Thursday30 January 2025
nurtoday.com

A new AI-based model for learning the Kazakh language, called "Tilqazyna," has been introduced.

Currently, the model is capable of performing tasks in various areas of the Kazakh language, including vocabulary, morphology, and semantics, as reported by Kazpravda.kz.
Представлена модель обучения казахскому языку «Tilqazyna», разработанная с использованием технологий искусственного интеллекта.

The National Scientific and Practical Center "Tilqazyna," under the Committee of Language Policy of the Ministry of Science and Higher Education, has introduced the initial results of the "Tilqazyna" Kazakh language learning model based on artificial intelligence.

Currently, the model is capable of performing tasks in various areas of the Kazakh language, including vocabulary, morphology, semantics, and more. Specifically, it can generate text in Kazakh, create paraphrases, work with context, summarize texts, correct grammatical and punctuation errors, clarify the meanings of idioms, and translate terms.

This sector-specific LLM model has already been uploaded to the Hugging Face platform and is accessible to all users. Utilizing this model will enable the development of numerous IT products in the Kazakh language using artificial intelligence. This aligns with President Kassym-Jomart Tokayev's message, which emphasized the importance of transforming Kazakhstan into a country that widely employs artificial intelligence and advances digital technologies.

In developing the model, experts from the "Tilqazyna" center employed natural language processing algorithms and conducted targeted analyses of large datasets.

This year, a voice communication feature will be added to the model, along with a user-friendly interface. The project will also be able to teach the Kazakh language at levels A1, A2, and B1, with plans to expand to levels B2 and C1 by 2026.

The ultimate outcome of the project will be a voice assistant capable of creating a personalized Kazakh language learning program based on the user's language proficiency level. It will be presented as a mobile application for both iOS and Android systems.