How to normalize sounds for a neural network?

Question

Python

Hello.
I am making a neural network, and I need to distinguish between 2 types of sounds in it.
But these 2 types of sounds are of different lengths, and I can't figure out how to normalize them so that they can enter the neural network normally. For example, one sound lasts 2 seconds, another 3, the third 1.5
If anyone has experience, articles, or ideas about this, I will be glad to hear everything ;)

Answer 1

So if they are by default of different lengths - why is there an ANN? Distinguish by length :-)
But in fact - it might be worth looking at the frequency characteristics of sounds. And compare them. In the general case - here it is necessary to build on how these sounds differ sobsno