Skip to main content

Table 9 Overall average SDR improvements (dB) in three-source case for the best parameter settings

From: Generalized independent low-rank matrix analysis using heavy-tailed distributions for blind source separation

Source

ILRMA

GGD-ILRMA

t-ILRMA

GGD-ILRMA w/ tempering

t-ILRMA w/ tempering

Music

1.76

3.24 (β = 1.94, p = 0.5)

3.19 (ν = 300, p = 0.5)

3.36 (β = 1.82, p = 1.0)

3.29 (ν = 1, p = 1.0)

Speech

2.79

3.14 (β = 1.94, p = 1.0)

2.94 (ν = 1000, p = 1.0)

3.32 (β = 1.40, p = 0.5)

3.22 (ν = 10, p = 2.0)