On this page ▾
Key takeaways
- BOSSA improved word recognition by up to 40 percentage points in a lab test with young adults.
- The test used simulated voices through headphones, so real-life performance is not yet known.
- BOSSA is not in any hearing aid yet and cannot follow shifts in attention.
- Today’s hearing aids can help in noise, and a hearing test shows what would suit you.
A brain-inspired algorithm called BOSSA improved word recognition in a lab test, by up to 40 percentage points compared with current hearing aid processing, for young adults with hearing loss1. It is not in any hearing aid yet, so it is a promising idea and not a product.
Following one voice among many is known as the “cocktail party problem”, and it is one of the biggest challenges for people with hearing loss in restaurants and at parties. If you find noisy places tiring, you are not alone. Standard hearing aids often use directional filters, but these work less well when several people talk at once or from similar distances.
BOSSA stands for Biologically Oriented Sound Segregation Algorithm. It copies how the brain uses sound from both ears to work out where voices are coming from, and then isolates the one the listener wants to follow1,2.
How was BOSSA tested?
The study, from Boston University, tested BOSSA with young adults with sensorineural hearing loss. They wore headphones that simulated several people talking from nearby positions, and listened with BOSSA, with a standard hearing aid algorithm and with no algorithm1,3. The standard approach gave little or no benefit in this test.
A useful way to picture it: standard directional systems are like a wide-beam torch, while BOSSA aims for a focused spotlight on one voice.
BOSSA is not perfect. It cannot yet follow your attention as it moves from one voice to another, and the trials used simulated voices in a lab, so real-life results are unknown. The team plans to test it in real hearing aids and is working on a version that uses eye tracking so users can steer it3.
What happens next for noisy-place hearing tech?
The researchers have patented BOSSA and are looking for companies to license it3. Whether it works as well outside the lab, and in a hearing aid, is still to be shown.
Other biologically inspired systems are also in development, including a physiologically inspired algorithm that models how the auditory system monitors and isolates sounds4, and systems that combine audio and visual cues5. Our report on smart glasses for hearing in noise covers a related idea, and so does how struggling to hear in loud places affects the brain.
You do not need to wait for future technology to get help with noisy places. Today’s hearing aids can include directional microphones and noise reduction, and a hearing test shows what would suit you.
“Too many people delay taking action on their hearing concerns, but early intervention can make a real difference,” said Michael Field, a Clinical Audiologist at Hearology. “Innovations like BOSSA are exciting, and in time they may reach more people. In the meantime, the biggest issue is that not enough people with hearing loss take that first step and seek help. Acting early opens the door to better long-term outcomes.”

Frequently asked questions
What is the cocktail party problem?
Which hearing aid solves the cocktail party problem best?
Is the BOSSA algorithm available in hearing aids yet?
Why is it hard to follow speech in noisy places with hearing loss?
References:
- [1] A. D. Boyd, V. Best and K. Sen, "A brain-inspired algorithm improves cocktail party listening for individuals with hearing loss," Communications Engineering, vol. 4, art. 75, 2025, doi: 10.1038/s44172-025-00414-5. Available:
- [2] K. F. Chou, A. D. Boyd, V. Best, H. S. Colburn and K. Sen, "A biologically oriented algorithm for spatial sound segregation," Frontiers in Neuroscience, vol. 16, 2022, doi: 10.3389/fnins.2022.1004071. Available:
- [3] Boston University, "New BU-developed algorithm could help hearing aid users," The Brink, Apr. 22, 2025. Accessed: Nov. 16, 2026. [Online]. Available:
- [4] A physiologically inspired model for solving the cocktail party problem, PMC. Available:
- [5] Solving the cocktail party problem using multi-modal speech enhancement, J. Acoust. Soc. Am., vol. 154, no. 4 suppl., 2023. Available:
