At the 2023 NSF NRT Annual Meeting, I presented a poster on native language processing impacts on human language technology. My collaborators and I used EEG to evaluate neural processing differences when native Spanish listeners hear matched versus mismatched native accent variants. We found evidence that unfamiliar variants incur cognitive costs. As language ID and speech recognition systems currently struggle with non-native speech, I’m integrating psychological and computational linguistics techniques to explore improvements. This project synthesizes varied disciplines - from psycholinguistics to natural language processing - to address inclusion gaps in speech technology across accent variants. Enhancing system robustness to diverse forms of speech is essential as virtual assistants and other AI applications continue permeating everyday life.
Slides can be added in a few ways:
slides parameter in the front matter of the talk file
static/ and link using
url_slides parameter in the front matter of the talk file
Further event details, including page elements such as image galleries, can be added to the body of this page.