Researchers at Texas A&M University are helping develop a new artificial intelligence-powered platform designed to improve communication and enhance quality of life for children with autism and older adults living with dementia.
The project, known as VoiceLove, is being developed through funding from the Advanced Research Projects Agency for Health (ARPA-H). Texas A&M University’s School of Public Health is one of three organizations selected to participate in the first phase of development.
Unlike traditional approaches that rely on generalized recommendations, the VoiceLove platform is designed to learn the unique communication styles, behaviors and preferences of each individual. By recognizing patterns over time, the system aims to help caregivers, therapists and healthcare providers better understand what a person is trying to communicate—even when words are difficult.
For families of children with autism, the technology could lead to more personalized support plans that strengthen communication skills, encourage greater independence, improve social interactions and build on each child’s individual strengths. Rather than treating autism as a one-size-fits-all condition, the platform is designed to adapt to each child’s unique needs as they grow and develop.
Researchers also see significant promise for people living with dementia. As memory loss and cognitive changes progress, many individuals struggle to express their needs, preferences or discomfort. The AI platform is intended to help caregivers recognize behavioral patterns that may indicate hunger, pain, anxiety, confusion or other unmet needs, allowing for more timely and compassionate care.
The system works by allowing caregivers or therapists to securely record short videos during everyday interactions. The AI analyzes both verbal and nonverbal communication—including speech patterns, facial expressions and behaviors—to identify meaningful trends over time. Caregivers can also record observations and respond to daily prompts, giving the system additional information to refine its recommendations.
Because the platform continually learns from each person’s daily experiences, its guidance becomes increasingly individualized. Researchers believe this adaptive approach could help caregivers better anticipate needs, reduce frustration for both patients and families, and create more proactive care plans tailored to each individual.
The research team plans to begin a pilot study involving a small group of children with autism and older adults living with dementia once institutional approvals are complete. Participants will use the platform for about one month while researchers evaluate its ease of use, effectiveness and overall user experience.
Following the pilot program, the team will continue refining the technology before expanding to larger clinical studies.
Researchers say the long-term goal is to create a first-of-its-kind AI platform that combines artificial intelligence, healthcare, behavioral science and human-centered design to support more personalized care. If successful, the technology could help families, caregivers and healthcare professionals better understand those living with autism or dementia, leading to improved communication, stronger relationships and better overall health outcomes.
