The dramatic piece reader for the blind and visually impaired

摘要

The paper presents the concept and realization of the intelligent audio-book reader for the visually impaired. The system is capable of presenting personalities of different characters. The synthesizer mimics the way how a puppeteer portrays different characters. A traditional puppeteer generally uses up to a dozen different marionettes in one piece. Each of them impersonates a character with its own typical voice manifestation. We studied the techniques the puppeteer uses to change his voice and the acoustical correlates of these changes. The results are used to predict appropriate settings of the parameters of the voice for every character of the piece. The information on the personality features of every particular character is inserted manually by a human operator. Similarly to the puppeteer's show only one speaker's voice is used in this concept and all the modifications are made using speech synthesis methods.