The report is concerned with the experimental study of the idiosyncrasy of utterance-into-phrase segmentation observed in the speech of a popular Russian TV-anchorman and two TV-news readers. Comparative statistical estimation of relative frequencies of occurrence of pauses of various duration, frequencies of occurrence of phrases and pairs of phrases with a different number of accent units were computed, as well as frequencies of occurrence of phrase break between different consecutive parts of speech. On the basis of the results obtained a stochastic algorithm of personalized utterance-into-phrase segmentation is developed. The algorithm is intended to be implemented to the system of individual voice cloning using a text-to-speech synthesis.