Hierarchical spike coding of sound

We develop a probabilistic generative model for representing acoustic event structure at multiple scales via a two-stage hierarchy. The first stage consists of a spiking representation which encodes a