Yasalma is a non-profit community building language and speech technologies for Tatar and other Turkic languages. Our work focuses on Natural Language Processing, speech technologies, LLMs, Text-To-Speech, and Machine Translation.
Yasalma проектының Hugging Face'тагы рәсми битенә хуш килдегез. Безнең максат — табигый телне эшкәртү һәм сөйләм технологияләре ярдәмендә татар һәм башка төрки телләр өчен заманча ЯФ коралларын булдыру.
Без телне саклау, үстерү һәм заманча технологияләрдә куллану өчен ачык датасетлар, модельләр һәм практик инструментлар өстендә эшлибез.
Welcome to the official Hugging Face page of Yasalma, formerly neurotatarlar. Our mission is to advance the Tatar language and other Turkic languages through cutting-edge AI technologies.
We focus on practical open resources for Natural Language Processing and Speech Technologies: datasets, models, benchmarks, and tools that help bring Turkic languages into modern AI systems.
We are building and improving open resources for Tatar language AI, with a focus on:
We welcome contributors interested in supporting Tatar language technology. You can help by creating and improving ML models, translating tasks, collecting and cleaning data, testing tools, or supporting the project financially.
No programming experience is required. If you speak Tatar or another Turkic language, your language knowledge is valuable for data collection, evaluation, translation, and community feedback.
Browse Yasalma's public Hugging Face work: models, datasets, Spaces, and curated collections for Tatar speech and language technology.
ASR, TTS, language modeling, and other experimental AI models for Tatar language tasks.
View modelsCorpora, speech datasets, OCR benchmarks, translation data, and other language resources.
View datasetsInteractive demos, annotation tools, experiments, and community-facing applications.
View Spaces