Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 3 Details of the joint training experiment of S-LTSS Net

From: Lightweight target speaker separation network based on joint training

Network

Loss function

Parameters of network

Joint training or not

S-LTSS Net-1

Combinative loss

0.55M

N

S-LTSS Net-2

Combinative loss

0.55M

Y

S-LTSS Net-3

Joint Loss-1

0.55M

Y

S-LTSS Net-4

Joint Loss-2

0.55M

Y

Back to article page