Study Uncovers Possible New Way to Help With Hearing Loss

Posted on December 27, 2013 by ECR Louisville in Blog, Hearing Loss


A study completed at Ohio State University has uncovered a breakthrough in solving a decades-long problem in helping those with hearing loss. Engineers and scientists who worked on the project have designed a possible new way to help these individuals understand what people are saying in the midst of background noise.The researchers utilized the most recent developments in neural networks to increase subjects’ ability to recognize words spoken to them from as little as 10 percent to as much as 90 percent, according to ScienceDaily. Their findings appeared in the Journal of the Acoustical Society of America.

How to get past background noise has for at least 50 years plagued designers of hearing technology. Often people with a hearing loss can’t make out conversation unless only one person speaks at a time.

According to PubMed Health, many factors cause hearing loss. Among the most common are age, a buildup of wax, a hole in the eardrum, fluid in the ear after an infection, regular exposure to loud noise, a scarred eardrum from multiple infections, and certain diseases like diabetes.

The National Institute on Deafness and Other Communication Disorders confirms that there’s a definite association between hearing loss and age. Among Americans between 45 and 64, 18 percent have experienced a loss. The figures rise to 30 percent for those between 65 and 74 and 47 percent for individuals at least 75 years old. Around 26 million Americans — 15 percent of the population — between 20 and 69 suffer high-frequency hearing loss as a result of being exposed to loud sounds or various types of noise. Only 20 percent of individuals whom a hearing aid could help actually use one.

The basis of the new technology is a computer algorithm that can analyze speech and delete a majority of background noise. The researchers initially tested 12 subjects with hearing loss and removed their hearing aids. The subjects heard speech recordings and repeated words they were fed.

Subjects repeated the test after scientists used the algorithm to remove any background noise from the recordings. Testing also involved determining the effectiveness of the algorithm against a constant noise, such as the hum of a machine, and then with a mixture of voices in the background.

The algorithm upped comprehension from 25 to nearly 85 percent on average for the background babble. For stationary noise like a hum, comprehension jumped from 35 to 85 percent.

Using 12 undergraduates without any hearing impairment, the team repeated the process. Scores for this group were lower than the results from the first group with the benefit of algorithm processing.

Additional work to refine and test the algorithm will be funded by a $1.8 million grant provided by the National Institutes of Health. The eventual goal is next-generation technology like a digital hearing aid, which could reside inside a smartphone.

Source: Vonda J. Sines

Avada Hearing Care Centers

Dec 10th, 2013