Methods of processing audio data arrays using Natural Language Processing

Коршун, Наталія Володимирівна and Крючкова, Лариса Петрівна and Соколов, Володимир Юрійович and Киричок, Роман Васильович (2024) Methods of processing audio data arrays using Natural Language Processing Телекомунікаційні та інформаційні технології (3). pp. 33-53. ISSN 2412-4338

[thumbnail of N_Korshun_L_Kryuchkova_V_Sokolov_R_Kyrychok_TIT_3_2024.pdf] Text
N_Korshun_L_Kryuchkova_V_Sokolov_R_Kyrychok_TIT_3_2024.pdf - Published Version

Download (506kB)

Abstract

The article is devoted to a comprehensive study of audio data processing methods using modern Natural Language Processing (NLP) technologies. It highlights important aspects of the development of this innovative field, in particular, focuses on the use of NLP to convert natural language instructions into executable code. This is achieved through the implementation of deep learning methods, semantic analysis, and compilation approaches that automate the process of generating software based on textual queries. An important part of the article is an overview of various approaches to audio data processing, including the selection of programs based on execution and semantic rules. These approaches significantly increase the accuracy and efficiency of code generation, which in turn leads to the creation of reliable and efficient systems capable of working effectively with large data sets. One of the central aspects discussed in this article is transfer learning. This method allows to increase the accuracy of audio data analysis, especially in highly specialized fields such as medicine or law. Transfer learning also reduces the need for large datasets for each specific task, which greatly facilitates the work with large volumes of audio arrays. In addition, the article emphasizes the importance of text preprocessing, which includes such steps as stop word removal, tokenization, and lemmatization. These processes allow to effectively structure the text for further analysis and reduce the likelihood of data processing errors. In general, NLP technologies are seen as a critical tool for processing audio data and large information arrays. Their application can have a huge impact in various industries, including business, medicine, information technology, and many other areas where data processing efficiency is key to decision making and process optimization. Thus, the article emphasizes the relevance and prospects for further development of NLP technologies in the context of audio data processing, which opens up new horizons for innovative solutions.

Item Type: Article
Uncontrolled Keywords: Natural Language Processing; audio data; authentication; code generation; security; data; deep learning; machine learning; text processing
Subjects: Статті у періодичних виданнях > Фахові (входять до переліку фахових, затверджений МОН)
Divisions: Факультет інформаційних технологій та математики > Кафедра інформаційної та кібернетичної безпеки ім. професора Володимира Бурячка
Depositing User: Наталія Володимирівна Коршун
Date Deposited: 07 Oct 2024 09:00
Last Modified: 07 Oct 2024 09:00
URI: https://elibrary.kubg.edu.ua/id/eprint/49790

Actions (login required)

View Item View Item