We understand the space around us by interacting with the environment. In that interaction, the hearing system plays a significant role in delivering spatial information to higher circuits in the brain. The literature proposes several computational models targeting specific components of that spatial-hearing process. However, the models’ implementation often relies on specific computational solutions limiting software reuse. In our talk, we describe Bayesian inference as a theoretical tool grounded in probability theory and apply it to models of human sound localization. Within this concept, we consider the extraction of perceptually relevant features, the integration of prior beliefs, and mechanisms of beliefs update. As a result, we propose a software framework based on Bayesian inference to develop computational models of human sound localization. We provide tools such as a flexible model template and functions to estimate the free parameters. In particular, the framework aims to help hearing scientists develop new models and run model- based analyses. We further discuss the advantages and the limitations of the proposed software framework.