Abstract
Objectives:Current hearing aids have a limited bandwidth, which limits the intelligibility and quality of their output, and inhibits their uptake. Recent advances in signal processing, as well as novel methods of transduction, allow for a greater useable frequency range. Previous studies have shown a benefit for this extended bandwidth in consonant recognition, talker-sex identification, and separating sound sources. To explore whether there would be any direct spatial benefits to extending bandwidth, we used a dynamic localization method in a realistic situation.Design:Twenty-eight adult participants with minimal hearing loss reoriented themselves as quickly and accurately as comfortable to a new, off-axis near-field talker continuing a story in a background of far-field talkers of the same overall level in a simulated large room with common building materials. All stimuli were low-pass filtered at either 5 or 10 kHz on each trial. To further simulate current hearing aids, participants wore microphones above the pinnae and insert earphones adjusted to provide a linear, zero-gain response.Results:Each individual trajectory was recorded with infra-red motion-tracking and analyzed for accuracy, duration, start time, peak velocity, peak velocity time, complexity, reversals, and misorientations. Results across listeners showed a significant increase in peak velocity and significant decrease in start and peak velocity time with greater (10 kHz) bandwidth.Conclusions:These earlier, swifter orientations demonstrate spatial benefits beyond static localization accuracy in plausible conditions; extended bandwidth without pinna cues provided more salient cues in a realistic mixture of talkers.
Highlights
Hearing prostheses historically have had a limited bandwidth of amplification in the extended high frequency region above 5 kHz due to multiple aspects of their design, from lower sampling rates to the acoustic tubing used in many models of hearing aids to the power required to provide audible output and gain to achieve audibility for users who have more than mild high frequency hearing loss
To reduce the number of measures necessary to report in orientation behavior, we looked for potential redundancies in the measures using a principal components analysis, averaged across conditions and angles, using varimax rotation
Future research is warranted to explore which measures capture all spatial benefits, but by using a suite of behavioral measures, potential spatial benefits can be more readily captured across different situations. These results show there is useful spatial information in the 5 to 10 kHz region without pinna cues that allow for an earlier, swifter orientation towards a target
Summary
Hearing prostheses historically have had a limited bandwidth of amplification in the extended high frequency region above 5 kHz due to multiple aspects of their design, from lower sampling rates to the acoustic tubing used in many models of hearing aids to the power required to provide audible output and gain to achieve audibility for users who have more than mild high frequency hearing loss. There have been mixed conclusions regarding the benefit of extended high frequency amplification in the literature investigating aspects of subjective preference and speech understanding. For those with residual hearing above 6 kHz, qualitative preference for extending the bandwidth from 5 to 5.5 to 8 to 11 kHz has been demonstrated in listeners with hearing impairment (Ricketts et al 2008; Brennan et al 2014; Van Eeckhoutte et al 2020). Amos and Humes (2007), found no difference in word recognition in mild noise [+5 and +20 dB signal-to-noise ratio (SNR)] for elderly hearing-impaired listeners when the upper frequency limit was extended from 3.2 to 6.4 kHz. Donai and Halbritter (2017) has shown talker-sex identification benefits for speech high-pass filtered above 5 kHz. Amos and Humes (2007), found no difference in word recognition in mild noise [+5 and +20 dB signal-to-noise ratio (SNR)] for elderly hearing-impaired listeners when the upper frequency limit was extended from 3.2 to 6.4 kHz. Silberer et al (2015) found limited improvement in speech recognition with extended bandwidth when congruent visual information was available for normal-hearing listeners
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
