Article View

Segmental-Prosodic elements in any language’s speech synthesis involves an automated procedure for blending phonetically structured meaningful units like melody, sound intensity, phoneme duration, and pause duration based. Each of these units varies in size and degree of completeness. This study aimed to examine the segmental-prosodic foundations of a database for synthesizing Kazakh speech. Using a qualitative approach, the data was collected through documentation research and content analysis of texts. The sampled texts continued elements of semantic segmentation and exemplified how an utterance can play a significant role in segmenting the speech streams. Results reveal that segmental-prosodic elements serve to unite parts of a phrase into a cohesive whole and function as meaning-differentiating tools, facilitating the clear and unambiguous understanding of text content. The boundaries of semantic-intonational and syntactic units—syntagms—can also be conveyed through syntagmatic stress. As the ultimate element of phrasing, the syntagm reflects syntactic-stylistic and lexical-phraseological segmentation of the speech stream. The boundaries of syntagms in speech are relatively fluid and are often reflected in punctuation; however, punctuation marks do not always align with syntagm boundaries. Ther study implies that speech signals from printed texts based on segmental-prosodic means can be one of the key conditions for developing speech synthesis systems.