espnet2.enh.layers.bsrnn.ChannelFreqwiseLayerNorm

source

class espnet2.enh.layers.bsrnn.ChannelFreqwiseLayerNorm(channel_size, shape='BDTF')

Bases: Module

Channel-and-Frequency-wise Layer Normalization (cfLN).

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(y)

Forward.

• Parameters:y – [M, N, T, K], M is batch size, N is channel size, T and K are lengths
• Returns: [M, N, T, K]
• Return type: gLN_y

reset_parameters()