|The Cape Peninsula University of Technology (CPUT) Electronic Theses and Dissertations (ETD) repository holds full-text theses and dissertations submitted for higher degrees at the University (including submissions from former Cape Technikon and Peninsula Technikon).|
Hearing loss simulation
MetadataShow full item record
This document gives a report on the research that has been done to simulate hearing loss. People working with the hearing impaired have no idea of what and/or how the hearing impaired person hears sound. An instrument that enables a normal hearing person to hear what a hearing impaired person hears, is referred to in this document as a Hearing Loss Simulator (HLS). An investigation of the feasibility and practicability of the abovementioned instrument, has led to the development of the HLS by making use of a distinct type of technology called Digital Signal Processing (DSP) technology. Before hearing loss can be simulated, the hearing loss first needs to be determined. A study of different procedures and methods for screening hearing has led to the incorporation of an existing instrument called an Audiometer. An audiometer is an instrument that determines the hearing loss by making use of pure tone sine waves. The results are then plotted on a graph called an Audiogram. The results of other methods that determine the hearing loss can also be transferred to the Audiogram. The Audiogram's information which is, in fact, the frequency response of the ear, is stored in a computer and is utilized to realize the HLS function. Six different DSP based methods were studied to shape any audio information according to a specific frequency response. The optimum method was identified and then implemented. The various methods are the following: Filter bank method Inverse Discrete Fourier Transforms (IDFT) method Inverse Fast Fourier Transforms (IFFT) method Chirp-z method Wavelet method Yule-walker method The ITFT method was identified as the optimum method and was therefore implemented. The algorithm to realize this method, was carried out by doing the IFFT calculation on computer and subsequently doing the filtering on a DSP processor called an ADSP-2181 processor. The audio information under investigation is fed into the instrument, it is then filtered according to the audiogram information and then sent out again. Any normal hearing person who wants to investigate the hearing impairment of the hearing impaired person can listen to sound reproduction through either a set of headphones or through a free field. This process is referred to in this dissertation as Hearing Loss Simulation.