Laura Orynbay

Ms. Laura Orynbay

Artificial Intelligence Technologies, L.N. Gumilyov Eurasian National University, Nur-Sultan, Astana, Kazakhstan
orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 April 2024 in Electronics
Verified
Journal Article
Open access
PDF file
LOLaura Orynbay
Laura Orynbay
BRBibigul Razakhova
Bibigul Razakhova
Peter PeerPeter Peer
Peter Peer
BMBlaz Meden
Blaz Meden
Žiga EmeršičŽiga Emeršič
Žiga Emeršič
In recent years, there has been increasing interest in the conversion of images into audio descriptions. This is a field that lies at the intersection of Computer Vision (CV) and Natural Language Processing (NLP), and it involves various tasks, including creating textual descriptions of images and converting them directly into auditory representations. Another aspect of this field is the synthesis of natural speech from text. This has significant potential to improve accessibility, user experience, and the applications of Artificial Intelligence (AI). In this article, we reviewed a wide range of image-to-audio conversion techniques. Various aspects of image captioning, speech synthesis, and direct image-to-speech conversion have been explored, from fundamental encoder–decoder architectures to more advanced methods such as transformers and adversarial learning. Although the focus of this review is on synthesizing audio descriptions from visual data, the reverse task of creating visual content from natural language descriptions is also covered. This study provides a comprehensive overview of the techniques and methodologies used in these fields and highlights the strengths and weaknesses of each approach. The study emphasizes the importance of various datasets, such as MS COCO, LibriTTS, and VizWiz Captions, which play a critical role in training models, evaluating them, promoting inclusivity, and solving real-world problems. The implications for the future suggest the potential of generating more natural and contextualized audio descriptions, whereas direct image-to-speech tasks provide opportunities for intuitive auditory representations of visual content.

15 November 2024
Verified
Conference Paper
Closed access
MYManas Yergesh
Manas Yergesh
ASAltynbek Sharipbay
Altynbek Sharipbay
ESEvren S. Sadi
Evren S. Sadi
BYBanu Zh Yergesh
Banu Zh Yergesh
LOLaura Orynbay
Laura Orynbay
KMKundyz Maxutova
Kundyz Maxutova
The article presents a system for the classification of Kazakh language documents based on subject areas, employing techniques such as tokenization, stemming, and term-matching. The system, named “Kazakh Language Document Classifier by Subject Area,” targets documents in education, employment, and medicine, aiming to address the challenges of document management and information retrieval in multilingual settings. The process begins with document preprocessing, including morphological analysis of Kazakh text, followed by the automatic identification of keywords using N-grams, TF-IDF, and syntactic tags. By leveraging databases of subject-specific terms, the system organizes documents into relevant categories, enhancing retrieval efficiency. The implementation also integrates the Apache POI library for handling MS Office Word documents. The classifier's accuracy is evaluated, showing initial success in categorizing documents. Further development aims to improve accuracy through data expansion and the establishment of classification thresholds, contributing to advancements in natural language processing for Kazakh. This system holds promise for educational applications and cross-disciplinary document analysis.

15 November 2024
Verified
Conference Paper
Closed access
ASAyaulym Sairanbekova
Ayaulym Sairanbekova
MAMamyr Altaibek
Mamyr Altaibek
ASAltynbek Sharipbay
Altynbek Sharipbay
GBGulmira Bekmanova
Gulmira Bekmanova
LOLaura Orynbay
Laura Orynbay
This article delves into the application of fuzzy logic for analyzing political discourse in the Kazakh language, specifically addressing the inherent ambiguity and uncertainty in political statements. By employing fuzzy logic, the study evaluates key rhetorical tactics such as accusation, manipulation, and criticism within the framework of Kazakh political discourse. The methodology incorporates fuzzy logic operations, including NOT, AND, and OR, to methodically assess the truthfulness and reliability of political statements, supplemented by mathematical modeling that provides practical examples from contemporary Kazakh political texts. The findings illuminate how fuzzy logic significantly enhances the comprehension of the complexities involved in political rhetoric. Additionally, the article outlines potential future research directions, such as comprehensive media analysis and public opinion studies. It underscores the significance of employing implicit logic to facilitate a more subtle and precise analysis of political discourse, thereby contributing valuable insights into the dynamics of political communication.

14 December 2024
Verified
Conference Paper
Closed access
GBGulmira Bekmanova
Gulmira Bekmanova
BYBanu Zh Yergesh
Banu Zh Yergesh
AOAssel Omarbekova
Assel Omarbekova
Gulsipat AbishevaGulsipat Abisheva
Gulsipat Abisheva
Tolegen AidynovTolegen Aidynov
Tolegen Aidynov
AZAlma Zakirova
Alma Zakirova
The Internet has changed dramatically over the past decade, and search engines have worked hard to adapt to these changes by providing better content to users. Many efforts have been made to assess document quality, including attempts to use semantic analysis to extract meaningful information from material presented on the Internet. This article describes the development of a text editor that uses morphological and syntactic analyzers to process website content. The main purpose of the system is to automatically mark and improve the text content of web pages by integrating semantic information obtained based on text structure and meaning analysis. The system uses preprocessing technology and deep NLP models (such as BERT), combined with a self-built Kazakh tree library (ConLL-U format) to restore the query's lemmatization. Optimize the matching of queries and documents through TF-IDF, Word2Vec and query expansion technology. Experiments were conducted on an autonomously collected Kazakh news dataset. Experimental results show that morphological analysis significantly improves retrieval accuracy during the text normalization process, while the deep learning model further enhances the ranking effect based on semantic correlation. This research provides new technical means for content optimization and information retrieval in low-resource languages, and helps improve the performance of search engines in multi-language environments.