A framework for quick testing and comparing multi-channel speech enhancement and separation methods, such as DSB, MVDR, LCMV, GEVD beamforming and ICA, FastICA, IVA, AuxIVA, OverIVA, ILRMA, FastMNMF.
141
stars
34
forks
source link
the reason FastMNMF2 is not good ---- Remove the sqrt() #2
https://github.com/KyleZhang1118/Voice-Separation-and-Enhancement/blob/4e98a2a6cb7d860cabe0b4cb2c2c8c122a11bb52/Process_FastMNMF2.m#L105