The University of Iowa
The University of Iowa News Services Home News Releases UI in the News Subscribe to UI News Contact Us

 

CONTACT: BECKY SOGLIN
2130 Medical Laboratories
Iowa City IA 52242
(319) 335-6660; fax (319) 335-8034
e-mail: becky-soglin@uiowa.edu

Release: Sept. 27, 1999

UI study: adjusting speech pitch may compensate for certain types of hearing loss

IOWA CITY, Iowa -- Amplifying sound through a hearing aid does not always help people with hearing loss better understand speech. However, compressing sound frequency -- decreasing its pitch -- through new computer technology seems to help people with high-frequency hearing loss better understand sounds, according to University of Iowa research findings.

"Theoretically, there was potential promise of taking speech to lower frequencies, so we wanted to try this technique one more time using new computer technology," said Christopher W. Turner, Ph.D., UI professor of speech pathology and audiology and lead investigator. "We were able to lower speech frequencies (pitch) to the region where certain individuals have usable hearing."

High-frequency hearing loss affects millions of people as they age and can also be caused by disease, injury, exposure to excessive noise or medication with adverse side effects. Hair cells in the ear normally transmit messages to the brain for interpretation. In people with severe high-frequency hearing loss, the hair cells at these levels are dead or injured. An option, as in this study, is to lower sound frequencies so that healthy hair cells at the lower frequencies can perceive the sound.

Individuals with high-frequency hearing loss have difficulty perceiving the softer sounds of speech or words with "f," "s" or "sh." They may also have trouble hearing women's and children's voices or comprehending others' speech amidst background noise.

Turner said that in contrast with earlier research, improved computer programs can change speech in a way so that it can be manipulated yet still be understandable. In addition, the UI study focused on people with hearing loss only in the high frequencies, so there was more likelihood of their benefiting from the technique.

Turner and co-investigator Richard Hurtig, Ph.D., UI professor and head of speech pathology and audiology, studied three people with normal hearing and 15 people who could hear lower-pitched sounds normally but had severe hearing losses for higher-pitched sounds.

The participants listened to nonsense syllables at normal and adjusted frequencies and volume. Using nonsense sounds allowed researchers to better track the effectiveness of the

frequency adjustment. Listeners might otherwise have "understood" familiar words because of prior knowledge.

Turner described how new technology helped make the research possible.

"You can't simply tape-record speech then slow it down because that would change the important timing cues of some sounds," he said. "Also, over time you couldn't keep up with the conversation. In addition, older technology shifted the frequencies in a way that muddled sounds and didn't maintain the frequency ratios within those sounds. The newer technology uses a commercially available musical algorithm to preserve the frequency ratios with a negligible change in the timing of speech."

Turner and Hurtig have also developed their own algorithm that would be even faster. Turner used a visual analogy to explain the importance of frequency ratios.

"The brain is set up to perceive ratios so we don't have to recognize things by remembering absolute numbers. When the brain perceives a square, it recognizes that all sides are equal, regardless if the sides are one inch or two inches long," he explained.

Similarly, the brain perceives frequency based on ratios, not absolute frequency. A syllable uttered by a man may show frequency peaks at 500, 1500 and 2500 hertz, while the same syllable uttered by a woman could peak at 700, 2100 and 3500 hertz. The computer technology will shift the woman's voice down about 70 percent so that it peaks at approximately the same points as the man's voice yet maintains its original frequency ratios. The speech cues are then perceived by working hair cells and can reach the brain. The manipulation also shifts the male's voice to even lower frequencies.

"Participants said the adjusted female voice was understandable but sounded unusual," Turner said. "However, they agreed it was better to understand the speech. I'm glad we didn't give up on this idea of frequency compression."

Turner said the compression technology is not a cure-all but might eventually be used to create more sophisticated hearing aids. The researchers' next step, however, is to test frequency adjustment using real sentences. Turner and Hurtig will also test how well the adjustment improves people's comprehension when background noise is present.

The study was supported in part by a grant from the National Institute on Deafness and Other Communication Disorders, a branch of the National Institutes of Health.

The findings appeared in the August issue of the Journal of the Acoustical Society of America. In September, Turner was appointed as an associate editor of the journal.

University of Iowa Health Care describes the partnership between the UI College of Medicine and the UI Hospitals and Clinics and the patient care, medical education and research programs and services they provide.