"Artistic and musical expression is an important part of human experience, and can be considered a universal language above any cultural diversities. In particular, the communicative power of voice is worldwide known, since voice is the first musical instrument ever exploited to create and perform music. During the last twenty years, novel scientific and technological achievements have given birth to several systems able to generate synthetic speech and singing voices, unleashing the exploration of innovative performative possibilities. Strong of the experience in the field of the involved researchers, the objectives of this project consist in the prosecution of these studies, extending the current “state of the art” with innovative technologies and techniques to enhance naturalness, intelligibility, expressivity and controllability of vocal synthesis, in thoughtful and insightful multimodal performances. The object of the first part of the project is the design and development of a novel musical instrument for excellent vocal synthesis in real-time; the core of the instrument will be an articulatory vocal synthesizer, composed by three parallel algorithms for physical modeling synthesis based on an unprecedented joint bio-mechanical and acoustic simulation of the human upper airways. Synthesis will be controlled by means of hand and finger gestures, making use of machine learning techniques to enhance usability and bandwidth of the instrument. The second part of the project aims at the integration of the instrument in a multimodal platform for audio/visual performances; making use of novel Virtual and Mixed Reality technologies, we want to define choreographies in which, by means of gestures, performers are able to control speech and song synthesis, multichannel sound diffusion, processed images and real-time 2D and stereoscopic projections. It is our intent to exhibit such performances on stages within theaters and concert halls, in Europe and North America."