Digital Nomads: Kamchatka Scientists Teach Neural Networks to Speak Koryak
In Russia, researchers have introduced a digital language course called KORYAK TUYU, and it can be downloaded anywhere – from Chukotka in the Far East to Kaliningrad on the Baltic coast.
The Koryak language, spoken today by only a few thousand people, has entered the digital age. KORYAK TUYU – a self-study platform designed to reconnect younger generations with the linguistic heritage of Kamchatka – represents a new attempt to preserve one of Russia’s indigenous languages through educational technology.
Ancestors Now Speak from a Smartphone
What if a language that echoed for centuries across tundra landscapes along the cold Bering Sea suddenly gained a digital form? Until recently it could mostly be heard in reindeer herding camps or in the homes of village elders. Today it exists in RuStore, Russia’s mobile application marketplace. At a congress of native language teachers in Surgut, Kamchatka researchers demonstrated the project in an unusual way. They took out their phones and invited colleagues to download the Koryak self-learning application KORYAK TUYU.
“To ensure a language survives and continues to exist in the modern world under the pressure of unification and globalization, parents need to speak it with their children even before they start school. If a child arrives at school knowing only Russian and then encounters a subject called ‘Native Language,’ that situation creates stress. In practice the language is no longer truly native for that child – it becomes only an ethnic language,” says Andrey Kibrik, Russian linguist and director of the Institute of Linguistics of the Russian Academy of Sciences.
The digital textbook developed by Kamchatka State University named after Vitus Bering is not simply an electronic copy of a printed manual. It is a product of the digital learning environment created by a team of scholars led by philologist Yulia Fayzrakhmanova. Linguists worked alongside native speakers of Koryak and schoolteachers, who helped review the content and ensured the material was understandable and easy to use for learners.
Across Years and Distances
In the past, studying the Koryak language required one of two options: travel to Kamchatka to find a native speaker or work as a professional linguist with access to rare printed dictionaries. Today that barrier is disappearing. The change reflects a broader process that the Arctic Council describes as the digitalization of linguistic and cultural heritage.
The goal goes beyond preventing books from gathering dust on library shelves. The issue concerns a living language. For a language to survive, people must speak it. For people to speak it, the language must remain relevant. And relevance today often means presence in the digital environment where younger generations spend their time. KORYAK TUYU moves the Koryak language into that digital space, giving Kamchatka a new tool to preserve its cultural memory.
Not a Lone Voice in the Wilderness
The Koryak project is not the first attempt to bring indigenous languages into digital ecosystems. The Mansi language entered the digital sphere earlier. In 2024, the Yugra Research Institute of Information Technologies created the Mansi Language Corpus – a database of parallel translations containing about 146,000 sentences. Researchers used it to train a neural machine translation system whose quality was later highly rated by the Institute of Linguistics of the Russian Academy of Sciences. By December the Mansi Language Portal had launched, combining a translator and dictionary in a single platform.
At the same time, major technology companies joined the effort. Within Yandex Translate, in partnership with the House of the Peoples of Russia and the Foundation for the Development of National Cultures, a large program was launched to preserve minority languages. During 2024–2025 the service added several new languages: first Komi and Tuvan, followed by Ossetian, Buryat, Erzya and Moksha. In September 2025 Karachay-Balkar and Kabardian-Circassian were added. Speakers of these languages can now search for information on smartphones as freely as users working in English or German.
In Karelia, researchers took a different approach by creating the TopKar – Toponimiya Karelii (TopKar – Toponymy of Karelia) geoportal. Specialists from the Karelian Research Centre of the Russian Academy of Sciences compiled and visualized nearly 60,000 geographical names across the region. In 2026 the project received the Heritage Prize as the best scientific initiative of the year. Researchers also published a reference guide to disappeared villages, rivers and hayfields – all documented in native languages. The result resembles a time machine, allowing users to see how the same places were named a century ago.
Russia and Norway have also launched an initiative aimed at digitalizing the linguistic and cultural heritage of the Arctic. Supported by the Arctic Council, the project brought together researchers from dozens of countries. Scientists conducted field expeditions and recorded speech from native speakers – in Yakutia alone they documented more than one hundred individuals – in order to build a multimedia portal and interactive map of Arctic indigenous languages, including Evenki, Chukchi, Dolgan, Nenets and others.
Keeping Cultural Roots Alive
Digital language preservation has clearly moved beyond the efforts of individual enthusiasts and has become part of broader public policy. The Key Word Prize, which includes a category dedicated to the best digital language projects, reflects growing institutional support for initiatives of this kind.
The next step may involve the creation of a unified digital ecosystem – a platform bringing together mobile applications, online courses, dictionaries and textual archives in the languages of Russia’s many ethnic groups. Projects emerging from Yugra, Karelia, Kamchatka and Yakutia could become building blocks of that larger infrastructure.
Technologies such as KORYAK TUYU may also attract international attention. Many countries in Africa, Southeast Asia and Latin America face similar challenges in preserving minority languages. Russia’s experience developing digital educational tools for dozens of indigenous languages could therefore provide valuable models for EdTech and digital heritage solutions worldwide.
