Audio Processing With Tensorflow Vs Pytorch