How to normalize the volume
1. Drop one audio file. Video and pictures are refused here. 2. Leave Speech for a voice note, or pick Music or Quiet. 3. Press “Normalize audio” and play the download next to the original.
GustKit uses FFmpeg loudnorm in one pass. Speech aims at -16 LUFS, music at -14 and quiet at -23, with a peak ceiling so a sudden spike does not clip as hard. The file is encoded again at 48 kHz in the same kind: WAV stays WAV, MP3 stays MP3.
A target level, not a percent
Change volume multiplies the samples by a percent you pick. A quiet memo and a loud song both get the same boost, so one of them can distort. This page measures the file and moves it toward one loudness. It does not remove noise, and it does not change speed or pitch.
The audio stays in this tab
The first visit downloads FFmpeg. Later visits reuse the browser cache. Files longer than the limit on the page are refused, because the whole file is re-encoded here. GustKit does not fetch audio from a link.
How to hear the new level before you replace the file
1. Play the download at the same device volume as the original. 2. If speech still feels quiet, the source may be noise rather than a low voice. This page does not remove noise. 3. Keep the original until the new file sounds right.
Speech is the usual target for a voice note. Music is a bit louder. Quiet is for a bed that should sit behind other audio. Change volume is the page for a percent. Nothing is uploaded.