Date of Degree
prosody, legendre polynomials, sarcasm, nativeness
This investigation demonstrates the effectiveness of Legendre polynomial coefficients representing prosodic contours within the context of two different tasks: nativeness classification and sarcasm detection. By making use of accurate representations of prosodic contours to answer fundamental linguistic questions, we contribute significantly to the body of research focused on analyzing prosody in linguistics as well as modeling prosody for machine learning tasks. Using Legendre polynomial coefficient representations of prosodic contours, we answer prosodic questions about differences in prosody between native English speakers and non-native English speakers whose first language is Mandarin. We also learn more about prosodic qualities of sarcastic speech. We additionally perform machine learning classification for both tasks, (achieving an accuracy of 72.3% for nativeness classification, and achieving 81.57% for sarcasm detection). We recommend that linguists looking to analyze prosodic contours make use of Legendre polynomial coefficients modeling; the accuracy and quality of the resulting prosodic contour representations makes them highly interpretable for linguistic analysis.
Rakov, Rachel, "Analyzing Prosody with Legendre Polynomial Coefficients" (2019). CUNY Academic Works.