REAL-TIME MULTIBAND DYNAMIC COMPRESSION AND NOISE-REDUCTION FOR BINAURAL HEARING-AIDS

A multi-signal-processor set-up is introduced that is used for real-ti me implementation of digital hearing aid algorithms that operate on st ereophonic (i.e., binaural) input signals and perform signal processin g in the frequency domain. A multiband dynamic compression algorithm w as implemented which operates in 24 critical band filter channels, all ows for interaction between frequency bands and stereo channels, and i s fitted to the hearing of the individual patient by a loudness scalin g method. In addition, a binaural noise reduction algorithm was implem ented that amplifies sound emanating from the front and suppresses lat eral noise sources as well as reverberation. These algorithms were opt imized with respect to their processing parameters and by minimizing t he processing artifacts. Different versions of the algorithms were tes ted in six listeners with sensorineural hearing impairment using both subjective quality assessment methods and speech intelligibility measu rements in different acoustical situations. For most subjects, linear frequency shaping was subjectively assessed to be negative, although i t improved speech intelligibility in noise. Additional compression was assessed to be positive and did not deteriorate speech intelligibilit y as long as the processing parameters were fitted carefully. All nois e reduction strategies employed here were subjectively assessed to be positive. Although the suppression of reverberation only slightly impr oved speech intelligibility, a combination of directional filtering an d dereverberation provided a substantial improvement in speech intelli gibility for most subjects and for a certain range of signal-to-noise ratios. The real-time implementation was very helpful in optimizing an d testing the algorithms, and the overall results indicate that carefu lly designed and fitted binaural hearing aids might be very beneficial for a large number of patients.