Trying a deep-learning noise-reduction feature for hearing aids
Evaluating a Deep Neural Noise-Reduction Algorithm for Hearing Aids in Varying Signal-to-Noise Conditions
This test tries a deep-learning noise-reduction feature in hearing aids to see if it helps people with mild-to-moderate cochlear hearing loss understand speech in noisy places.
Quick facts
| Phase | Not applicable |
|---|---|
| Study type | Interventional |
| Enrollment | 50 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Purdue University Academic / other |
| Locations | 1 site (West Lafayette, Indiana) |
| Trial ID | NCT07287774 on ClinicalTrials.gov |
What this trial studies
Participants with mild-to-moderate cochlear hearing loss will listen to speech stimuli presented with various levels of background noise while wearing a hearing-aid configuration that isolates single-microphone noise-reduction. The study compares several processing modes (no noise reduction, low, and high DNN-based noise reduction) across multiple signal-to-noise ratios to represent easier and harder listening conditions. All participants experience every condition in a counterbalanced order to limit order effects, and speech intelligibility is measured under each setting. The monaural, omnidirectional setup removes directional microphone benefits so the effects of the noise-reduction algorithms can be isolated.
Who should consider this trial
Good fit: Ideal candidates are hearing-aid candidates with mild-to-moderate cochlear (sensorineural) hearing loss, especially with higher-frequency loss and who have difficulty understanding speech in noise.
Not a fit: People with normal hearing, severe or profound hearing loss, conductive hearing loss, or neural hearing loss are unlikely to benefit from or qualify for this test.
Why it matters
Potential benefit: If successful, this approach could help some listeners understand speech better in noisy everyday settings and reduce listening effort.
How similar studies have performed: Prior research has shown that DNN-based noise-reduction can improve speech clarity for some listeners, but benefits vary with noise type, SNR, and individual hearing profiles.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * A hearing aid candidate with mild-to-moderate cochlear hearing loss, based on audiometric profile (at least 20 dB of hearing loss at 2000 Hz, with progressively worse hearing levels at higher frequencies). Exclusion Criteria: * Normal hearing * Severe or profound hearing loss * Conductive hearing loss * Neural hearing loss
Where this trial is running
West Lafayette, Indiana
- Purdue University — West Lafayette, Indiana, United States (Recruiting)
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.