-term loudness (3 s) over the track Integrated loudness Your target

Measured in your browser. The file never leaves your device.

## What normalization does to this track {#ldm-targets-h}

A service that normalizes loudness turns each track up or down to its own level before it plays. The rows are the levels Spotify publishes, the EBU's broadcast target, and one you set.

| Target | Level | What happens | Gain | Plays at |
| --- | --- | --- | --- | --- |
| Spotify, Loud | −11 LUFS | Turned up 2.5 dB A limiter catches the peaks at this level, so the loudest moments are changed. | +2.5 dB | −11.0 LUFS |
| Spotify, Normal | −14 LUFS | Turned down 0.5 dB | −0.5 dB | −14.0 LUFS |
| Spotify, Quiet | −19 LUFS | Turned down 5.5 dB | −5.5 dB | −19.0 LUFS |
| EBU R 128 (broadcast) | −23 LUFS | Turned down 9.5 dB | −9.5 dB | −23.0 LUFS |
| Your target | −16 LUFS | Turned down 2.5 dB | −2.5 dB | −16.0 LUFS |

## How to read the loudness figures {#ldm-read-h}

### The four figures

**Integrated loudness**, in LUFS, is the loudness of the whole track measured to ITU-R BS.1770-4: the signal is weighted the way hearing weighs frequencies, averaged in 400 ms blocks, and quiet passages are gated out so a long fade does not pull the figure down. A higher figure (closer to zero) is a louder track: −8 LUFS is much louder than −16 LUFS.

**True peak**, in dBTP, is the highest point the waveform reaches between its samples, found by interpolating the signal. It can sit above the highest sample, and above 0 dBTP a converter or a lossy encoder can clip a file whose samples never touch full scale.

**Loudness range (LRA)**, in LU, is how far the loudness of 3 second stretches moves through the track: the 95th percentile minus the 10th, as EBU Tech 3342 defines it. **Peak to loudness ratio (PLR)** is the true peak minus the integrated loudness.

### What the figures say about a master

Compressing and limiting a master raises its loudness and brings its peaks closer to it, so a master pushed harder shows a higher LUFS figure, a lower PLR and usually a smaller loudness range. Measure an original release beside its remaster and the three figures show which was mastered harder.

None of them is a quality score. A quiet master with a wide range can sound dull, and a loud one can sound exactly as its makers intended. What the figures settle is the level question: once a service normalizes playback, a master made loud by heavy limiting plays no louder than a dynamic one, and it keeps the limiting.

To compare two masters by ear with the loudness matched, load both into the [ABX blind test](https://theaudiostuff.com/tools/abx-test/). To see what a file's spectrum holds, such as a lossy source or padded bit depth, use the [FLAC and hi-res audio checker](https://theaudiostuff.com/tools/hires-audio-checker/), which prints these loudness figures for every file it reads.

## Published loudness targets {#ref-ldm-title}

The targets in the table above, as their publishers state them. Spotify's are from its help page for artists, "Loudness normalization on Spotify"; the broadcast target is EBU R 128. Apple's mastering guide describes Sound Check without a target figure, and YouTube's help pages state none, so this page prints no row for either: type the level you want to test into Your target.

| Target | Level | Rule for a quieter track |
| --- | --- | --- |
| Spotify, Loud (Premium setting) | −11 LUFS | Turned up to the level, with a limiter on the peaks |
| Spotify, Normal (the default) | −14 LUFS | Turned up only as far as leaves the true peak 1 dB under full scale |
| Spotify, Quiet (Premium setting) | −19 LUFS | Turned up only as far as leaves the true peak 1 dB under full scale |
| EBU R 128 (broadcast) | −23 LUFS | Normalized to the level, with a maximum true peak of −1 dBTP |

## What a loudness meter measures {#tprimer-title}

A peak meter shows the highest sample in a file, which says almost nothing about how loud the file sounds: a single drum hit can reach full scale in a quiet track. A loudness meter measures what hearing responds to, the energy of the signal over time, weighted toward the frequencies the ear is most sensitive to.

The method is ITU-R BS.1770, the one broadcasters and streaming services use. The signal passes through a filter that lifts the treble a little and removes the deepest bass, its energy is averaged in 400 ms blocks, and the blocks are averaged again with the quiet ones left out. The result is in LUFS, loudness units relative to full scale: 0 LUFS is the loudest a signal could measure, and every figure for real music is negative.

One LU is one decibel, so a track at −10 LUFS has to be turned down 4 dB to match one at −14 LUFS. That is what a streaming service does when it normalizes playback: it measures each track once and applies one gain to the whole of it, which changes the level and leaves the sound of the master as it was.

### The ideas behind the controls {#tprimer-ideas}

- **LUFS**: Loudness units relative to full scale, the unit of a loudness measured to BS.1770. LKFS is the same thing under its older name.
- **Integrated loudness**: The loudness of the whole file as one number, with silences and very quiet passages gated out.
- **Short-term and momentary**: The loudness of the last 3 seconds and of the last 0.4 seconds. The chart draws the short-term loudness through the track.
- **True peak (dBTP)**: The highest point of the waveform between the samples, found by interpolation. It is the peak a converter has to reproduce.
- **Loudness range (LRA)**: The spread, in LU, between the quiet and the loud stretches of a track, with the extremes left out.
- **PLR**: Peak to loudness ratio: the true peak minus the integrated loudness. Heavy limiting lowers it.

### Common mistakes {#tprimer-mistakes}

- Mastering to a streaming target as if it were a rule. A service turns a louder master down to its level and plays it; the target only says where that level is.
- Reading the sample peak as the true peak. A file whose samples stop at −0.1 dBFS can reach above 0 dBTP between them, and clip a converter or a lossy encoder.
- Comparing two masters by ear without matching their loudness. The louder one tends to win a quick comparison whatever else differs.
- Taking a low loudness range for a fault. Dense, steady music has a small range by nature; the figure describes the music as well as the mastering.
- Measuring a clip and calling it the track. Integrated loudness is defined over the whole program, and an intro or a chorus alone reads differently.

### What this tool cannot tell you {#tprimer-limits}

- It measures files, never live sound. For the level of a room through a microphone, the decibel meter is the tool.
- The true peak is an estimate, as every interpolating meter's is: the signal is interpolated four times below 96 kHz and twice below 192 kHz, and the worst case between those points can sit a fraction of a decibel higher.
- The targets are what their publishers state. A service can change its level, and a service that publishes none has no row here.
- A file is measured as your browser decodes it. A lossy file can decode with peaks above full scale, which is why its true peak can read above 0 dBTP.
- DSD files are converted to PCM first, with full modulation at 0 dBFS, so they read 6 dB lower than in a player that adds 6 dB of gain.

### Where the math comes from {#tprimer-sources}

- [ITU-R BS.1770-4: Algorithms to measure audio programme loudness and true-peak audio level](https://www.itu.int/rec/R-REC-BS.1770)
  Standard International Telecommunication Union
  The K-weighting, the gated integrated loudness in LUFS and the true peak in dBTP that the page measures.
- [EBU Tech 3342: Loudness Range: A measure to supplement EBU R 128 loudness normalization](https://tech.ebu.ch/docs/tech/tech3342.pdf)
  Standard European Broadcasting Union PDF
  The loudness range: short-term loudness over 3 s windows, gated, the 95th percentile minus the 10th.
- [EBU R 128: Loudness normalisation and permitted maximum level of audio signals](https://tech.ebu.ch/publications/r128)
  Standard European Broadcasting Union
  The broadcast target in the table: program loudness normalized to -23 LUFS, with a maximum true peak of -1 dBTP.

## Loudness meter FAQ {#ldm-faq-title}

1. ### What is LUFS? {#ldm-faq-what-is-lufs}
   LUFS stands for loudness units relative to full scale. It is the unit of loudness measured to ITU-R BS.1770, the method broadcasters and streaming services use: the audio is filtered to match how hearing weighs frequencies, its energy is averaged, and quiet passages are left out of the average. One loudness unit equals one decibel, and figures are negative, so −9 LUFS is a loud modern master and −20 LUFS a quiet, dynamic one. LKFS is the same unit under an older name.
2. ### How do I check the LUFS of a song? {#ldm-faq-how-do-i-check-the-lufs-of-a}
   Press Choose audio files and pick the song, or drop it on the card. The page decodes it in your browser and shows its integrated loudness in LUFS within a few seconds, with the true peak, the loudness range and the peak to loudness ratio beside it. MP3, FLAC, WAV, AAC, M4A, ALAC, OGG, AIFF and DSD files work, and nothing is uploaded.
3. ### What is a loudness penalty? {#ldm-faq-what-is-a-loudness-penalty}
   It is the amount a streaming service turns a loud track down when it normalizes playback. A master at −8 LUFS played on a service that levels to −14 LUFS is turned down 6 dB, so the loudness gained by limiting it hard is taken back, and the limiting stays. The table under the meter shows the gain for each of Spotify's published levels, for EBU R 128 and for any target you type.
4. ### What LUFS does Spotify normalize to? {#ldm-faq-what-lufs-does-spotify-normalize-to}
   Spotify's help page for artists states −14 LUFS, measured to ITU 1770, as its default level. Premium listeners can choose Loud, at −11 LUFS, or Quiet, at −19 LUFS. A louder track is turned down to the level. A quieter one is turned up only as far as leaves its true peak 1 dB under full scale, except at Loud, where a limiter catches the peaks. An album is leveled as a whole, and Spotify does not normalize in its web player.
5. ### What about YouTube, Apple Music and other services? {#ldm-faq-what-about-youtube-apple-music-and-other-services}
   This page prints a row only for a figure the service itself publishes. Apple's mastering guide describes Sound Check without stating a target, and YouTube's help pages state no figure, so neither has a row. To test a level you have read elsewhere, type it into Your target: the row shows the gain it would apply and warns when that gain would push the true peak above −1 dBTP.
6. ### What is true peak, and how can it be above 0 dB? {#ldm-faq-what-is-true-peak-and-how-can-it}
   A digital file stores the waveform as samples, and the waveform between two samples can rise higher than either of them. True peak is an estimate of that highest point, found by interpolating between the samples, and it is written in dBTP. A file whose samples stop just under full scale can have a true peak above 0 dBTP, which a converter or a lossy encoder may clip. Leaving about 1 dB of room, a true peak of −1 dBTP, avoids it.
7. ### What is loudness range (LRA)? {#ldm-faq-what-is-loudness-range-lra}
   Loudness range says how much the loudness changes through a track. The meter measures the loudness of every 3 second stretch, leaves out silence and the quietest passages, and takes the distance between the level 10% of the stretches fall below and the level 95% fall below, in LU. A film or an orchestral piece can exceed 15 LU, and a dense pop or rock master often sits between 3 and 8 LU.
8. ### Can it tell me if a track is too compressed? {#ldm-faq-can-it-tell-me-if-a-track-is}
   It gives the evidence without a verdict. A master that was limited hard shows a high integrated loudness, a low peak to loudness ratio, often under 8 dB, a small loudness range and sometimes clipped samples, which the page counts. Whether that suits the music is a judgment the figures cannot make. Comparing the original release of an album with a remaster shows which of the two was pushed harder.
9. ### Can I measure a whole album? {#ldm-faq-can-i-measure-a-whole-album}
   Yes. Choose or drop several files at once, up to 100. Each file gets its own row with its loudness, true peak, loudness range and peak to loudness ratio, and the page adds the loudness of all the files measured as one program, which is how a service that levels a whole album measures it. Press Show on a row to see that track's chart and its normalization table.
10. ### Is my audio file uploaded? {#ldm-faq-is-my-audio-file-uploaded}
    No. The file is decoded and measured by your own browser on your device. Nothing about it is sent anywhere or stored, and closing the tab clears it.

## You might also use these tools {#rt-title}

- Generator **Which test tone do you need?** [Online Tone Generator](https://theaudiostuff.com/tools/signal-generator/) Sine, square, triangle, sawtooth and pulse tones at any frequency or musical note, white, pink, brown, blue and violet noise, frequency sweeps, a channel test and a polarity check. Runs entirely in your browser.
- Test **Are my speakers or headphones wired right?** [Speaker & Headphone Test](https://theaudiostuff.com/tools/speaker-test/) Check that left and right come from the correct side, that a pair of speakers is in phase, how low the bass goes and how high the treble reaches, then sweep 20 Hz to 20 kHz for rattles.
- Test **How low do my speakers or subwoofer really go?** [Bass & Subwoofer Test](https://theaudiostuff.com/tools/bass-test/) Step a tone down from 200 Hz to 16 Hz, find the note where your bass fades out, set a subwoofer's phase by ear and sweep the bass for rattles and booms.
