Directional microphone effects on measurement of the HINT

Directional microphones are one of the few solutions available to improve the signal-to-noise ratio (SNR) in noisy environments for listeners with hearing loss. In addition to changes in SNR, directional hearing aids have a characteristic +6 dB per octave low frequency reduction in gain compared to omnidirectional hearing aids. (Killion et al., 1998) These gain reductions are sometimes compensated through a low frequency gain boost during the hearing aid fitting process. This research examined the effect of directional microphone equalization on sentence intelligibility in noise.; Digital recordings were mastered through a directional and omnidirectional microphone mounted in a full shell hearing aid case worn by the author, who was positioned in a sound field between four loudspeakers. Sentences from the Hearing In Noise Test (HINT) were delivered from a front loudspeaker (0 degrees) while the speech noise portion was delivered from three loudspeakers positioned at 90, 180, and 270 degrees. The speech level was presented for each HINT sentence at a constant level (65 dB SPL), while the competing noise was varied in 2 dB steps from −10 to +10 dB SNR. Digital sound recordings were made for omnidirectional, directional, and frequency-response equalized directional microphones with multiple sentences recorded at each SNR. These recordings were mastered to three CD-R disks, one for each microphone configuration. The audio output was passed through a +6 dB/octave high pass filter before being fed to a clinical audiometer in order to provide some frequency shaping for the subjects' hearing losses. A 20-sentence HINT was administered using each CD-R as the source through a clinical audiometer into one ear of 10 subjects, each having a similar high frequency hearing loss in the tested ear. SNR scores were thus obtained for each subject in all three microphone configurations.; Results showed that subjects performed 3.5 dB more poorly on the HINT when tested by the omnidirectional recordings than with the directional recordings. No statistical differences were present between the unequalized and equalized directional test results. Therefore, the use of gain equalization in clinical fittings for patients with near normal low frequency hearing is not recommended. Low frequency gain compensation is not suggested for patients with low frequency hearing thresholds falling in the moderate range or better.