Towards audio language modeling - an overview