Embedded Machine Learning (TinyML) model to build a system that can recognize audible events, particularly voice/speech.