Podcast guide

What loudness should a podcast be?

Aim for about -16 LUFS integrated loudness, with true peaks no higher than -1 dBTP. That is what Apple Podcasts recommends in its published audio requirements: loudness around -16 LKFS, give or take 1 dB, and a true peak that does not exceed -1 dB FS.

LKFS and LUFS are the same unit under two names. Apple words these figures as recommendations, not as a rule that rejects an episode, and its stated reason is to avoid volume adjustments driven by the content. Finality's podcast review works from the same figures.

Last checked against Apple's published audio requirements on 30 September 2026.

What -16 LUFS actually measures

Integrated loudness is one number for a whole episode: how loud it sounds on average, weighted for how the ear hears and with near-silent passages left out, as defined in ITU-R BS.1770. It is measured down from digital full scale, so the numbers are negative and closer to zero is louder. -14 LUFS is louder than -16 LUFS.

It is not the loudest moment. That is the peak, and an episode can have a high average loudness with a modest peak, or a low average with a very high one.

Why a true-peak limit sits beside the loudness target

A sample peak is the highest stored sample. A true peak estimates the highest level the waveform reaches between samples, which can be higher, and which is what can clip after conversion or encoding. It is measured in dBTP. A ceiling of -1 dBTP leaves a little room for that.

The two limits work against each other. The gap between an episode's true peak and its integrated loudness is its peak-to-loudness ratio. To sit at -16 LUFS without crossing -1 dBTP, that gap can be at most 15 dB.

Can a quiet episode just be turned up?

Sometimes. It depends on that gap.

• An episode at -22 LUFS with a true peak of -9 dBTP has a 13 dB gap. Raising it by 6 dB reaches -16 LUFS, and the peak lands at -3 dBTP, under the ceiling. Turning it up is enough.
• An episode at -25 LUFS with a true peak of -3 dBTP has a 22 dB gap. Reaching -16 LUFS would take 9 dB of gain and put the peak at +6 dBTP. Only 2 dB fits under the ceiling, so gain alone gets it to -23 LUFS.

Speech recorded without compression often looks like the second case, because a few short, sharp sounds, such as a plosive, a laugh or a bump, set the peak while the voice sits much lower.

How to measure your own episode

Any loudness meter that follows ITU-R BS.1770 will give you both numbers. If you have ffmpeg installed, this prints them:

ffmpeg -i episode.mp3 -af ebur128=peak=true -f null -

Read the summary at the end of the output. Integrated loudness is the line labelled I:, in LUFS. The true peak is the line labelled Peak: under True peak; ffmpeg shows the unit as dBFS, but with peak=true it is the true-peak value.

What to do when the peaks are in the way

• Find the specific sounds that set the peak, such as a plosive, a cough or a laugh, deal with them, and measure again. Sometimes a few moments are responsible.
• Reduce the peaks with a limiter or compressor on the voice, then raise the level. This changes how the voice sounds, so it is an editing decision, and Apple's guidance cautions against heavy compression and amplification.
• Accept a quieter episode. That is a legitimate choice. The cost is that it will play quieter than episodes that meet the target.

Where Finality fits

Finality's podcast review reports integrated loudness and true peak beside Apple's figures, says how much gain the true-peak ceiling allows, and lists plosives and long pauses with timestamps so you can listen to the moments that matter. It can return a level-corrected file, but it will not compress your voice: where gain alone cannot reach the target, it reports the shortfall and stops.