While many people say they have no rhythm, most humans when listening to music can easily discern and move to a beat. On the other hand, many of us are not so adept at actually generating and maintaining a constant beat over a period of time. Demonstrating a beat is a very complicated task. Among other things, it involves the ability of our brains to estimate time intervals and to make physical movements, for example hitting a drum, in coordination with the time estimates that we make. How the complex system comprised of our brain and body solves this problem is an open and active area of research. In this talk, I will introduce a neuromechanistic model of a beat generator, which is defined here as a group of neurons that can learn to keep a constant beat across a range of frequencies relevant to music. The model is a biophysical manifestation of two different types of models: error/correction and neural entrainment models, both of which will be reviewed. The goal of the talk is not just to introduce a new way of thinking of beat generation, but also to raise a series of questions about the nature of time and the role of perception in our ability to make decisions.