cjwbw/demucs

Demucs Music Source Separation

Input
Configure the inputs for the AI model.

Only separate audio into the chosen stem and others (no_stem).

Input audio

0
100

Number of random shifts for equivariant stabilization.Increase separation time but improves quality for Demucs. 10 was used in the original paper

Save wav output as float32 (2x bigger).

0
100

Overlap between the splits.

Strategy for avoiding clipping: rescaling entire signal if necessary (rescale) or hard clipping (clamp).

Choose a model

0
100

Bitrate of converted mp3

Choose the output format

Output
The generated output will appear here.

No output yet

Click "Generate" to create an output.