Deep Audio Priors Emerge from Harmonic Convolutional Networks

Please use Safari or Chrome to play the files.

=> Select task:

Fitting Example 1-5 Audio Restoration Audio Restoration From Quantization Sound Separation
Speech Denoising Music Denoising
Index Noisy Clean Our approach Wiener Wavelet DFL DNP+lsa DNP+wiener WaveUet Regular Dilated Our approach(shallower) Our approach(deeper)
1
2
3
4
5


Data source: Test dataset from LJspeech and MUSIC .