The Voice of the Steppe: How the Buryat Language Is Entering Cyberspace
The national language has become available through a Yandex service, making it possible to translate texts from Buryat into more than 100 other languages.
Literary Buryat
“Yandex aims to ensure that all spelling rules and lexical compatibility are respected. Once we reach one million sentences, artificial intelligence will be able to learn the Buryat language. At that point, it will gain a form of permanence,” said Bair Baldanov, director of the Belig Republican Center. On the path toward this ambitious goal, the Center’s specialists completed a critical phase of work by translating and editing three datasets into Buryat. This marked an important step toward building a full-fledged digital infrastructure for the language, opening the door to a wide range of new technological opportunities. The project is not limited to text translation. It is designed to create working tools for AI-driven tasks such as machine translation, speech recognition, and educational applications.
Large-scale technical work was carried out in cooperation with Yandex. On Buryat Language Day, October 24, 2025, a milestone was reached – Buryat was added to Yandex Translate. This made the language available for translation into more than 100 other languages. The current implementation focuses on the standardized “literary” variant of Buryat used in the Republic of Buryatia, which may feel unfamiliar to speakers of regional dialects. Even so, the launch represents a major step forward. At the same time, Yandex Translate remains in beta, and full-scale performance will require hundreds of thousands of additional phrases, meaning further sustained effort lies ahead.
Saving Languages From the Red List
The translated datasets are only the beginning. Over time, they can serve as the foundation for an entire ecosystem of digital solutions, including translation tools, learning interfaces, and applications that support the Buryat language. Plans include the development of training courses and educational materials for schools, creating new opportunities for Buryat-language instruction.
However, for the project to reach maturity, several technical and cultural challenges remain. Among them is the need to develop a transliteration system for users unfamiliar with Cyrillic, as well as to improve translation quality, which still struggles to capture the full nuance of Buryat’s many dialects. In parallel, work continues on voice technologies, which would allow Buryat to be integrated into mobile apps and virtual assistant systems.
Notably, the development of NLP modules for the Buryat language is attracting attention beyond the local context. In recent years, the number of projects focused on endangered languages has grown worldwide. These initiatives demonstrate how language digitalization can become a powerful instrument for preservation and long-term development.
Millions of Sentences, Billions of Words
To understand the scale of the current effort, it is worth recalling several key milestones from recent years. In 2024, work began on building a Buryat language corpus for integration into Yandex services, laying the groundwork for machine translation and other digital platforms. The corpus reached a volume of 2,112.97 megabytes, representing more than two million Buryat-language sentences. Another major direction has been the development of digital educational resources in native languages, including school curricula and online courses.
Special attention is being paid to translating textbooks into Buryat with sensitivity to cultural context. In 2026, work is expected to begin on translating textbooks in mathematics, environmental studies, physical education, and other subjects, enabling Buryat to be incorporated directly into the core educational process.
“A People Without a Language Is Not a People”
“While visiting the Chukotka Autonomous Okrug, Russian President Vladimir Vladimirovich Putin spoke about the importance of preserving and developing the native languages of all the peoples of Russia. Our work in Buryatia fully aligns with this goal. A people without a language is not a people, and a language without a people is not a language,” emphasizes Belig Center head Bair Baldanov.
The development of the Buryat language in the digital environment is a multi-stage process. Continued work on transliteration for international users and improvements in translation quality, accounting for dialect diversity, remain critical priorities. This is why one of the main challenges lies in the need for fundamental linguistic research. Only a deep scientific approach to studying dialects, grammar, and pronunciation can ensure that the language becomes not just digitally present, but digitally rich and expressive.
