Wilson et al. (1992) have developed signal processing strategies which have produced improvements in speech understanding for patients who use the Ineraid cochlear implant. The long term goal of the research proposed here is to understand some of the processes which underlie the improvements.
The specific aims of the research are: (i) to assess and compare frequency resolution in three groups of patients (those with excellent, average and poor speech understanding with their present Ineraid) when those patients are fitted with the (i) Ineraid processing strategy, (ii) with the CIS strategy and (iii) with the CIS/virtual channel strategy. Resolution will be assessed at three levels of loudness and in experimenter-controlled loudness balanced conditions (our usual approach) and in processor-controlled loudness balanced conditions, (ii) to determine, by measuring the outputs of the processors, the minimum differences in signal levels across channels which support frequency resolution, (iii) to relate changes in speech understanding, when patients are first fitted with the two new processing strategies, to changes in frequency resolution, (iv) to relate longitudinal changes in speech understanding, for patients fitted with the CIS or CIS/virtual strategies, to changes in frequency resolution, (v) to determine whether a hybrid CIS/peak picker processor will improve both speech recognition and low frequency resolution (vi) to investigate the role of dynamic range in speech understanding by systematically reducing the dynamic range available for signal coding, (vii) to assess which aspects of amplitude envelope cues contribute to voicing and manner distinctions in consonant identification.
Showing the most recent 10 out of 46 publications
