Use of Modulation-based Digital Noise Reduction in Hearing Protection Devices

A new study published in the journal Human Factors: The Journal of the Human Factors and Ergonomics Society concluded that modulation-based digital noise reduction may potentially enhance the performance and user acceptance of hearing protection devices. The abstract is provided here;for a copy of the complete study or a subscription to the journal, please visit www.hfes.org

Title: Modulation-based digital noise reduction for application to hearing protectors to reduce noise and maintain intelligibility (2009;51,1, 78-89).

Authors: K. Chung, J. Tufts, and L. Nelson

Objective: The objective of this study was to test the effects of modulation-based digital noise reduction (MB-DNR) on noise levels, speech intelligibility, and listening preference in four real-world noises.

Background: A significant challenge in hearing protection device (HPD) design is to reduce ambient noise levels while maintaining users' ability to understand speech. MB-DNR technology, currently employed in hearing aids, potentially could be used to achieve these objectives.

Methods: Speech and noise calibrated to signal-to-noise ratios (SNRs) of 0, -5, and -10 dB in the sound field were recorded at the outputs of two digital hearing aids in the ears of a Knowles Electronic Manikin for Acoustic Research with and without MB-DNR activated (i.e., NR and noNR, respectively). Listeners' speech intelligibility scores and sound quality preferences were evaluated while they listened to the recordings presented via ER-3A insert earphones at 85 dB SPL.

Results: MB-DNR reduced the overall noise level by approximately 4 to 7 dB. Listeners obtained significantly higher speech intelligibility scores in the NR condition at an SNR of -10 dB in two noises and similar scores in the noNR and NR conditions in all other cases. They preferred the NR condition in all cases.

Conclusion: MB-DNR reduced overall noise level, enhanced sound quality, and maintained or improved speech intelligibility in the four military noises tested.

Applications: MB-DNR algorithms potentially could be incorporated into HPDs to enhance performance and increase user acceptance.