Clean up voice recordings and videos in your browser. Remove background noise with a neural network, then trim, level, retime and fade. Your files never leave this device.
RNNoise, a recurrent neural network trained on speech, removes fans, hum, traffic and room noise. It's made for voices; it also removes music and tones.
Drag across the Before waveform to select a region, or type exact times.
Normalize sets the loudest peak to a target level. Gain turns everything up or down by a fixed amount.
Change tempo and pitch independently. Speed keeps the voice's pitch; Pitch keeps the timing.
Ramp the start up from silence and the end down to silence.
Choose channels and the sample rate for the downloaded WAV. Load a video file to pull out its soundtrack.
Apply keeps a change and makes it the new Before, so you can chain tools: remove noise, trim, normalize, then download. Undo steps back. Downloads are 16-bit WAV.