Text-to-speech systems typically include an automatic language processing module for input text, which allows you to extract the language information needed for the speech generation process. The development of these modules requires the incorporation and formalization of linguistic knowledge (mainly phonetic and phonological, but also orthographic, lexical, morphological and syntactic) often in the form of rules and dictionaries. It is a job in which engineers and linguists usually work together.
In this line, I have worked on different projects that have involved the development or improvement of different language processing modules for text-to-speech conversion.
Incorporation of English into the language processing module of Telefónica Investigación y Desarrollo, a project carried out in collaboration with the Voice and Language Group of the Barcelona Media Foundation.
Dvelopment of the linguistic processing modules in Spanish and Catalan for Cerevoice, the text-to-speech engine of Cereproc
Development of TextAFon, its own language processing module developed jointly by the Voice and Language Group of the Barcelona Media Foundation.