This article is about the work being done by Michael Running Wolf, who is a clinical instructor of computer science at Northeastern University’s Khoury College of Computer Sciences, on developing methods for documenting and maintaining Indigenous languages through automatic speech recognition software. This work is a precursor to his long term goal of providing a way for Indigenous youth to learn their language by way of technological immersion, using technologies such as virtual reality or augmented reality.
Part of the difficulty of developing automatic speech recognition for Indigenous languages is that in the field of computational linguistics, relatively little research has been devoted to Indigenous languages. An additional challenge is that many Indigenous languages are “polysynthetic” meaning that they have words that contain many morphemes, or units of language that cannot be further divided. As Michael Running Wolf points out, “polysynthetic languages often have very long words – words that can mean an entire sentence, or denote a sentence’s worth of meaning.”