期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2018
卷号：96
期号：23
出版社：Journal of Theoretical and Applied
摘要：The article is dedicated to the development of an algorithm of autonomous speech segmentation of the given text using dynamic programming method and creation of base unit bases of a certain announcer. Base units are used to form speech-like signals in systems of active speech information protection. The algorithm is based on the use of text, spoken by one of the announcers, marked and divided manually into structural units, which is the model for an automated speech segmentation of other announcers for the same text. For the speech segmentation of other announcers and formation of structural base units of their speech and to be used with the method of dynamic programming according to the model as done manually before that. The formation of speech-like signals is conducted using compilation method of speech synthesis of the text, which is automatically synthesized including linguistic, phonetic and prosodic features of the given language.
