Biography

I am a tenure-track Assistant Professor in the College of Information Science at the University of Arizona, where I direct the CRISP Lab (Conversational Research in Interdisciplinary Speech Processing). 

I earned my Ph.D. in Electrical and Computer Engineering and B.S. in Electrical Engineering from the University of Illinois Urbana-Champaign. Following my Ph.D, I also spent time visiting the Language Technologies Institute at Carnegie Mellon University. During my Ph.D., I interned at Meta AI Research and Amazon Web Services.

I'm actively recruiting self-motivated PhD, M.S, undergraduate students to join my group. Please read more recruiting information here!

Research Vision

My research vision is to develop interdisciplinary speech technologies that enable meaningful real-world applications in healthcare, education, and human development. I am particularly interested in developing robust speech AI systems for analyzing multi-speaker interactions in complex, real-world environments. My goal is to build methods that can handle diverse vocal behaviors and noisy social contexts, extract clinically and socially meaningful behavioral signals, and support scalable screening, assessment, and other novel applications across populations and settings.

During my doctoral study, I focused on speech applications for early childhood (<4 years old), especially for supporting the early identification of developmental conditions such as autism, delayed speech maturity, and language disorders. My work centered on several foundational speech processing tasks for child-adult interactions: speaker diarization, vocalization classification, and phoneme recognition. I developed these methods across different application domains, including daylong home recordings and clinical settings. My prior work has contributed to:

• monitoring infant psychological development through naturalistic family audio analysis

• identifying speech and developmental markers associated with autism risk and speech maturity

• improving phoneme recognition performance for young children’s speech.

More broadly, this line of work reflects my interest in creating speech technologies that bridge machine learning, speech science, clinical practice, and education. Looking ahead, I aim to advance the core speech processing capabilities needed for real-world human-centered applications, while expanding to new problems in healthcare and education where speech can serve as an accessible, scalable, and informative signal.

Miscellaneous: 

Academic family tree:  me – Hasegawa-Johnson – Stevens – Beranek – Hunt – Chaffee – Pierce – Macfarlane – Tait – Hopkins – Sedgwick – Jones – Postlethwaite – Whisson – Taylor – Smith – Cotes – Newton.