Work

The Pangloss Collection: Opening up Research Data on Endangered and Underdocumented Languages

Year: 2025

Type: article

Abstract: In this article, we present the Pangloss Collection, a collection of digital corpora of (mostly) endangered or underdocumented languages, developed in France since the 1990s in the context of a global realization of the considerable potential of digital technologies. The Pangloss Collection currently hosts 1,180 hours of audio and video recordings of about 200 languages. These materials are archived for the long-term using a suite of French public services for digital humanities. The Pangloss Collection can be freely accessed through a bilingual English-French website that was reshaped in 2021 to offer a general-audience interface mode and a professional interface mode (see https://pangloss.cnrs.fr/).

Source: Language

DOI: 10.1353/lan.2025.a954237

Authors: Evangelia Adamou, Séverine Guillaume, Alexis Michaud

Language: English

Topic: Linguistic Studies and Language Acquisition

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Agence Nationale de la Recherche

Awards: Probing Neural Representations for Typological Signal, Empirical Foundations of Linguistics : data, methods, models

OpenAlex ID: W4408719224

FWCI: 0.854

Cites:

Cited by: 2

Related to: 10