The Discrete Element Method (DEM) captures the collective behavior of a granular material by tracking the kinematics of the constituent grains. DEM simulations using measured particle configurations as input, e.g., from X-ray imaging, allows for a micromechanical interpretation on the acoustic response of a given granular system. This talk covers the fundamentals of DEM and the applications to wave propagation in dry and saturated granular media. For the calibration of DEM models, we use an iterative Bayesian filtering approach to infer the posterior probability distribution of grain properties, conditioned to experimental data. Once calibrated, time and frequency domain techniques can be applied to investigate the acoustic properties such as dispersion relations. The numerical modeling of wave propagation in saturated granular media requires a coupling of hydrodynamics with particle motion. The coupling scheme is briefly presented and its powerful capability shown in the comparison with Biot's analytical solution of wave velocities in saturated granular-elastic media.