Automated speech and language detection for Alzheimer’s and frontotemporal dementia in Spanish and English speakers
An automated machine learning approach to language changes in Alzheimer’s disease and frontotemporal dementia across Latino and English-speaking populations
This project uses automated speech and language analysis to help spot signs of Alzheimer’s disease and frontotemporal dementia in Spanish- and English-speaking people, with special focus on Latino communities.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of California, San Francisco NIH-funded |
| Lab location | 1 site (San Francisco, United States) |
| Project ID | NIH-11404660 on NIH RePORTER |
What this research studies
You would provide short samples of natural speech that are recorded and converted into acoustic and language data. The team uses machine and deep learning to find patterns in sound waves and word use that link to Alzheimer’s disease or frontotemporal dementia. They are collecting a large, diverse group of participants from five Latin American countries and the United States to make sure the tools work across languages, dialects, and backgrounds. The project also compares these speech markers with standard clinical measures and accounts for factors like bilingualism, sex, and brain profiles.
Who could benefit from this research
Good fit: Ideal participants include Spanish-speaking adults from the participating Latin American countries and English-speaking adults in the US who have memory or language concerns, have been diagnosed with Alzheimer’s disease or frontotemporal dementia, or are healthy controls willing to provide speech samples.
Not a fit: People who cannot produce usable speech (for example, severe nonverbal aphasia), those who do not speak Spanish or English, or those with conditions not targeted by the project are unlikely to benefit directly from this work.
Why it matters
Potential benefit: If successful, this could lead to low-cost, scalable language-based tools for earlier and more accessible detection and monitoring of dementia across languages and cultures.
How similar studies have performed: Previous smaller studies using automated speech and language analysis have shown promising signals for detecting dementia, but large multilingual and culturally validated efforts like this are relatively new.
Where this research is happening
San Francisco, United States
- University of California, San Francisco — San Francisco, United States (Active)
Researchers
- Principal investigator: Gorno Tempini, Maria Luisa — University of California, San Francisco
- Study coordinator: Gorno Tempini, Maria Luisa
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page below.