The development of speech processing strategies for the University of Melbourne/cochlear multiple channel implantable hearing prosthesis.

Publisher's permission requested and denied.The speech processing strategies that have been used with the University of Melbourne/Cochlear multiple channel implantable hearing prosthesis have been developed systematically from the inaugural one that extracted the second formant and presented this on a place coding basis and the voicing frequency which determined the rate of stimulation. Speech processing has also depended heavily on biological research to ensure that the stimulus parameters used or the operative approach did not damage the spiral ganglion cells it was hoped to stimulate. The advances in speech processing from Melbourne primarily have been to extract more features and spectral information and present this on a place coding basis. This has led to a progressive improvement in speech perception, and a small number of patients can achieve nearly 100% correct scores for open sets of phonetically-balanced words using electrical stimulation alone