You are here
Frank Rudzicz, PhD
Toronto Rehabilitation; University of Toronto
Areas of Interest:
Assistive technologies, natural language processing, machine learning, speech, diagnostics, cognition
Biography & Research:
Dr. Frank Rudzicz, (PhD, University of Toronto, 2011) is a scientist at the Toronto Rehabilitation Institute with an assistant professor status in Computer Science at the University of Toronto. Dr. Rudzicz focuses on improving speech models for people with communication disabilities. One focus of his research has been on speech disorders resulting from disruptions in the neuro-motor interface (e.g., caused by cerebral palsy) that result in unintelligible speech – a condition called dysarthria. He designed and collected one of the first and most extensive databases of dysarthric speech articulation, Torgo, which is now distributed internationally by the Linguistic Data Consortium. Ongoing work includes developing unique algorithms that transform hard-to-understand speech signals to be better understood by typical listeners, almost doubling the number of words that can be correctly understood. This invention was nominated by the University of Toronto for an NSERC Innovation Challenge award and a company, Thotra Incorporated, has been founded to commercialize it. Currently, Dr. Rudzicz is increasingly working with lexical-semantic communication disorders, e.g., related to dementia. This work recently received a Best Student Paper award at Interspeech 2013 for results that showed that combining acoustic and linguistic measurements of an adult’s speech can accurately diagnose aphasia up to 88% of the time. Dr. Rudzicz is recipient of the Alzheimer Society of Canada’s Young Investigator award and is a Network Investigator of the AGE-WELL National Centres of Excellence.