Study On Key Technology Of Digital Hearing Aid Based On DSP

This paper presents research on hardware and software of DHA (digital hearing aid) based studing of overseas advanced technology,and it also describes the key technology for design of DHA based DSP TMS320VC5416 from TI with theory about physiological acoustics and speech signal processes.The DHA system introduced consists of circuit and algorithms. Construction of circuit is include of DSP TMS320VC5416 as central MCU and TLV320AIC23B from TI as Analog Interface Circuit.Function of the system is performed in DSP mostly,so design of hardware is easier than algorithms in a certain extent. Emphasis of this paper is research in algorithms for speech signal processing.The DHA system applies WRDC (wide dynamic range compression), frequency transposition compression and adaptive noise reduction etc. Firstly we study the algorithms used for digital signal processing of speech for the hearing impaired currently,than we present devise of spectrum dependent WRDC which can remedy deficiency of previous algorithms.The devise perform functions for different loudness compensation in different frequence area facility.Most basic properties of speech signal remain invariant across procession,while it receives nonlinear amplification variously. we also give an algorithm of frequency transposition compression especially for persons with hearing loss seriously. Frequency spectrum in high frequency area is compressed linearly to low to fit imparied's residual dynamic range in accodance within their audiologic curves and properties of speech signal remain invariant relatively. A noise reduction algorithm used in speech signal processes which is referred to as LMS adaptive noise reduction, is also designed to reduce wide-band noise but improve Signal to Noise Ratio. All algorithms aims for less distortion and better comfort.Than we emulate them on MATLAB,and obtain evaluation result to prove availability ofthe algorithms.At last the algorithms are replanted on DSP experimental kit for speech signals processing.we build project with C and Compile language in CCS.A11 programms runs correctly and exact result is gained to prove availability DHA system based DSP.The system can set parameter of frequency response depending imparied's audiogram and fit their need for loud compensation.This system is well available for study of advanced algorithms of DHA.