Improving speech intelligibility in hearing aids. Part I: Signal processing algorithms

The improvement of speech intelligibility in hearing aids is a traditional problem that still remains open and unsolved. Modern devices may include signal processing algorithms to improve intelligibility: automatic gain control, automatic environmental classification or speech enhancement. However, the design of such algorithms is strongly restricted by some engineering constraints caused by the reduced dimensions of hearing aid devices. In this paper, we discuss the application of state-of-theart signal processing algorithms to improve speech intelligibility in digital hearing aids, with particular emphasis on speech enhancement algorithms. Different alternatives for both monaural and binaural speech enhancement have been considered, arguing whether they are suitable to be implemented in a commercial hearing aid or not.