Peak reduction

Tame protruding sample peaks while leaving below-threshold samples unchanged.

Audio editor showing a real recording waveform, spectrogram and tools
A real recording in the editor. This screenshot demonstrates the workflow; evaluate treatment results using your own audio.

When should you choose this tool?

A knock or brief spike may be disproportionately loud compared with speech. Peak reduction reduces only the amplitude portion exceeding the chosen threshold. An impact’s longer tail may remain audible.

How it changes audio

Below-threshold samples remain unchanged; excess amplitude is reduced by the chosen percentage. This is sample-wise amplitude processing, without reconstruction, makeup gain or channel linking.

What do the parameters control?

Threshold
−60–0 dBFS. Only the waveform above this level changes. Use dragging or numeric controls and inspect actual affected regions.
Reduction
0–100% of excess amplitude above threshold. At 100%, sample peaks are capped at the threshold; this is not a true-peak guarantee or a dB compression ratio.

Try a short passage

  1. Select the spike with quieter surroundings. Place the threshold above speech you want preserved, if the recording permits that distinction.
  2. Increase reduction gradually during Preview. The graph and affected-sample summary refer to the current preview scope.
  3. Listen to Residual and the whole event. If a harsh edge appears, reduce strength or narrow the time selection.

What to listen for

The spike should be less intrusive without making surrounding speech crackle. Residual contains the difference above threshold.

Limits of the method

Strong settings can introduce distortion. It does not reconstruct clipped peaks; De-clip is the separate repair method.

Preview covers a short passage. After applying, review the full treated section; you can undo the step from history.

Related problems and workflows

Other useful tools