LUFS File Checker

Knowledge Base

Integrated LUFS vs. a Live Meter

A live LUFS meter (like the site's Podcast Level Meter) shows momentary and short-term loudness as you record, useful for real-time gain staging, but it can't tell you your episode's true overall Integrated Loudness until the whole file exists. This tool measures the complete, final file using the same gated algorithm broadcast and streaming platforms use to check compliance.

What Integrated Loudness, LRA, and Peak Estimates Tell You

• Integrated Loudness (LUFS): The single number platforms check against their target. This is what determines whether your file gets turned up or down on playback.
• Loudness Range (LRA): How much your loudness varies over time, in LU. A very high LRA on a podcast usually means inconsistent mic levels between speakers or segments.
• Peak estimate: This browser result uses linear interpolation and cannot reveal all inter-sample peaks. Use a true-peak meter with a reconstruction filter for formal delivery checks.

FAQ

Q: Why does this reading differ from my DAW's loudness meter?
A: Differences can come from gating, filters, sample rate, channel layout, and true-peak implementation. Use this tool for a local estimate and a validated meter for delivery acceptance.

Q: My file measures louder than my target, what do I do next?
A: Use the site's Loudness Normalizer to automatically adjust gain to your target LUFS and download a corrected file.

What a loudness measurement tells you

LUFS measures perceived loudness rather than electrical level, which is why it replaced peak metering for delivery. A peak meter reports the single highest sample and says nothing about how loud something sounds. Two files can peak identically and differ enormously in perceived loudness, and only one of them will survive a platform's normalisation unchanged.

Three measurements get reported and they answer different questions. Integrated loudness covers the whole file and is the number a delivery specification refers to. Short-term covers the last three seconds and shows how loudness moves through a programme. Momentary covers the last four hundred milliseconds and is closest to what a meter feels like while you work.

Loudness range describes how much the loudness varies across the file. A drama with whispers and shouting has a wide range, and a podcast recorded consistently has a narrow one. A very wide range on spoken word usually means levels drifted rather than that the material is dynamic, and it is the figure to look at when listeners report having to adjust volume.

Every major platform normalises on playback, turning your file up or down toward its own target. This means delivering louder than the target gains nothing, since the platform simply turns it down, and doing so by compressing hard leaves you with a quieter, flatter result than a competitor who delivered at the target. Targets differ between services and get revised, so check the current published specification for wherever you are delivering rather than working from a remembered number.

Browser measurement is an estimate. A full implementation of the loudness standard includes filtering and gating stages, and any calculation running in a page approximates some of that. Treat the figure as close enough to make decisions and use a certified meter for anything being delivered under contract.

Vocabulary

LUFS. Loudness Units relative to Full Scale, the standard measure of perceived loudness. Weighted to approximate human hearing, so a LUFS figure predicts how loud something sounds in a way a peak reading cannot.

Integrated loudness. The loudness of an entire file measured as one figure, with silent passages excluded by a gating rule. The number delivery specifications refer to when they state a loudness target.

Loudness range. How much loudness varies across a programme, in loudness units. A narrow range suits spoken word, and a wide one suits drama. On a podcast a wide range usually indicates drifting levels rather than intent.

True peak. The highest level a waveform actually reaches between samples, which can exceed the highest sample value. Measured because converters and lossy encoders reconstruct those in-between peaks and can distort where a file peaks at maximum.