Please read our student and staff community guidance on COVID-19
Skip to main content Skip to navigation

Paper No. 11-27

Download 11-27

PZ Hadjipantelis, JAD Aston and JP Evans

Charaterizing fundamental frequency in Mandarin: A functional principal component approach utilizing mixed effect models

Date: September 2011

Abstract: A model for fundamental frequency (F0, or commonly pitch) employing a functional principal component analysis (FPCA) framework is presented. The language in the presented study is Taiwanese Mandarin; this Sino-Tibetan language is rich in pitch-related information as the relative pitch curve is specified in the syllable of each word's lexical entry. The original 5 speaker corpus is preprocessed using a locally weighted least squares smoother. These smoothed curves are then utilized as input for the computation of the final FPC scores and their corresponding eigenfunctions. These scores are finally utilized in a series of penalized mixed effect models to build meaningful categorical prototypes. These prototypes appeared to confirm known tonal characteristics of the language, as well as suggest the presence of a sinusoid tonal component that is previously undocumented. PACS numbers: 43.60.Cg, 43.60.Uv, 43.66.Hg

Keywords: Phonetic Analysis; Functional Data Analysis; Linear models