The Complete Guide to Loudness Meters: LUFS, LRA & True Peak in the Browser
Your mix peaks under 0 dBFS but still sounds quiet next to other YouTube videos. Or it sounds loud and fatiguing after a minute. Peak meters alone cannot explain that — loudness meters can.
This free loudness meter measures integrated LUFS, loudness range (LRA), and true peak when FFmpeg’s ebur128 filter reports them. Everything runs in your browser with the same local engine used by our other audio tools. No signup. No upload of your file for analysis. First run may download about 30MB of engine assets from this site.
This guide explains LUFS in plain language, how to use the tool, how to read peaks, common reference targets (streaming, podcast, broadcast), creator workflows, and honest limits — including that this is not a certified broadcast compliance certificate.
After you measure, use Audio Trim, Audio Converter, or Silence Remover for edits — gain riding still happens in your DAW or limiter.
In this guide
FoundationsWhat is a loudness meter?
A loudness meter estimates how loud audio sounds to human ears over time, using standards related to EBU R128 / ITU-R BS.1770 (K-weighting, gating, integration). The headline number most creators watch is integrated loudness in LUFS (or LKFS in US broadcast wording — same idea for practical purposes).
Unlike a simple peak meter that only cares about the tallest sample, loudness accounts for duration and perception. Two tracks can share the same peak while one feels twice as “present.” Streaming platforms use loudness-aware processing so hyper-smashed masters do not always win the volume war.
What this browser tool reports
- Integrated LUFS (I) — overall programme loudness
- LRA — loudness range (how much level varies)
- True peak (dBTP) — when the engine logs it
- Sample peak (dBFS) — when present
Measurement, not mastering
The meter does not normalize, limit, or “fix” your file. It gives numbers so you can adjust gain, compression, or limiting in tools designed for that. For delivery contracts, use the meter your client specifies.
WorkflowHow to measure LUFS with this free meter
Add audio or video
Drop MP3, WAV, M4A, FLAC, or a video file. Video is fine — only the soundtrack is analyzed. Soft size guidance ~200MB.
Optional preview
Use the on-page player to confirm you loaded the right export.
Measure loudness
Click Measure. First visit may load the FFmpeg engine (~30MB). The tool runs ebur128 and parses the log for I, LRA, and peaks.
Compare to a target
Pick YouTube-ish (−14), podcast (−16), or broadcast EBU (−23) references. The delta in LU shows whether you sit louder or quieter than that ballpark.
Adjust elsewhere, re-measure
Change gain/limiting in your DAW or host, export again, measure again. Iterate until you like the balance of loudness and dynamics.
MetricsIntegrated LUFS and loudness range (LRA)
Integrated loudness answers: “How loud is this whole piece if I listen straight through?” More negative numbers are quieter (for example −20 is quieter than −12). LRA answers: “How much does loudness swing?” Dialogue-heavy podcasts often have moderate LRA; film mixes can be wider; hyper-compressed pop can be very narrow.
| Metric | Unit | What it tells you |
|---|---|---|
| Integrated (I) | LUFS | Overall programme loudness |
| LRA | LU | Dynamic variation of loudness |
| Momentary (M) | LUFS | Very short window (if logged) |
| Short-term (S) | LUFS | Short window (if logged) |
Short files
Very short clips (a few seconds) can produce odd integrated readings because gating and integration assume programme-like material. Prefer measuring a full episode, full song, or a representative section.
PeaksTrue peak vs sample peak
Sample peak is the highest digital sample value in the file (dBFS). True peak estimates peaks that can appear after reconstruction filtering — sometimes higher than sample peak. Leaving true-peak headroom helps avoid distortion after lossy encode (YouTube, podcast hosts, social).
| Weak habit | Better habit |
|---|---|
| Only watch sample peak at −0.1 dBFS | Also check true peak headroom |
| Crush to max LUFS with no peak margin | Hit a LUFS target with clean true peak |
| Ignore encode after limiting | Re-measure the file you actually upload |
If true peak is missing from the log, sample peak may still appear. Treat missing fields as “not reported,” not as zero.
ReferencesYouTube, podcast, and broadcast targets
Targets change with platform policy and genre. Use them as orientation, not law. YouTube applies loudness normalization — extremely hot masters may simply get turned down. Podcast networks often publish a show loudness target in their tech specs.
| Context | Ballpark integrated | Notes |
|---|---|---|
| Streaming / YouTube-ish | Around −14 LUFS | Platform still normalizes |
| Many podcasts | Around −16 LUFS | Check your host’s guide |
| EBU R128 programmes | −23 LUFS | Broadcast reference |
The on-page target chips compute how many LU you sit above or below the selected reference. Louder than target (positive delta) often means turn down or ease limiting; quieter means the track may feel soft next to peers.
Consistency beats chasing one number
For a series, pick a show target and stick to it episode after episode. Viewers notice jumpiness between episodes more than a 0.5 LU difference from a blog post’s recommendation.
PlaybooksCreator workflows
YouTube voiceover or talking head
Export mixdown → measure integrated + true peak → adjust limiter ceiling / makeup → re-export → measure again → upload. Pair with clear packaging (titles, chapters).
Podcast episode
After edit and noise cleanup, measure full episode loudness. If LRA is extreme, check whether music beds or ads dwarf speech. Silence cleanup (Silence Remover) changes timing; re-measure after major edits.
Music master preview
Use the meter as a sanity check before a mastering engineer or before self-mastering. Competitive loudness is genre-dependent; do not force −14 if it kills your dynamics without a reason.
AvoidCommon loudness mistakes
- Mixing only to peak — peak-safe but quiet or harsh overall.
- Assuming −14 always wins on YouTube — normalization and content quality still matter more than 0.3 LU.
- Measuring a 2-second sting — use a representative programme length.
- Treating this as legal QC — use certified tools for contractual delivery.
- Expecting auto-fix — measurement only; change gain in a DAW.
- Ignoring true peak before lossy encode — leave headroom for AAC/MP3.
TrustPrivacy, engine load, and honest limits
Analysis is local. We do not market a cloud loudness API. The FFmpeg wasm engine loads from the same origin under /ffmpeg. Soft file-size guidance protects browser memory. Mobile phones may struggle with long masters — desktop is happier.
- Not a stream ripper — only files you open.
- Not automatic loudness normalization.
- Not a substitute for a rack meter in a broadcast chain.
- ebur128 availability depends on the wasm build; failures get a clear error.
Cheat sheetQuick reference
| Item | Guideline |
|---|---|
| Integrated (I) | Overall loudness in LUFS |
| LRA | Loudness variation in LU |
| True peak | dBTP when reported |
| Streaming ballpark | ~−14 LUFS (reference only) |
| Podcast ballpark | ~−16 LUFS (check host) |
| EBU programme | −23 LUFS reference |
| Engine | FFmpeg ebur128 in browser |
| Privacy | No upload for analysis |
| Action | Measure → adjust DAW → re-measure |
Ship checklist
Right export · full programme measured · integrated + peaks noted · target chosen · headroom OK · re-measure after limiting · upload the file you measured.
Wrap-upMeasure first, then decide
Loudness meters turn “sounds quiet” into a number you can act on. Use this free browser meter to read integrated LUFS, LRA, and peaks privately — then fix levels with intention instead of endless A/B guessing.
Scroll to the tool above, drop your mix, and measure before you publish.