Keywords

Utah English variation, corpora, automated transcriptions

Abstract

Utah English Pronunciation

Early 20th Century Utah English:

Early work (Cook 1969, Helquist 1970) compares rural Utah to Salt Lake City.
Bowie (2003, 2008, etc.) analyzes archival recordings of religious sermons.
Primary focus in those studies: the decline of the CORD-CARD merger.

Introducing the Kohler Tapes

Background:

Norm Kohler, a middle school teacher in Heber in 1980s–2000s.
He had students to interview a local older person, like a grandparent.
He provided questions to ask, akin to sociolinguistic interviews.
Collected 1200+ cassette tapes.
Intended to write a history based on these oral narratives.

Processing:

Digitized in 2021 at BYU’s Office of Digital Humanities.
Metadata extracted from each tape (by Jessica Shepherd)
We filled in the rest using a genealogy website.
Manual transcription has been very slow and arduous.

Recent developments in transcription!

We are training Whisper to transcribe the rest.

This is possible thanks to BYU CS PhD student, Alex Lyman.
Preliminary output is very encouraging.

Many fascinating applications once those transcriptions are done.

Summaries of each interview.
Social network analysis based on named-entity recognition.
High-dimensional clustering to predict language variation.

A proof-of-concept for other large-scale transcription tasks!

Original Publication Citation

Joseph A. Stanley & Hallie Davidson. “Variation in Early 20th Century Rural Utah English”. Poster presentation at the American Dialect Society Annual Meeting. Philadelphia, Pennsylvania. January 11, 2025