AI institute for speech, language disorders to help children

Progress abounds for the National AI Institute for Exceptional Education.

The $20 million, UB-led initiative — funded in January by the U.S. National Science Foundation and Institute of Education Sciences — will create artificial intelligence systems that ensure children with speech and language disorders receive timely, effective assistance.

The work addresses the nationwide shortage of speech-language pathologists (SLPs) and provides services to millions of children, ages 3 to 10, who are at increased risk of falling behind in their academic and socio-emotional development.

The institute’s research team — comprised of dozens of investigators from nine universities that specialize in artificial intelligence, natural language processing, social robotics, diversity and inclusivity, and other fields — held its inaugural in-person meeting in April at UB.

Since then, researchers have begun lab researching and building prototypes of two AI systems: the AI Screener, which will identify potential speech and language impairments by listening to and observing children in classrooms; and the AI Orchestrator, which acts as a virtual teaching assistant to SLPs by providing students with ability-based interventions.

“The team has been making consistent progress thus far with a lot of community excitement around our work,” says Jinjun Xiong, scientific director and co-principal investigator at the institute. A SUNY Empire Innovation Professor in the Department of Computer Science and Engineering, Xiong came to UB in 2021 after working as a program director and senior research scientist at IBM’s Thomas J. Watson Research Center.

In addition to creating the AI technologies, the research team is establishing robust guidelines to ensure that data collected by those systems meets ethical and privacy guidelines.

Eventually, the institute will place each system in roughly 80 classrooms, reaching 480 students, with the majority of them being in Western New York. To ensure the system’s smooth rollout, institute leaders will be meeting with public school administrators and educators in Buffalo and beyond.

The institute builds upon a rich history of AI research at UB that dates back decades, when computer scientists developed a handwriting-recognition system that has saved the U.S. Postal Service hundreds of millions of dollars by automating the sorting of mail.

More recently, UB has made strategic investments in both people and programs to enhance the university’s national leadership in AI, including forming the Institute for Artificial Intelligence and Data Science and the Center for Information Integrity.

These efforts are addressing major societal challenges including combating dis- and misinformation, especially among older adults and other populations more susceptible to online scams; utilizing machine learning to reduce health inequities; advancing safe and secure biometrics; and more.

“With the National AI Institute for Exceptional Education quickly coalescing, as well as other established artificial intelligence programs the University at Buffalo progressing, we are poised to bring forth even more technologies with immense societal benefit,” says the institute’s director and principal investigator Venu Govindaraju, vice president for research and economic development.