What LUFS should your podcast or video be? Loudness normalization explained

For podcasts, target -16 LUFS (stereo) or -19 LUFS (mono) — the standard Apple Podcasts recommends. For video going to YouTube or a social platform, target -14 LUFS instead, which matches what those platforms actually normalize toward. LUFS (Loudness Units relative to Full Scale) measures how loud audio actually sounds to human ears, not just its raw peak volume — which is why it's the number that matters for consistent playback.

You can normalize to either target free, in your browser: normalize your audio free here, no upload required.

Loudness normalization vs. peak normalization

Peak normalization looks at the single loudest instantaneous sample in your audio and scales the whole file so that peak hits a target level (often just under 0dB, to avoid clipping). It says nothing about how loud the audio sounds overall — two files with the same peak level can sound very different in perceived loudness depending on their dynamics.

Loudness normalization instead measures perceived loudness across the whole file, using a standard (LUFS, based on the EBU R128 spec) that accounts for how human hearing actually works — quiet passages and loud passages get averaged in a way that correlates with what listeners actually perceive as "how loud is this." This is the measurement streaming platforms use to keep volume consistent from one piece of content to the next.

What LUFS target should you use?

-16 LUFS (stereo) / -19 LUFS (mono): The established podcast standard, recommended by Apple Podcasts. Leaves headroom for dynamic speech without sounding quiet next to other podcasts.

-14 LUFS: What YouTube, Spotify's music catalog, and most social platforms (Instagram, TikTok) target. Louder than the podcast standard — use this if you're publishing video content or music rather than a podcast feed.

Note that Spotify's own podcast player normalizes toward -14 LUFS even though Apple Podcasts still recommends -16 — so if you distribute to both, -16 with a bit of headroom is the safer middle ground, since content above the target gets turned down (no artifacts) while content below target sometimes gets turned up (which can add noise if you normalized too conservatively).

Why your audio might still sound too quiet after normalizing

Two common causes. First: you may have targeted the wrong platform's LUFS value — normalizing to -16 and then listening on a platform that expects -14 will genuinely sound quieter by comparison, even though the normalization itself worked correctly. Second: some tools only measure and report loudness without actually applying gain — worth double-checking that your output file's measured loudness has actually changed, not just been read.

Normalization won't fix distortion

Loudness normalization adjusts overall level — it doesn't repair distortion or clipping that's already baked into a recording. If a section was recorded too hot and clipped, no amount of loudness adjustment brings back the audio information that got clipped off; that has to be fixed (or re-recorded) before normalizing, not after.

Normalize before or after removing background noise?

After. If you normalize loudness first, the noise floor sitting underneath your speech gets factored into the measurement, which throws off where your actual speech ends up landing relative to the target. Remove background noise first, then normalize — the loudness measurement reflects your real speech, and the target lands where you actually intended.

Does this work on video, or just audio?

Loudness normalization applies to a video's audio track — drop in a video file and only the sound gets processed, with the picture passed through unchanged. This is useful for the same reason it matters for podcasts: viewers shouldn't have to adjust volume between your video and whatever they watched right before it.

The bottom line

LUFS is the number that actually predicts how loud your content sounds to a listener, unlike peak level. Pick -16 for podcasts, -14 for video and social, clean up noise first, and normalize your loudness free with WavyVid — done entirely on your device, no upload required.