Octav Chipara
University of Iowa, IA·Computer Science
Tens of millions of US adults have hearing loss. Left untreated, it can lead to social isolation, depression and dementia.
- Fields
Computer Science
Engineering
Biomedical Engineering
- Topics
hearing aids
hearing loss
personalization
- Grants
- 1 known grant, through Aug 2027
What they work on
- Develop shallow personalization methods that set hearing-aid band gains, compression and noise reduction according to the listening context, for users with mild-to-moderate sensorineural hearing loss.
- Develop deep personalization methods for hearing aids that use deep neural networks to amplify sound, tuned by combining user feedback with EEG (electroencephalogram) recordings of brain activity.
- Compare the two approaches on how much personalization they allow, how much user effort they require, and how well they work across levels of hearing-loss severity.
More about the research
Hearing aids are the main treatment, yet only 15–30% of people who could benefit from them use them. One reason is that a hearing aid has many adjustable settings, and finding the right ones for a given person is hard. This project develops signal-processing algorithms and machine-learning methods that let people with hearing loss tune their hearing aids to their own needs. The goals are clearer speech, better sound quality and settings the user actually likes.
Why it matters
Many people with hearing loss abandon or never adopt hearing aids because the devices are hard to configure. Easier personalization could raise hearing-aid satisfaction and use, including for over-the-counter devices and for people with severe hearing loss. That in turn may reduce the communication problems and social isolation linked to depression and dementia, and help patients take a more active role in their own hearing care.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Kevin Sitek
Northwestern University, IL·Neuroscience·Early-career
Use the frequency-following response, a scalp recording of how faithfully early auditory structures encode a sound, to determine when and where auditory-motor interactions occur during self-generated sounds, subcortically or in primary auditory cortex.
Carol Espy-Wilson
University of Maryland, MD·Engineering
Extend a speech inversion tool, which already estimates tongue and lip movements from audio, to recover nasality and voicing by training neural networks on acoustic features.
Susan Voss
Smith College, MA·Biomedical Engineering
Expand and update the online wideband acoustic immittance database, now over 9,900 subjects from 41 studies, to define normal measurements and show how specific middle-ear diseases change them.
The school
University of Iowa: 32 labs on PI Pages, $703M a year on research, 418 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.