Wolfram Language

Build a Custom Audio Encoder

The built-in audio NetEncoder can be used as an efficient starting point for custom encoders. This way, you can leverage the capabilities of the encoders and add custom functionality on top of that.

Create an encoder that partitions the conformed audio data. This is a very simple operation, but incredibly useful for transforming one-dimensional audio data into a two-dimensional signal so that it can be used more effectively as input to a convolutional layer.

Apply the encoder to a recording of a bird.

Plot the computed features.

Create a function encoder that standardizes the computed power spectrum of the training inputs according to a pre-determined distribution.

Plot the features computed with the encoder.

Related Examples

de es fr ja pt-br zh