Podcast and Spoken-Word Audio

How to get your podcast to -16 LUFS in REAPER (measure it, then hit it)

By StudioRescue Editorial

Published July 18, 2026 · Updated July 18, 2026

“How loud should my podcast be?” is one of the most common questions from people producing their own show, and the answer is more specific than it sounds. Streaming platforms turn everything toward a target loudness on playback, so a podcast that is far off that target gets turned up or down automatically, and if you were much too loud the platform can leave your audio squashed. Getting close to the target yourself is how you keep control of your sound.

The number people quote most for podcasts is -16 LUFS. That comes from Apple Podcasts, which recommends the overall loudness sit around -16 dB LKFS with a tolerance of about ±1 dB, and the true peak stay under -1 dB FS. Below is what those terms mean and how to hit them in REAPER without buying anything.

What LUFS actually measures

LUFS stands for Loudness Units relative to Full Scale. It is a loudness measurement defined by the ITU-R BS.1770 standard, and it is built to track how loud something actually seems to a listener, not just how tall the waveform is. You will also see the label LKFS, which Apple uses. For practical purposes LKFS and LUFS are the same measurement, so a reading of -16 LKFS and -16 LUFS mean the same thing.

There are a few flavors of the reading, and the difference matters:

• Integrated (LUFS-I). The average loudness across the whole episode. This is the number platforms mean when they list a target, and the one you are trying to land on.
• Short-term (LUFS-S). Loudness over the last few seconds. Useful for watching a meter while you work, not the number you deliver.
• Momentary. A fast, moment-to-moment reading. Good for spotting a loud spike, not for setting your final level.

True peak is separate from all of these. It is the highest actual sample value in the file, and it is why the target comes with a peak ceiling. Keeping true peak under -1 dBFS leaves headroom so that lossy encoding, which most podcast apps use, does not push a peak into clipping.

Fix the sound before you fix the number

Loudness is the last step, not the first. Hitting a target on a recording that is noisy, uneven, or harsh just gives you a noisy, uneven, harsh episode at the correct volume. Get the voice sitting right first: clean obvious problems, then balance levels between speakers with compression so the episode is reasonably consistent. If you are still building your processing, the stock-plugin REAPER vocal chain walks through a beginner build order, and if a voice sounds thick or boomy the guide to clearing muddy low mids covers that first. Once the episode sounds the way you want, measure and set the loudness.

Measure and hit -16 LUFS in REAPER

REAPER has had built-in loudness measurement and loudness normalization on render since version 6.30, and they are in current REAPER (7.77 at the time of writing). The Render dialog can measure your integrated loudness and adjust the whole file to a target in one step.

If your render misses the target

Normalizing on render is reliable, so a miss usually points to something upstream.

1. You rendered a selection, not the whole show. Integrated loudness is measured across whatever you render. Set Bounds to Entire project.
2. The true-peak limit is fighting the target. If a few extreme peaks force heavy limiting, the perceived loudness can shift. Tame obvious spikes with compression or a clip-gain reduction first, then let normalization do the gentle final move.
3. You measured momentary or short-term by mistake. Only the integrated (LUFS-I) reading is the delivery number. A short-term meter bouncing around -16 is not the same as an integrated -16.
4. Mono versus stereo. Apple’s recommendation is referenced to stereo. Most podcast exports are stereo or dual mono, so target -16 LUFS-I and keep it simple unless your host specifies a separate mono figure.

What hitting -16 LUFS will not do

Loudness normalization changes level, and only level. It is not a cleanup or a quality step.

• It will not remove background noise, hum, or room echo. Raising a quiet, noisy recording to -16 raises the noise with it. Clean first, level second.
• It will not fix a clipped or distorted recording. Turning distortion down just gives you quieter distortion.
• It will not even out two speakers at very different levels. Integrated loudness is one average for the whole file, so a soft guest and a loud host still sound uneven. That is a job for leveling and compression on the way in.
• It is a target, not a seal of quality. An episode can measure exactly -16 LUFS and still sound thin or harsh. The number only means the overall volume is in the right neighborhood.