A real time text-to-speech system designed for italian language is described in this paper. The system is based on synthesis by segment approach. The various components of the system are described in terms of theoretical and implementation aspects. The real time is achieved by using a PC-based board expressely designed for implementing speech processing algorithms.
