Adaptive signal processing for interference cancellation in hearing aids

Interference signals which superimpose the desired speech signals are a severe problem for hearing impaired people, since they suffer from a reduced speech intelligibility which cannot be completely compensated by an increase of the input signal level with hearing aids. Directional signal processing with beamformers and noise reduction try to increase the speech intelligibility by enhancing the signal-to-noise ratio for the disturbed input signals.In this contribution, first differential beamformers are described and analysed under the consideration of hearing aid constraints, such as the microphone distance and head shading effects. Differential beamformers are analysed, adaptation procedures are deduced and real-word hearing aid problems are addressed. Also, a performance analysis tool is described for the measurement of adaptive beamformers.In a second step, a combined noise reduction and hearing aid gain optimization procedure is presented which optimizes the speech intelligibility for hearing impaired persons under consideration of the hearing loss and the instantaneous signal-to-noise ratio of the ambient sound field.