Quantization analysis for fixed-point implementation of speech processing for the hearing impaired

Over the past decade, real time digital signal processing (DSP) has found several new applications in the biomedical arena. One of these has been in the area of hearing aids. This paper presents quantization analysis of speech processing algorithms, for the hearing impaired, for implementation on Texas Instruments' (TI) TMS320C54x (C54x) fixed point DSP chip. These algorithms were initial implemented on TI's floating-point processor, the TMS320C3x (C3x). Portable C3x units were tested on hearing impaired subjects to assure the efficacy of the various speech processing algorithms. The algorithms that proved successful are currently being ported to the CS4x since this chip is being rapidly reduced in size and is on track to be an extremely low power DSP chip, ideal for eventual use in a behind the ear (BTE) or in the ear (lTE) hearing aid.