About me

My name is Xinyu Leslie Liao. I’m a first-year PhD student in Linguistics at the University of Toronto. The central focus of my research is perceptual flexibility in speech perception and computational speech models, bridging perception experiments, computational modeling, and theories of phonetic learning and language evolution. Human perceptual and memory systems are remarkably adaptive—listeners continuously recalibrate their category boundaries (e.g., for speech sounds) in response to novel input. I am particularly interested in how listeners adapt to or compensate for the innovative phonetic input, how explicit computational models can be built to probe human cognition, and whether computational models can exhibit similar perceptual biases as human does.

With my past training, I apporach these questions from an interdisciplinary perspective, especially from phonetics/phonology, psycholinguistics, computational modeling, and variationist sociolinguistics. my CV is here

Research

Phonetic biases and word frequency effects in perceptual learning: implications for sound change

By exposing listeners to innovative phonetic input, I explore biases in the phonetic recalibration of phonemic categorization, with implications for episodic exemplar models of phonology, and the role of acoustic memory and word frequency encoding in diachronic sound change

I’ll keep updating my research outputs here. Feel free to reach out :)