AI-DRIVEN TEXT AND AUDIO SYNTHESIS FOR LOW-RESOURCE LANGUAGE DATASETS

Aidana Karibayeva; Balzhan Abduali

doi:10.26577/jpcsit4220268

Authors

Aidana Karibayeva Al-Farabi Kazakh National University, Almaty, Kazakhstan https://orcid.org/0000-0002-2023-1573 (unauthenticated)
Balzhan Abduali Al-Farabi Kazakh National University, Almaty, Kazakhstan https://orcid.org/0000-0003-0140-4181 (unauthenticated)

DOI:

https://doi.org/10.26577/jpcsit4220268

Keywords:

low-resource languages, methodology for creating datasets, parallel text corpora, AI systems, audio synthesis

Abstract

This paper presents an AI-driven methodology for constructing parallel text corpora and synthesizing parallel audio datasets for low-resource Turkic language pairs, namely Kazakh-Kyrgyz and Kazakh-Uzbek. The scarcity of high-quality linguistic and speech resources for these languages poses significant challenges for the development of neural machine translation and automatic speech recognition systems. The paper proposes and validates a methodology for parallel dataset construction driven by artificial intelligence, in which the selection of a translation system is guided by a set of criteria encompassing free accessibility, translation quality, and processing efficiency, assessed through experiments on a 2000-sentence test set. Based on this evaluation, large-scale parallel corpora for the Kazakh–Kyrgyz and Kazakh–Uzbek language pairs were generated using the selected AI system. A subsequent manual error analysis revealed that approximately 3.7% (Kazakh–Kyrgyz) and 4.3% (Kazakh–Uzbek) of the translations contained inaccuracies, indicating the need for post-editing and further model refinement. Audio synthesis experiments using MMS-TTS and TurkicTTS systems demonstrated that synthetic speech of near-natural quality can be generated for both languages, with NISQA scores reaching up to 4.44 for Uzbek. The findings confirm that the proposed methodology provides a practical and scalable foundation for expanding linguistic resources for low-resource Turkic languages and supporting further research in machine translation and speech processing.

Downloads

Download data is not yet available.

Author Biographies

Aidana Karibayeva, Al-Farabi Kazakh National University, Almaty, Kazakhstan

Aidana Karibayeva, PhD, is an Acting Associate Professor at the Department of Information Systems of al-Farabi Kazakh National University (Almaty, Kazakhstan; karibayeva.aidana@kaznu.edu.kz). Dr. Karibayeva has more than 13 years of experience in artificial intelligence and machine learning. Her research interests include neural networks, deep learning applications, data mining, and speech recognition.
Balzhan Abduali, Al-Farabi Kazakh National University, Almaty, Kazakhstan

Balzhan Abduali is Master of Engineering Science, Senior Lecturer at the Department of Artificial Intelligence and Big Data of al-Farabi Kazakh National University (Almaty, Kazakhstan; abduali.balzhan@kaznu.kz). She has more than 12 years of experience in machine learning. Her research interests include artificial intelligence, machine learning, and data mining.