Most silence removers ask for a threshold in dBFS, which means you have to guess a number that suits this one recording: too high and the room tone survives, too low and the whole quiet take is deleted. This measures how loud your recording actually is, then treats anything a chosen distance below that as silence — so the same setting behaves the same way on a whispered voice memo and on a loud interview. Gaps shorter than the minimum are left alone because they are the rhythm of speech, a short margin of each pause is kept so words do not end abruptly, and every join is crossfaded so the edit does not click.
Upload
Adjust & Preview
Pixlane is loading on your device. Supported file contents are not uploaded for processing. Site analytics and the assistant are separate; tool-specific limits may apply.
Cut the dead air out of a recording using a threshold measured against the material itself, with crossfaded joins.
Most silence removers ask for a threshold in dBFS, which means you have to guess a number that suits this one recording: too high and the room tone survives, too low and the whole quiet take is deleted. This measures how loud your recording actually is, then treats anything a chosen distance below that as silence — so the same setting behaves the same way on a whispered voice memo and on a loud interview. Gaps shorter than the minimum are left alone because they are the rhythm of speech, a short margin of each pause is kept so words do not end abruptly, and every join is crossfaded so the edit does not click.
Because a fixed dBFS threshold only ever suits one recording level. A quiet phone memo sits entirely below a threshold set for a studio take, so the tool deletes everything; a loud one has room tone above it, so nothing is cut. Measuring the programme first and working relative to it makes one setting behave the same on both.
It decides what counts as dead air rather than a breath. Speech is full of short pauses; removing them makes the result sound rushed and unnatural, so anything shorter than this is left in place.
No. Butting two pieces of audio together leaves a step in the waveform, which is heard as a click. Each join is crossfaded with an equal-power curve, so the level stays steady through the transition instead of dipping.
Yes — that is the common case. The result panel reports how much was removed and how many gaps were closed, so you can adjust the settings and run it again if it took out too much or too little.
Supported audio content is not uploaded for processing. Reading, detection, and export run in your browser; site analytics and the assistant are separate.