Soliciting judgments of prosodic prominence and boundary from naive listeners in a corpus of African American and Latine English
Presentation Type
Abstract
Faculty Advisor
Jonathan Howell
Access Type
Event
Start Date
25-4-2025 1:30 PM
End Date
25-4-2025 2:29 PM
Description
Prosodic elements such as pitch and rhythm play a crucial role in shaping how listeners interpret a speaker’s intentions. While the study of prosody offers valuable insight into how language conveys meaning, there remains a lack of data on the prosody of marginalized English varieties. Dr. Jonathan Howell’s ongoing work addresses this gap by collecting prosodically annotated speech samples from non-standard dialects, such as African American English and Latine English, to improve the representation of linguistically significant variation in both theoretical and applied contexts. The current project builds upon this foundation by integrating corpus data with the Language Markup and Experimental Design Software (LMEDS), an established tool made for interactive speech analysis. The piloted method explores how untrained listeners perceive and identify both prominence and boundary in the data systematically selected from a large corpus of spontaneous speech. Reflecting on the challenges encountered during development, the project proposes new directions and advocates for the goal of making linguistic tools and data more accessible for both research and education. As the project remains in its pilot phase, current efforts focus on fine-tuning the interface and improving usability for broader application.
Soliciting judgments of prosodic prominence and boundary from naive listeners in a corpus of African American and Latine English
Prosodic elements such as pitch and rhythm play a crucial role in shaping how listeners interpret a speaker’s intentions. While the study of prosody offers valuable insight into how language conveys meaning, there remains a lack of data on the prosody of marginalized English varieties. Dr. Jonathan Howell’s ongoing work addresses this gap by collecting prosodically annotated speech samples from non-standard dialects, such as African American English and Latine English, to improve the representation of linguistically significant variation in both theoretical and applied contexts. The current project builds upon this foundation by integrating corpus data with the Language Markup and Experimental Design Software (LMEDS), an established tool made for interactive speech analysis. The piloted method explores how untrained listeners perceive and identify both prominence and boundary in the data systematically selected from a large corpus of spontaneous speech. Reflecting on the challenges encountered during development, the project proposes new directions and advocates for the goal of making linguistic tools and data more accessible for both research and education. As the project remains in its pilot phase, current efforts focus on fine-tuning the interface and improving usability for broader application.
Comments
Poster presentation at the 2025 Student Research Symposium.