How to feed the input to a Memory Augmented Neural Network (MAAN) to do one shot learning?

In this paper by Deep-Mind on one shot learning they have published an architecture explaining how the system works with an external meory. I understand the mechanism perfectly. But what I don't understand is how they feed the data in to the algorithm.

In this part of the paper they explain as follows ,

From this part I actually don't understand is do do the concatenate inputs with labels ? And what is label shuffling ?