Ressources.ircam

The subject of automatic speech synthesis began to be popularised as early as the 1990s. Each of us has already had to deal with automatic answering machine voices that made us all suffer in the beginning. Today, however, the progress made both in terms of language comprehension and the acoustic quality of speech synthesis approaches have helped us make giant leaps forward, and new vocal services are currently seeing their quality and capabilities improve significantly with increasingly more human-sounding and expressive voices.In this presentation, I will briefly review recent advances in speech synthesis. After this introduction, I will discuss topics related to the customization of synthetic voices to the customer's needs; and this on several levels. First, at the level of the main components of oral expression: language, speech style, language register and gender for example. Then, I will address issues at the level of the utterance; prosodic for the most part (pitch and flow manipulation). Finally, I will finish by discussing the subsidiary elements to be taken into consideration in order to best meet the needs of customers and end-users of synthetic voices in our constantly changing world.