Brey, Robert H.

Thirty‐six weanling guinea pigs were fed either a low (600 ppm) or normal (3000 ppm) diet of magnesium for 8 weeks. One half of each diet group received intramuscular injections of magnesium‐depleting drugs, furosemide and gentamicin. The other half were controls and received equal intramuscular injections of saline. Auditory brainstem responses were obtained from all animals before and after 8 weeks of treatment of diet and drugs to examine the effects of treatment upon hearing and auditory brainstem function. A three‐way analysis of variance of dietary magnesium, by drug and by sex, showed no significant differences in auditory brainstem wave V thresholds, wave V latencies, or interpeak wave I‐V latencies between the control and experimental groups. The low magnesium diet group, which received drugs, had significantly greater wave V auditory brainstem response amplitudes. Results can be explained on the basis of magnesium influencing the uptake of calcium into both the hair cells and associated brainstem pathways.

A two microphone adaptive digital noise cancellation technique was used to improve word-recognition ability of normally hearing and hearing-impaired subjects in the presence of varying amounts of multitalker speech babble noise and speech spectrum noise. Signal-to-noise ratios varied from -8 dB to + 12 dB in 4 dB increments. The adaptive noise cancellation technique resulted in reducing both the speech babble and speech spectrum noises 18 to 22 dB. This reduction in noise resulted in average improvements in word recognition, at the poorest signal-to-noise ratios, ranging from 37% to 50% for the normally hearing subjects and 27% to 40% for the hearing-impaired subjects. Improvements in word recognition in the presence of speech babble noise as a result of adaptive filtering were just as large or larger than improvements found in the presence of speech spectrum noise. The amount of improvement of word-recognition scores was most pronounced at the least favorable signal-to-noise ratios.

A nonlinear model of the human aural system is proposed. Utilizing a homomorphic transformation, the model is implemented in a digital computer and accounts for the Fletcher critical bands in hearing, the nonlinear relationship between input sound intensity and perceived loudness known as recruitment, as well as the thresholds of audibility and the maximum tolerance threshold. The transformed or output signal domain is identified as a perceptual space. It is shown that the model can be used to reproduce the Fletcher Munson equal loudness contours for normal hearing populations. Modification of easily measured parameters results in the generation of equal loudness contours for hearing impaired individuals. Through a combination of the forward homomorphic transformation for normal hearing to the perceptual domain and the inverse mapping utilizing parameters of a hearing impaired individual, it is shown that hearing intelligibility is dramatically improved. It is postulated that signal processing performed in the perceptual domain will produce results more pleasing and intelligible to the listener. Results of auditory testing utilizing the proposed model for both normal and hearing impaired populations are presented. Speech intelligibility scores are shown to increase by more than 40% (i.e., from 40% to over 80%) over preprocessed scores under headphones for hearing impaired subjects used in these tests.