Using voice to detect and track health conditions
Bridge2AI: Voice as a Biomarker of Health - Building an ethically sourced, bioaccoustic database to understand disease like never before
This project gathers voice recordings from people and links them to medical records and other health data to build AI tools that could help screen and monitor conditions like Alzheimer's, Parkinson's, depression, lung disease, and childhood speech or autism.
Quick facts
| Grant type | NIH-funded research |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of South Florida NIH-funded |
| Lab location | 1 site (Tampa, United States) |
| Project ID | NIH-11376382 on NIH RePORTER |
What this research studies
The team will ask people at multiple medical centers to record their voice using a smartphone app and to allow those recordings to be linked to their electronic health records and other tests such as scans and genetic data. They plan to build a large, diverse, and ethically sourced voice database centered on five groups of conditions: vocal/laryngeal problems, neurological and neurodegenerative disorders, mood and psychiatric disorders, respiratory diseases, and pediatric speech/autism. Privacy-preserving methods like federated learning will be used so models can be trained without sharing raw personal data. Researchers will use the linked data to develop AI models intended to help screen, diagnose, or track these conditions over time.
Who could benefit from this research
Good fit: People who can use a smartphone and are willing to record their voice and permit linkage of those recordings to their medical records and other health tests are ideal candidates.
Not a fit: People without smartphone access, those who do not want to share medical records, or those with conditions not related to voice changes may not receive direct benefit from this project.
Why it matters
Potential benefit: If successful, this could enable noninvasive, low-cost screening and earlier detection or monitoring of many conditions from simple voice recordings.
How similar studies have performed: Smaller studies have shown voice features can signal Parkinson's, Alzheimer's, depression, and some speech disorders, but large, diverse, clinically linked voice databases are still limited.
Where this research is happening
Tampa, United States
- University of South Florida — Tampa, United States (Active)
Researchers
- Principal investigator: Bensoussan, Yael Emilie — University of South Florida
- Study coordinator: Bensoussan, Yael Emilie
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.