A speech detector decides
A pause is found by whether anybody is speaking, not by how loud the moment is. A whispered line survives; a silent four seconds does not.
Gaps found by a speech detector rather than a volume threshold, so a quietly delivered sentence is never mistaken for a pause. Set how tight you want the pacing and it holds it across the whole file.
A pause is found by whether anybody is speaking, not by how loud the moment is. A whispered line survives; a silent four seconds does not.
Tighten the gaps to the rhythm you want and it stays consistent from the first minute to the last, which hand-trimming never quite does.
Speech boundaries come from the detector, so the trim happens between words rather than through one.
The classic "strip silence" tool measures level. Anything under the threshold goes. That works on a loud, close-mic'd read and fails on everything else: the end of a sentence trails off, a thoughtful line is delivered softly, and both get cut.
Deciding by speech instead of by volume removes the guesswork, because the question being asked is the right one — is anyone talking here?
Dead air is the most tedious editing there is: hundreds of two-second gaps, each individually trivial, together an afternoon. Automating it is the difference between editing a weekly episode and dreading one.
Yes — you set how tight the pacing should be, rather than removing every gap outright.
It works on the file you give it. Silence is silence regardless of who was about to speak next.
Every tool below runs in the same job, on the same file, in the order that actually sounds right.
Remove breath sounds from audio, on your own PC
AI noise removal that runs offline on Windows
LUFS loudness normalisation for podcasts and video
Try it on your own recording
The free tier runs the same models on the same machine, with no account and no licence key. Install it and put a real file through it.