How to Normalize Audio Loudness (EBU R128 / LUFS, Explained)

To make every audio file sound equally loud, use Zro7 Normalize Audio. It runs FFmpeg's loudnorm filter (an EBU R128 implementation) to target a specific LUFS loudness — the same standard used by Spotify (-14 LUFS), Apple Music (-16 LUFS), YouTube (-14 LUFS), and podcast networks (-16 LUFS).

LUFS vs peak: why peak normalization sounds wrong

Old-school "normalize to -1 dB" scales every file so its loudest sample hits -1 dB. But peak has nothing to do with perceived loudness — a drum hit and a whispered vowel can share the same peak while sounding wildly different in volume. LUFS (Loudness Units relative to Full Scale) measures perceived loudness across the whole file, using human-hearing weighting curves. Two files at -14 LUFS truly sound equally loud.

Target LUFS by platform

• Spotify / YouTube / Amazon Music: -14 LUFS integrated.
• Apple Music / Tidal: -16 LUFS.
• Podcasts (AES recommendation): -16 LUFS mono, -19 LUFS stereo.
• Broadcast TV (EBU R128): -23 LUFS.
• Cinema: -27 LUFS (there's a lot of headroom for explosions).

How Zro7 does it

1. First pass: loudnorm measures the file's integrated LUFS, true peak, and loudness range.
2. Second pass: applies gain + limiter to hit the target LUFS while keeping true peak below -1 dBTP.
3. The result is fully compliant with EBU R128 and Spotify/Apple loudness specs.

Steps

1. Open Normalize Audio.
2. Drop one file or a batch.
3. Pick target LUFS (default -16 for podcast, -14 for music).
4. Click Normalize.
5. Download.

Batch normalize a podcast season

Consistency matters: listeners get frustrated when episode 12 is 6 dB quieter than episode 11. Drop all episodes in at once; each is normalized to the same LUFS target independently. Combine with Podcast Cleaner to also denoise and trim silence in one pass.