Mark Hasegawa-Johnson, Ph.D.

Mark Hasegawa-Johnson is a researcher at the Beckman Institute for Advanced Science and Technology and a professor of electrical and computer engineering at the University of Illinois Urbana-Champaign. He is the William L. Everitt Faculty Scholar in ECE and holds affiliations in the Department of Speech and Hearing Science, Coordinated Science Lab, Department of Linguistics, and Department of Computer Science. He also leads the Speech Accessibility Project, a new research initiative to make voice recognition technology more useful for people with a range of diverse speech patterns and disabilities. 
Hasegawa-Johnson has been on the faculty at the University of Illinois since 1999. His research addresses automatic speech recognition with a focus on the mathematization of linguistic concepts. His group has developed mathematical models of concepts from linguistics including a rudimentary model of pre-conscious speech perception (the landmark-based speech recognizer), a model that interprets pronunciation variability by figuring out how the talker planned his or her speech movements (tracking of tract variables from acoustics, and of gestures from tract variables), and a model that uses the stress and rhythm of natural language (prosody) to disambiguate confusable sentences. Applications of his research include:

Speech recognition for talkers with cerebral palsy. The automatic system, suitably constrained, outperforms a human listener.

Provably correct unsupervised ASR, or ASR that can be trained using speech that has no associated text transcripts.


Equal Accuracy Ratio regularization: Methods that reduce the error rate gaps caused by gender, race, dialect, age, education, disability and/or socioeconomic class.


Automatic analysis of the social interactions between infant, father, mother, and older sibling during the first eighteen months of life.


Hasegawa-Johnson is currently Senior Area Editor of the journal IEEE Transactions on Audio, Speech and Language and a member of the ISCA Diversity Committee. He has published 308 peer-reviewed journal articles, patents, and conference papers in the general area of automatic speech analysis, including machine learning models of articulatory and acoustic phonetics, prosody, dysarthria, non-speech acoustic events, audio source separation, and under-resourced languages.
Education

Postdoctoral fellow, University of California at Los Angeles, 1996-1999

Ph.D., Massachusetts of Technology, 1996

M.S., Massachusetts Institute of Technology, 1989

Honors

2023: Fellow of the International Speech Communication Association for contributions to knowledge-constrained signal generation

2020: Fellow of the IEEE, for contributions to speech processing of under-resourced languages


2011: Fellow of the Acoustical Society of America, for contributions to vocal tract and speech modeling


2009: Senior Member of the Association for Computing Machinery


2004: Member, Articulograph International Steering Committee; CLSP Workshop leader, "Landmark-Based Speech Recognition”, Invited paper


2004: NAACL workshop on Linguistic and Higher-Level Knowledge Sources in Speech Recognition and Understanding


2003: List of faculty rated as excellent by their students


2002: NSF CAREER award


1998: NIH National Research Service Award


Personal website: http://www.ifp.illinois.edu/~hasegawa/
CV: http://www.ifp.illinois.edu/~hasegawa/2021MarHasegawa-Johnson.pdf