Bakhtiyor Akmuradov et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 4657 – 4664 4657 ABSTRACT Demand for multimedia systems, including speech synthesis systems, is growing rapidly to simplify and facilitate the use of rapidly evolving modern technologies. Designing text-to-speech systems capable of producing natural-sounding speech segments in the Uzbek language is a challenging and open problem. In this paper, we propose an algorithm based on concatenative methods for dividing Uzbek language words into syllables for text-to-speech synthesis. First, the electronic dictionary of the Uzbek word and syllables was developed based on the proposed algorithm. Then, the structure and characteristics of these words and syllables were analyzed. This electronic dictionary has more than 31.5 thousand words and 3 thousand syllables. Key words : Concatenative, Electronic Dictionary, Phonetics, Sounds, Syllables, Text-to-Speech Synthesis, Uzbek language, Words. 1. INTRODUCTION The invention of new technology makes life more comfortable for people, however, numerous problems appeared related to human-machine interaction, storage, and data transmission. History has confirmed that voice communication is essential for social communication because it is a simple and useful system of communicating. Therefore, from the earliest times of computer technology, efforts have been made to train computers on how to communicate with humans utilizing a natural speech interface. Several powerful methods and algorithms have been developed in this field over the years. However, the creation of the Uzbek language speech synthesizer remains an open problem. Speech synthesis is the method of artificially generating human speech. A machine operation applied for this goal is described as a speech synthesizer and can be utilized in hardware and software. Speech synthesis, more specifically identified as text-to-speech (TTS), is a complete technology that incorporates various disciplines such as digital signal processing, sentiment text classification [1], statistics, linguistics, and acoustics. It is cutting-edge technology in the area of data processing, especially for the advanced smart speech interaction systems. The principal task is to transform text input into speech output [2]. Nowadays, there are many projects for industrial TTS systems with various characteristics and performance. Therefore, it seems that a comparative study of them would be quite useful for signal processing researchers, among others. With the advancement of digital signal processing technologies, the research purpose of speech synthesis and text analysis has been developing from intelligibility and clarity to naturalness and expressiveness. Intelligibility describes the accuracy of the synthesized speech, whereas naturalness refers to the ease of listening and global stylistic consistency [3], [4]. We propose an algorithm for dividing words into syllables for the Uzbek language text-to-speech synthesizer that uses the concatenative method. The main contributions of the proposed method are summarized as:  The first Uzbek language electronic dictionary that has more than 31.500 words and 3.000 syllables is created.  Uzbek language words are divided into syllables using the proposed algorithm for concatenative Uzbek language text-to-speech synthesizer. The rest of the article is organized as follows. Section 2 gives an overview of speech synthesis including its basic concept, history, and technologies. In Section 3, this paper discussed the methods of speech synthesis. Brief information is given in Section 4 about phonetic units of speech in the Uzbek language. In Section 5, the Uzbek language electronic dictionary is developed. The proposed algorithm for dividing Uzbek words into syllables is explained in Section 6. provides discussions on new research directions. Finally, Section 7 concludes the article. 2. LITERATURE REVIEW Several TTS systems are currently available for application, but in this section, we are going to describe some of them. There are various articles where the most appropriate TTS systems are listed [5], [6]. The research work of A. Kaliyev et A Novel Algorithm for Dividing Uzbek Language Words into Syllables for Concatenative Text-to-Speech Synthesizer Bakhtiyor Akmuradov 1 , Utkir Khamdamov 2 , Mukhriddin Mukhiddinov 3 , Elbek Zarmasov 4 1,2,3,4 Department of Hardware and Software of Control Systems in Telecommunications, Tashkent University of Information Technologies named after Muhammad al-Kwarizmi, Tashkent, Uzbekistan. E-mails: b.u.akmuradov@gmail.com (B.A.), utkir.hamdamov@mail.ru (U.K.), mmuhriddinm@gmail.com (M.M.), zarmasov.elbek@mail.ru (E.Z.) ISSN 2278-3091 Volume 9, No.4, July – August 2020 International Journal of Advanced Trends in Computer Science and Engineering Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse67942020.pdf https://doi.org/10.30534/ijatcse/2020/67942020