Dog Language or Barking Classification task using Deep Learning

The BiLSTM_FCN model consists of the following components:

Fully Convolutional Block: A series of 1D convolutional layers followed by batch normalization and global average pooling to extract spatial features.

BiLSTM Block: A Bidirectional LSTM layer to capture temporal dependencies in the input sequence.

Concatenation: Outputs from the Fully Convolutional Block and the BiLSTM Block are concatenated.

Softmax Classification Layer: A dense layer with softmax activation for classification.

The model can be customized with the following parameters:

num_classes: Number of output classes.

num_conv_layers: Number of convolutional layers in the FCN block.

num_filters: Number of filters in each convolutional layer.

kernel_size: Size of the convolutional kernel.

lstm_units: Number of units in the LSTM layer.

The following code demonstrates how to instantiate and train the BiLSTM_FCN model: