The goal of speech generation is to generate an oral message in response to the sentence spoken by the user. Two processes are usually used for this. The first is intended to create phrases in text format, while the second is responsible for transforming these phrases into voice format.