From: Explicit-memory multiresolution adaptive framework for speech and music separation
Method
Bass
Drums
Others
Vocals
RGT1 [40]
2.70
3.44
2.63
3.84
JY3 [27]
3.67
4.66
3.40
5.74
UHL2 [41]
5.03
5.92
4.19
5.93
\(L_{1+2-SI}\)
4.35
5.55
3.69
6.42
\(TD_{1+2-SI}\)
4.71
5.95
3.91
6.74