Effectiveness of Hearing-aid Based Wind-noise Algorithm

Overview

Wind-noise is highly disturbing to hearing impaired individuals wearing hearing aids who wish to participate in outdoor conversations where wind is present or during activities such as walking or running. In these situations, wind noise significantly reduces signal-to-noise ratio and, consequently, the intelligibility of speech and sounds may be significantly impaired. This negative effect is exacerbated with the use of directional microphone schemes in the hearing iads. The objective of this project is to determine the efficacy of the MH Acoustics' multi-microphone wind-noise reduction invention for the digital hearing aids market. MH Acoustics' wind noise reduction technology is unique since it provides instantaneous convergence while maintaining directionality of the microphone array. Current commercial technologies do not provide this feature. We are hypothesizing that, due to the design of the algorithm, speech perception ability and sound quality perception will be better than that available with traditional directional and/or omnidirectional microphone schemes in windy environments.

Two groups of subjects will be recruited to participate: Normal hearing adults and adults with mild to moderately severe hearing loss, ages 18-65. Pure tone audiometrics (re ANSI, 1996) will be done to ascertain the hearing sensitivity through 6ooo Hz. Normal hearing will be defined as thresholds at or better than 20 dB HL (re ANSI, 1996). The only exclusion criterion for the group exhibiting hearing loss is that no thresholds up to and including 3000 Hz will exceed 75 dB, so as to minimize the inclusion of subjects with "dead regions" in the cochlea. Subjects will be seen for four visits to the laboratory. The first visit will involve documentation of informed consent, and measurement of hearing thresholds. The second, third and fourth visits will consist of testing with the following measures (in random order for each subject) to determine if the various implementations of the wind noise reduction algorithm 1) impact speech perception ability, and/or 2) impact sound quality perception. Each session will take approximately 1.5 hours, with a maximum of six hours over all the sessions. The two speech perception tests that will be utilized include : 1) Connected Speech Test and Hearing in Noise Test, 2) The Hearing in Noise Test (HINT) (Nilsson, Soli, \& Sullivan, 1994; Koch, Nilsson \& Soli, 1995. Overall Impression of Sound Quality and Ratings of Annoyance will be used as subjective, or self-reported, measures of preference. Overall Impression will be influenced by the audibility and masking effect of the noise bursts, whereas Annoyance ratings are significantly correlated to the high frequency emphasis of the stimulus, a potential impact of the extreme suppression conditions (e.g., -18 dB) (Warner \& Bentler, 2002; Miedema \& Vos, 2003). Both measures will be analyzed as a function of the different time constants, gain reduction levels, and level of presentation.