How to remove background noise from a voice recording
Traffic, a busy room, a laptop fan or a hum from the cable: Tailwave has two ways to take noise out of a recording, and both run in your browser.
For speech: voice noise reduction
The quickest way for podcasts, interviews, voice memos and video calls. A small neural network trained on speech keeps the voice and removes the rest; you don't need to tell it what the noise sounds like.
- Open the editor and drop your recording onto the page.
- Select the speech: press Ctrl+A to select everything.
- Choose Keep at 100%, or lower it if the voice starts to sound thin. Press to listen first.
- Apply it, then export with .
Made for voices: on music it can take detail away along with the noise.
For steady hum and hiss: noise reduction with a profile
Best for a constant sound (hiss, hum, air conditioning) under music or speech. You first show Tailwave what the noise alone sounds like.
- Find a moment of noise only, such as the pause before someone speaks, and select one or two seconds of it.
- Choose .
- Select the whole recording (Ctrl+A) and choose
- Set how much to reduce: 12 dB is a good start. Raise it for more cleaning; if the sound turns watery, go lower or raise . Preview, then apply.
Finishing touches
- silences the quiet gaps between phrases.
- with a high-pass around 80 Hz removes rumble from handling and wind.
- brings the cleaned voice to a steady loudness: see make audio louder.
Your recording stays with you. The noise reduction runs on your computer, not on a server.