Connecting Documentation and Revitalization: A New Approach to Language Apps - Abstract

Aug 29, 2026

Upload by Cart

7000 Languages is a nonprofit effort that adapts commercial language-learning software for free use by endangered language groups. The work compares advantages and challenges of using donated technology and outlines how corpus and archive data can be converted into language-learning courses. It showcases the software’s ability to support both language revitalization and documentation, while proposing directions for program expansion and inviting feedback from practitioners in endangered language work.

LanguageEnglish

CategoryPresentations

FormatPDF

Pages5

Reading Time2 min

AccessFree to download

Introduction

The Programs

Partnership Program

Archives Program

Connecting Documentation and Revitalization: A New Approach to Language Apps Alexa N. Little 7000 Languages 12 Murphy Drive Nashua, NH 03062 [alittle@7000languages.org](alittle@7000languages.org) Abstract This paper introduces 7000 Languages, a nonprofit effort to adapt commercial language-learning software for free use by endangered language groups. We discuss the advantages and challenges of our approach and suggest ways of converting corpus and archive data into languagelearning courses. We also demonstrate the capabilities of the software, both for the revitalization and the documentation of endangered languages. Finally, we discuss ideas for future expansion of our programs and seek feedback from others involved in endangered language work. 1 Introduction Many endangered language communities are interested in producing “language apps”: digital tools for teaching and learning their languages. Although these language apps are useful for language revitalization efforts, building them requires considerable time, funding, and technical skill. Even successful community projects may lack the resources for future updates to their language apps. Researchers who study endangered languages also want to provide digital tools, so that their research and data will be helpful to communities. Creating and maintaining an entire languagelearning system, however, is more than a researcher can reasonably accomplish. The current compromise is the digital archive. In a digital archive, language data is secure, accessible, and generally portable. XML-formatted data could even, with some manipulation, be integrated into a language app. However, the time and expense of building the rest of such an application remains a challenge for both sides. We introduce a nonprofit effort, 7000 Languages, that seeks to resolve this problem. Our approach uses technology donated by the language-learning industry to produce free commercial-grade language apps in partnership with endangered language advocates. 2 The Programs We have organized our approach into two programs – one from a revitalization-first perspective, and one from a documentation-first perspective. Below is a brief description of these two programs, followed by an in-depth discussion of how projects starting from either perspective will ultimately benefit both sides. 2.1 Partnership Program Our Partnership Program is a free program intended for groups, such as endangered language communities, who have a revitalization-first perspective. The system works as follows: 7000 Languages has an agreement with a forprofit language-learning company, Transparent Language. This agreement allows 7000 Languages to use Transparent Language’s internal tools to develop online courses for endangered and low-resource languages. The tools include: 􀁸 a program that converts an Excel template filled with language content (text, images, and recordings) into a functioning course 􀁸 a program for designing a custom unit, lesson, and activity layout 􀁸 a graphical user interface (GUI) for creating individual lessons, ideal for those with limited technical experience (see Figure 1) 151 Proceedings of the 2nd Workshop on the Use of Computational Methods in the Study of Endangered Languages, pages 151–155, Honolulu, Hawai`i, USA, March 6–7, 2017 . 􀀍c2017 Association for Computational Linguistics Figure 1. A GUI for lesson authoring. 7000 Languages trains interested groups, called Partners, to use these tools. The Partner decides what they want to teach, and they retain ownership rights of all the content they create. Once the course is finished, 7000 Languages contacts Transparent Language, who sets up a“node” for the Partner in their systems. This publishes the course online and lets the Partner create user accounts. The Partner can add as many users as they want, for free. The course will also be made available, for free, on the 7000 Languages website and published through Tra

FAQs

What problem does 7000 Languages aim to solve for language communities?

It targets the time, funding, and technical effort required to build and maintain language-learning apps, and the difficulty researchers face in creating full language-learning systems for communities.

How does the Partnership Program create and publish language courses?

It uses tools donated through an agreement with Transparent Language to convert content into online courses, trains partner groups to author lessons, and then has Transparent Language set up a partner node to publish and allow free user accounts.

How do the courses support both revitalization and documentation?

Courses store language data in an XML format, enabling export for archival, corpus creation, or reuse in other digital applications such as searchable dictionaries.

Aug 29, 2026

Upload by Cart