Abstract
Algorithms designed to improve speech intelligibility for those with sensorineural hearing loss (SNHL) by enhancing peaks in a spectrum have had limited success. Since testing of such algorithms cannot separate the theory of the design from the implementation itself, the contribution of each of these potentially limiting factors is not clear. Therefore, psychophysical paradigms were used to test subjects with either normal hearing or SNHL in detection tasks using well controlled stimuli to predict and assess the limits in performance gain from a spectrally enhancing algorithm. A group of normal-hearing (NH) and hearing-impaired (HI) subjects listened in two experiments: auditory filter measurements and detection of incremented harmonics in a harmonic spectrum. The results show that NH and HI subjects have an improved ability to detect incremented harmonics when there are spectral decrements surrounding the increment. Various decrement widths and depths were compared against subjects’ equivalent rectangular bandwidths (ERBs). NH subjects effectively used the available energy cue in their auditory filters. Some HI subjects, while showing significant improvements, underutilized the energy reduction in their auditory filters.
Cite
CITATION STYLE
DiGiovanni, J. J., & Nair, P. (2006). Auditory filters and the benefit measured from spectral enhancement. The Journal of the Acoustical Society of America, 120(3), 1529–1538. https://doi.org/10.1121/1.2221554
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.