Skip to main navigation Skip to search Skip to main content

Collaborative Research: Prosodic Analysis and Visualization of Phonetic Samples for Improved Understanding of Stress and Intonation

  • University of Washington

Project: Research

Project Details

Description

Rhythm and intonation are important part of pronunciation. While there are some basic similarities in rhythm and intonation patterns across languages, there are also interesting differences. Some of these differences are linked to the way words are structured in languages, for example, are the words short or are they composed of many sub-parts. This project investigates and documents the rhythm and intonation, or prosodic system, of under-documented languages to better understand how humans modulate pitch movement (i.e. intonation) and segmental duration (i.e. the length of vowels and consonants). The project uses archived and new recordings to arrive at acoustic measurements and visualizations of pitch and duration, creating a database that can be mined for the study of the interplay of prosody and other domains such as the syllable, word, foot, prosodic word, prosodic phrase, and intonation unit. The project includes training for graduate students and undergraduate students from Aaniiih Nakoda College.

The investigators mine archived recordings or new recordings of words, phrases and connected speech produced by native speakers to improve and understand the prosodic system of the morphologically complex languages for analysis. They represent their findings through the software in development by the team for this purpose. A major product of this project is a database with phonetic analyses across different prosodic domains, e.g, the syllable or word. These materials will be made available through an online webservice and will be useful for researchers working to analyze and represent other languages with significant modulations of pitch and duration. The software includes a self-pronunciation check feature allowing the user to listen, repeat, and check accuracy in reproducing the pitch and duration values of native speakers. This feature will be used to understand how well such overt input impacts second language pronunciation acquisition.
StatusFinished
Effective start/end date09/1/21 → 02/28/25

Collaborative partners

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.