The process of using computer algorithms to generate synthetic speech that sounds natural and expressive.