Skip to main content

Advertisement

Table 1 Learned values of the prior hyper-parameters

From: Spatial location priors for Gaussian model based reverberant audio source separation

T 60 50 ms 130 ms 250 ms 500 ms
m 2.1 2.1 3.4 5.3
σ 1 2 0.009 0.033 0.068 0.148
σ 2 2 0.002 0.024 0.063 0.139