Multi-Channel Dynamic-Range Compression Techniques for Hearing Devices

Abstract

Dynamic-range compression (DRC) is an important component in hearing aid devices (HADs). Research of multi-channel DRC design and real-time implementation has been carried out in the last few decades. The trade-offs of every DRC system include the frequency resolution, computational complexity and processing time delay. In this thesis, a crossover filter bank based nine-channel DRC with an optimized structure is proposed. A polyphase extension of this approach is then applied and a compensation filter is proposed to reduce the distortion. Then a subband filter bank is designed and implemented for multi-channel DRC to reduce the computational complexity. The final method to optimize a multi-channel DRC is to use Equalization technique in frequency domain, which further reduces the computational complexity. In this work, all the methods are implemented on a smartphone to work as an assistive device to hearing aids. Objective and subjective evaluation of the developed methods show the improvement in quality and intelligibility.