Abstract

The linguistic landscape of Central and Northern Asia, stretching from the Pontic steppes to Siberia and the Far East, represents one of the most diverse yet critically endangered regions on Earth. This panel examines the intersection of traditional documentary and corpus linguistics with contemporary computational strategies and pedagogical innovations for "low-resource" languages. While major regional languages like Uzbek or Kazakh are making strides in the digital domain, dozens of minority languages, including Turkic, Mongolic, Tungusic, Ugric, Samoyedic, and Paleoasiatic varieties, remain on the "digital periphery".

The panel explores the entire research cycle: from data collection to corpus design and its application in teaching and learning. Key areas of focus include:

- Collaborative Documentation and Pedagogy: Analyzing community-engaged models that prioritize "language-in-use" over static literary standards.

- Technological Adaptation: Evaluating the efficacy of transfer learning, cross-lingual embeddings, and OCR in creating corpora for languages with limited annotated digital data.

- Curriculum Integration: Developing frameworks to enhance language education by integrating corpus-based resources, such as spoken corpora, into curricula and teacher-training programs.

- Ethical Data Sovereignty: Addressing tensions between open-access archiving and the intellectual property rights of speaker communities.

By bringing together linguists, computational scientists, and educators, this panel highlights how digital humanities can modernize Central Asian languages. Ultimately, we argue that preservation depends on transitioning from "passive archiving" to "active resource development," ensuring these voices are functionally integrated into both the global digital ecosystem and the classroom.