SynapSpeech
Project Overview
About
Alzheimer’s disease is often diagnosed late because neuroimaging and biomarker tests are costly and not widely available. SynapSpeech uses natural language processing and machine learning to analyse spontaneous speech for subtle signs of cognitive decline, including hesitation, less varied vocabulary and disorganised storytelling. People can submit voice recordings through a mobile device or messaging platform, allowing screening without specialist equipment. Trained on Brazilian Portuguese data, the system is designed for linguistic and cultural relevance. As a clinical decision-support tool, it could help identify at-risk individuals earlier and prioritise further assessment.
Team
João Pedro Madureira Sales
João Pedro Madureira Sales is a Computer Engineering student at the Federal University of Technology – Paraná (UTFPR), Brazil, with a strong interest in artificial intelligence, data science, machine learning, and bioinformatics. He is the creator of SynapSpeech, a project that applies artificial intelligence and speech analysis to support cognitive screening. His work focuses on developing accessible, human-centered technologies with real-world impact, combining research, software development, and data-driven approaches to address complex healthcare and social challenges.
Robson Parmezan Bonidia
Robson Parmezan Bonidia is a researcher, educator, and social innovator dedicated to democratizing Artificial Intelligence for the Global South. With over 40 awards and recognitions, his groundbreaking projects—such as InteliGente and BioAutoML—have impacted thousands of people. His work has been featured in major media outlets and endorsed by institutions like Google, AI4PEP, HundrED.org, and FRIDA. A firm believer in AI for social good, Robson empowers educators, students, and scientists to harness AI ethically and responsibly, bridging the gap between cutting-edge technology and underserved communities. He holds a Ph.D. in Computer Science and Computational Mathematics from the University of São Paulo (USP). He also earned a degree in Information Security Technology from the Faculdade Estadual de Tecnologia de Ourinhos (FATEC – Centro Paula Souza), and holds a specialization in Computer Networks and a Master’s degree in Bioinformatics, both from the Federal University of Technology – Paraná (UTFPR). Currently, he serves as a member of the Steering Committee of the Global South Artificial Intelligence for Pandemic and Epidemic Preparedness and Response Network (AI4PEP) and is a professor at UTFPR – Cornélio Procópio campus.
Marina Carvalho Arruda Barreto
Marina Carvalho Arruda Barreto is a postdoctoral researcher at the University of São Paulo (ICMC-USP), Brazil, a member of the AI4PEP network (Brazilian Hub – Auto AI-Pandemics), and of the INCT IAPROBEM. Trained as a physiotherapist, she holds a PhD in Public Health from the Federal University of Ceará-Brazil. Her work bridges public health and artificial intelligence, focusing on the use of machine learning to investigate diseases of public health interest, including speech-based cognitive screening, epidemiological surveillance, and the analysis of mortality and disease burden.
Bruno Rafael Florentino
Formed in Physical and Biomolecular Sciences, I have a solid foundation in physics, molecular biology, and quantitative methods. Since graduation, I have pursued interdisciplinary research at the interface of computer science and biology, focusing on machine learning and artificial intelligence for biological applications, particularly the development of accessible models and datasets for predicting protein–protein and RNA–protein interactions, with applications in bioinformatics and biomedical research.