Back to post
Demo · libdse

Speech Enhancement using Deep Learning

Upload a noisy speech recording, select a model, and listen to the enhanced result. Spectrograms are rendered in real-time directly in the browser.

Wave U-Net is currently untrained. Selecting it will produce degraded output.


Drop your noisy audio file here, or click to browse

WAV · MP3 · FLAC · OGG and other common formats

or
api.libdse.nikpau.io · LibDSE · Read the post →