Brain-based wireless hearing aid enhancer using language-model priorities
Official title: Semantic-based auditory attention decoder using Large Language Model
Funded at Rochester Institute of Technology · NIH-11381057 · Data updated 2026-11-12
A new project aims to help people with hearing loss better follow conversations by using brain signals and smart language models to focus on the person speaking.
Quick facts
| Grant type | NIH-funded research |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Rochester Institute of Technology NIH-funded |
| Lab location | 1 site (Rochester, United States) |
| Project start | August 1, 2025 |
| Project end | July 31, 2029 |
| Funding year | 2026 |
| Award (this year) | $185,000 |
| NIH project number | 5R16GM159795-02 |
| Project ID | NIH-11381057 on NIH RePORTER |
Who could benefit from this research
✓Could help
Ideal candidates are adults with hearing loss who rely on hearing aids or cochlear implants and who experience difficulty understanding speech in noisy environments.
✕Less likely to help
Individuals with normal hearing or those without difficulty in noisy listening situations may not directly benefit.
What this research studies
This project will explore how the brain’s attention to a speaker can be decoded from EEG data to improve how hearing devices amplify the person you want to hear. By combining brain signals with AI language-model cues, the system aims to boost the target voice while reducing background noise. The work will use existing EEG datasets to identify patterns that indicate what a listener is trying to focus on, with the goal of guiding real-time signal processing in hearing devices. If successful, it could make conversations easier for people with hearing loss and reduce cognitive load during noisy conversations.
Why it matters
Potential benefit: If successful, it could improve speech understanding in noise and reduce listening effort for people with hearing loss.
How similar studies have gone: Prior EEG attentional decoding research shows promise, but applying it to real-time hearing aid enhancement is still early-stage.
Where this research is happening
-
Rochester Institute of Technology
Rochester, United States Active
Researchers
- Principal investigator: Shim, Hwan — Rochester Institute of Technology
- Study coordinator: Shim, Hwan
Related research
Other NIH-funded projects on the same topics.
-
Deep learning approach to diagnosing chronic headaches and facial pain
A new AI-based method aims to improve how doctors diagnose chronic headache and facial pain by using a large, long-term set of patient data.
-
Understanding how jumbo phages build their giant protein shells
Researchers are studying how extremely large bacteriophages assemble their big protective heads to better use phages against drug‑resistant bacteria.
-
Navigating damaged tissue: how cells use topographical clues to move
This grant studies how healthy or diseased cells find their way through complex tissue by learning to remember guidance cues from the environment.
-
Smartphone coaching to prevent falls after stroke and train researchers
A smartphone-based coaching tool is being developed to help stroke survivors move more safely and reduce fall risk, with training opportunities for students at a major technical university.
-
AI-assisted digital twin for personalized phonosurgery planning
A new AI-powered digital twin will simulate a patient’s vocal fold biology to help tailor phonosurgery for clearer, more reliable voice restoration.
-
3D AI-driven system to map heart flutter circuits for safer ablation
A new AI-assisted, 3D model helps doctors locate the hidden heart circuit that causes dangerous ventricular tachycardia to guide safer, more effective ablation.
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.