Text this: Sequential modeling, generative recurrent neural networks, and their applications to audio