Voice Analysis Coding

Linguistics

WHO: Grant McGuire, Tom Patterson

WHAT: This project lays the foundation for future research on voice typicality and memory. We are finishing coding a set of voice data for future analysis.

WHY: Speech perception relies heavily on expectations around voices. We know very little about how listeners sort and identify them.

WHAT'S NEXT: By the end of the quarter, we should have three data sets for 40 speakers: a word list, a sentence list, and a passage. Each of these data sets will be phonetically analyzed for the qualities of each voice and their acoustic similarity.

THE WOW: The voices are pretty diverse! We’ve learned that the specific ways we ask listeners about voice characteristics can result in significant changes in our results.