A multilingual ASR model that can recognize ten Turkic languages—Azerbaijani, Bashkir, Chuvash, Kazakh, Kyrgyz, Sakha, Tatar, Turkish, Uyghur, and Uzbek.
-
Updated
Aug 1, 2025 - Python
A multilingual ASR model that can recognize ten Turkic languages—Azerbaijani, Bashkir, Chuvash, Kazakh, Kyrgyz, Sakha, Tatar, Turkish, Uyghur, and Uzbek.
Kyrgyz language processing software, models and datasets.
Kazakh full-text search extension for PostgreSQL
NLP Toolkit for Turkic Languages
Cross Lingual Word Embeddings for Turkic Languages
KyrgyzNLP: Browsing, Bibliography, and Scientometrics
Multi-label 24.kg Kyrgyz news articles classification into 20 topics: data & baselines
Open-source Automatic Speech Recognition (ASR) pipeline for Bashkir (Bashkort), Kazakh, and Kyrgyz languages with deterministic orthography correction.
An open parallel Azerbaijani-English LLM benchmark. Kazakh fine-tuning produces negative transfer to Azerbaijani, and the mechanism is orthographic.
Karakalpak language toolkit for Python — Latin/Cyrillic script conversion, number-to-words, and string utilities
On the history and evolution of culture and institutions
Apertium monolingual language package for Urum. Features include morphological analysis/genertion, and part-of-speech tagging.
Turkic Transliterator: Fast, cross-platform Python transliteration and IPA conversion for Turkic languages.
An interactive 3D digital museum and Old Turkic (Göktürk) translation engine built with Next.js and FastAPI.
To associate your repository with the turkic-languages topic, visit your repo's landing page and select "manage topics."