Research & ReadAlongs

My research centres the authority of communities to direct the development and use of technologies for their own languages. Beginning from a community’s own relational understanding of the world reshapes what the technology is for, how it is built and what counts as success. Responsibilities to people, land, language and knowledge shape its design and operation, making accountability part of every stage of the work. The aim is to support language learning without reproducing the colonial assumptions and extractive practices that have contributed to language endangerment.

View my ResearchGate profile →

MA / UBC Okanagan / 2024

Kʷu sqilxʷ:
an interactive digital edition.

My MA in Interdisciplinary Graduate Studies developed an interactive digital edition of Kʷu sqilxʷ / We Are the People, bringing nsyilxcən recorded speech into alignment with written text.

The work considered how a language technology could be developed through accountable relationships with the people, stories and institutions involved. Its technical and methodological questions continue to shape my practice.

The stories, recordings and permissions belong within their own community and rights-holder relationships; this public page describes the research and links to the thesis.

Thesis

Seeding alignment between language technology and Indigenous methodologies

The making of Kʷu sqilxʷ an interactive digital edition.
Craig Carpenter, University of British Columbia, 2024.
DOI: 10.14288/1.0442345.

Explore →

Language learning tools

Hear the voice.
Follow the words.

ReadAlong Studio, developed by the National Research Council Canada with Carleton University and Indigenous collaborators, supports speech–text alignment for more than 30 Indigenous languages. It highlights words during playback so learners can follow a recording and select individual words to hear them again.

With linguist Tony Mattina’s help, we developed a grapheme-to-phoneme (g2p) model for Interior Salishan languages. The model maps written forms to speech sounds, helping ReadAlongs align transcriptions with recordings for language learning and revitalization.

Listen

Use a recording that the language team has authority to work with.

Follow

Review the text and timing with speakers and educators who can assess the language.

Create

Build a documented process so the community can produce and maintain new resources.

ReadAlong training covers preparation, alignment, review and delivery, with the community deciding where the resulting resource can be used.

Try a ReadAlong: Horn Lake

Horn Lake, told by Edna Jack and transcribed with Clara Jack in 1988, is a public example on Tony Mattina’s website. Press play to follow the highlighted words, or pause and select a word to hear it again.

Published paper / SIGUL 2024

Seeding Alignment

Seeding Alignment between Language Technology and Indigenous Methodologies: A Decolonizing Framework for Endangered Language Revitalization.

Craig John Carpenter, John Lyon, Miles Thorogood and Jeannette C. Armstrong. Proceedings of the 3rd Annual Meeting of the Special Interest Group on Under-resourced Languages @ LREC-COLING 2024, pp. 318–324. ELRA and ICCL.

The paper develops a relational framework for language-technology research, drawing on respect, relevance, reciprocity and responsibility, and on the need for methodological accountability to the language community.

Bring research into a practical conversation.

I welcome conversations about community-directed archives, ReadAlong training and the careful development of language supports.