Acoustic analyses of TH-fronting

Much research exists on the environments where /θ/ fronting occurs (Green 2002; Hinton & Pollock 2000; Thomas 2007), but there has been little discussion about the acoustic realizations of this feature and its parameters of variation. This production study investigates the acoustic realizations of /θ/ and /f/ in the speech of 10 women. Production data for 133 lexical items were elicited from 10 Black women. This project aims to analyze how phonetic features, such as speech rate, duration, and spectral moments (i.e., variance, center of gravity, skewness, and kurtosis), vary across instances of /θ/ and /f/. Along with understanding the production difference between the variables, this research also analyzes the acoustic differences between resultative [f] (the product of /θ/ fronting) and canonical /f/. The central prediction is that /θ/ fronting displays systematic acoustic differences conditioned by both linguistic and social factors, revealing patterns within the ethnoracial variety. This study offers a new perspective on a well-known linguistic feature by analyzing its production and providing descriptive data on this phonological process, thereby contributing to our understanding of speech variability.