Deploying Technology to Save Endangered Languages

TL;DR

This paper explores how advanced speech recognition technology, particularly neural networks, can be used to automate linguistic tasks to help preserve endangered languages.

Contribution

It proposes integrating neural network-based speech recognition with linguistic efforts to automate language annotation and analysis.

Findings

Neural networks can effectively automate speech tagging.

Automated tools can accelerate endangered language documentation.

Collaboration between technologists and linguists enhances preservation efforts.

Abstract

Computer scientists working on natural language processing, native speakers of endangered languages, and field linguists to discuss ways to harness Automatic Speech Recognition, especially neural networks, to automate annotation, speech tagging, and text parsing on endangered languages.