OPTIMIZATION OF SPEECH SYNTHESIS SOFTWARE FOR VOCAL COMMUNICATION

SEVERELY SPEECH-IMPAIRED INDIVIDUALS NEED AFFORDABLE COMMUNICATION DEVICES WITH NATURAL-SOUNDING SYNTHETIC VOICES THAT ARE APPROPRIATE IN TERMS OF SEX, AGE, NATIONALITY, AND DIALECT. THE CREATION OF SUCH VOICES HAS BEEN HAMPERED BY THE HIGH COST OF DEVELOPING SPEECH SYNTHESIS SOFTWARE AND ADDING NEW VOICES, PARTICULARLY DIALECTS. WE HAVE BEEN WORKING TO REDUCE THESE DEVELOPMENT COSTS VIA (1) A NOVEL MODULAR APPROACH TO THE DEVELOPMENT OF SYNTHESIS RULES FOR MULTIPLE LANGUAGES AND DIALECTS, AND (2) THE DELTA SYSTEM, A SOFTWARE SYSTEM FOR SPEECH SYNTHESIS THAT MAKES THE MODULAR RULES POSSIBLE. WE WILL ENHANCE THE DELTA SYSTEM SO THAT SYNTHESIS PROGRAMS GENERATED WITH IT, INCLUDING THE MODULAR RULES FOR SEVEN AMERICAN ENGLISH DIALECTS AND OTHER VOICES TO BE DEVELOPED WILL BE OPTIMIZED TO RUN ON LESS POWERFUL COMPUTERS OF THE SORT REQUIRED IN COMPACT, AFFORDABLE COMMUNICATION DEVICES. THE RESULT WILL BE A WIDE VARIETY OF HIGH-QUALITY SYNTHETIC VOICES FOR VOCAL COMMUNICATION AIDS AT AFFORDABLE PRICES.