Development of methods, algorithms of machine learning and mobile applications for Kazakh speech recognition [Text] : Dis. ... Dr. of Philosophy (PhD) : 6D070400 - Department of Computer Sciences
Faculty of Engineering and Natural Sciences / D. N. Kuanyshbay ; scientific advisors: Y. N. Amirgaliyev, T. Shevchenko, Y. V. Krak ; Ministry of Education and Science of the Republic of Kazakhstan ; Suleyman Demirel University... – Kaskelen : [s. n.], 2021. - 114 p., tabl., diagr.;+ 1 electronic optical disk (CD-ROM) . . – Bibliogr.: p. 97-105 . – [n. c.] (hardcover) : 0 tg.
The purpose of this work is to build a correct ASR system avoiding the problem of data shortage, provide a well-designed environment for collecting speech data, collect a significant amount of speech data with transcription for the Kazakh language, apply post-processing techniques to the collected data and structure files for neural network operation, apply a multilingual technique with Russian using the transfer learning method
speech signal processing, speech enhancement, types of noise, speech enhancement techniques, analysis of existing ASR systems, language models in ASR, speech corpora in kazakh language
Development of end-to-end system for automatic recognition of speech in agglutinative languages [Electronic resource] : Thesis for the Degree of Doctor of philosophy (PhD) : 8D06103 - Management information systems / A. Bekarystankyzy ; scientific consultants: O. Mamyrbayev, M. Mateus ; Satbayev University Institute of Automation and Information Technologies... – Алматы : [s. n.], 2023. - 107 p., graphs, diagrams;+ 1 electron. opt. disk (CD-ROM) . . – Bibliogr.: p. 86-96 . : 0 tg.
The present dissertation was developed with the aim of studying the ways of improving ASR performance for agglutinative languages on the example languages from Turkic family