A real model, running locally
Full-band speech separation: the same class of published model the cloud services build on, running on the processor you already own.
Traffic, air conditioning, computer fans and room reverb pulled out from under the voice by a neural de-noiser — not by a gate that silences the quiet parts and leaves the loud ones untouched.
Full-band speech separation: the same class of published model the cloud services build on, running on the processor you already own.
A gate mutes audio below a threshold. The noise never goes away — it simply pumps in and out under the speech, which is the sound people describe as "over-processed".
The enhance pass restores body and clarity afterwards, so a cleaned-up laptop microphone sounds like a recording rather than a survivor of one.
Most of it is the room. Air conditioning, a fan, a fridge two doors away, and the reflections off every hard surface between you and the microphone. None of it is loud, and all of it is constant, which is why the ear stops noticing it while recording and cannot stop noticing it on playback.
A model trained to separate speech from everything else removes the floor rather than hiding it. That is why the result stays clean when the speaker pauses — the moment a gated recording gives itself away.
Run separately and in the wrong order, cleanup tools fight each other: a de-noiser chasing a breath the silence pass already cut, a loudness stage normalising to a floor about to be removed.
So they do not run separately. One press runs the lot as a single job, the audio stays lossless the whole way through, and it is encoded once at the end.
That is the case it helps most. Pair it with the enhance pass, which restores the body a small microphone never captured.
No. It runs on the CPU of an ordinary Windows 10 or 11 machine.
No. The free tier gives 2 hours of processing a month with no account, and a licence adds hours. Nothing renews itself.
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
Remove silence and dead air from audio automatically
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.