endangered-languages 开源项目推荐（48 个）

Tatoeba /tatoeba2

Tatoeba is a platform whose purpose is to create a collaborative and open dataset of sentences and their translations.

RichardLitt /low-resource-languages

A curated list of resources for the conservation, development, and documentation of low resource (human) languages.

livingtongues /living-dictionaries

Speeding the availability of language resources for endangered languages. Tools such as this have the power to shift how we think about endangered languages. Rather than perceiving them as being antiquated, difficult to learn and on the brink of vanishing, we see them as modern, easily accessible for learning online in text and audio formats.

RichardLitt /thesis

My thesis on "Open Source Code and Low Resource Languages" for an MSc in Language Science and Technology at Saarland University

CoEDL /vad-sli-asr

A pipeline to isolate and transcribe one language in mixed-language speech

Halvani /Alphabetic

A Python module for retrieving script types of writing systems including alphabets, abjads, abugidas, syllabaries, logographs, featurals as well as Latin script codes

108charlotte /Living-Flashcards

A spaced repetition flashcard app for learning endangered languages

dnnyhr /Piko-Aplicacion-para-el-aprendizaje-de-idiomas-

Piko es una aplicación desarrollada para Hackathon Nicaragua 2026, dando solución al reto de aprendizaje de idiomas en zonas rurales del pais mediante aplicaciones y asistentes fisicos de IA en las aulas de clases.

Louisianish /weatherkv

Weather app originally created by CodeExplained (https://github.com/CodeExplainedRepo/Weather-App-JavaScript) to which I have added translations of weather descriptions in Kouri-Vini (Louisiana Creole) and a location search bar. Please let me know if any of my translations aren't showing up correctly. Byin mèsi. :)

jo-valer /machine-translation-ladin-fascian

Repository of our paper Nesciun Lengaz Lascià Endò: Machine Translation for Fassa Ladin.

areebhirani-beep /lantern

A language dies every two weeks. Lantern is Duolingo for dying languages: it turns the words a community still remembers into a real course, with AI that never invents a word.

William-Robert-Robertson /Scottish-Gaelic-Spellchecker-GOC

Scottish Gaelic Spellchecker - GOC (Gaelic Orthographic Convention)

terma-heritage /termatype

The first word processor built for English and Tibetan. Free, open source, 100% offline. Wylie and TCRC-Bodyig input, Tibetan-English-Tibetan dictionary,

schyuler /Chamorro-English-Kindle-Dictionary

A Python script supporting Chamorro language preservation through the creation of a custom Chamorro-English dictionary for Kindle devices—making reading in Chamorro more accessible.

dearabhin /LipiPala

Saving endangered Indian languages with open AI innovation

EmilyMB /la-joya

Nahuatl language family album

cutthroat78 /Oneida-Canada-to-English-Dictionary

A Oneida (Canada) to English Dictionary

fahim-ling /Sylheti-Bangla-English-Russian-German-Parallel-Corpus-for-NLP

The first open-source 𝗺𝘂𝗹𝘁𝗶𝗹𝗶𝗻𝗴𝘂𝗮𝗹 (5 languages) corpus for low-resource NLP, boldly bridging three distinct language branches. Built by a 𝗻𝗮𝘁𝗶𝘃𝗲 𝗦𝘆𝗹𝗵𝗲𝘁𝗶 Linguistics undergrad at 𝗞𝘂𝗿𝘀𝗸 𝗦𝘁𝗮𝘁𝗲 𝗨𝗻𝗶𝘃𝗲𝗿𝘀𝗶𝘁𝘆, Russia. Targeting a 10K+ sentence dataset for MT/ASR training to computationally revitalize Sylheti.

William-Robert-Robertson /Scottish-Gaelic-Spellchecker-Universal

Scottish Gaelic Spellchecker (Universal)

schyuler /Chamorro-Dictionary-Scraper

A project to scrape and process online Chamorro language dictionaries to support language analysis and revitalization efforts. (WIP)