Stefano Sandri

Real-time text processing for Italian speech synthesis

The paper describes the software architecture of an Italian text-to-speech synthesis system based on the joining of LPC coded diphones. The automatic voice response system is designed according to multichannel and real time criteria. For each output channel, the following operations are performed: pre-processing of the input string of characters, translation into the proper sequence of diphones, generation of prosodic contours and real-time control of a hardware speech synthesizer.

Automatic stress assignment for Italian text-to-speech synthesis

This paper describes in detail a set of automatic rules determining the correct position of stress in Italian words. The rules are based on a reverse scanning of the word characters, and consist of a set of simple automata, each for the recognition of a different stress location. The reverse scanning is limited to a maximum of 6 characters. The words making exception are stored in a scheduled area. The performance of the rules has been evaluated by means of a statistical analysis of the Italian vocabulary and some real texts.