Skip to content
Media 1 for listing Audio Normalizer

Description

Version 2.0 is out!!

A new offline rendering engine up to 200x faster than real time, and rebuilt fully native in C++ and Slate. Details under New in 2.0.

Video tour: overview

Video tour: in depth

Website and documentation

Support: [email protected]

Consistent loudness across hundreds of assets, in one pass.

Audio Normalizer measures your Unreal audio with industry-standard BS.1770-5 K-weighted LUFS metering and brings every asset to your target. Procedural sources are rendered offline with their random logic intact and averaged across variations, so MetaSounds and Sound Cues get a real reading, not a guess.

Features

  • BS.1770-5 integrated LUFS metering with true K-weighting derived per sample rate

  • True peak, 4x oversampled

  • Sound Waves, Sound Cues and MetaSounds, with per-type filters and Select All

  • Offline rendering engine, up to 200x faster than real time

  • MetaSounds rendered with their random logic intact and averaged across variations; Sound Cues rendered through their full graph

  • Low, Mid and High presets or custom LUFS and true-peak targets, with a tolerance to skip assets already in range

  • Analyze, preview every asset's current and target loudness, then apply

  • Every gain change re-rendered and re-measured before it is kept, rolled back if it is off

  • The whole batch applied as one Unreal undo step

  • Non-modal: keep working in the editor while it runs

  • Text, CSV and filtered-assets reports, one switch each

  • Blueprint API for analysis and normalization, synchronous and asynchronous

  • Native Unreal window: Enter runs, Escape cancels

New in 2.0

  • Offline rendering, up to 200x faster than real time. Analysis no longer waits for the sound to play. Large batches finish in seconds.

  • Fully native C++ and Slate. Rebuilt on the engine's own code and widgets: instant startup, engine-native look, and fast moves to each new engine version.

  • BS.1770-5 metering with true K-weighting. The filter is derived for each file's own sample rate, and the implementation is checked against FFmpeg's ebur128.

  • True peak. 4x oversampled true-peak measurement replaces sample peak.

  • Verified gains. Every change is re-rendered and re-measured before it is kept. If the result is off, it is rolled back. No more accidental +80 dB.

  • Preview, then apply, then undo. See every asset before anything changes, apply as one undo step, undo as one step.

  • Reports. Text, CSV and a filtered-assets report.

Included formats