Speech-Processing Strategies and Their Implementation

Since 1979, the Australian speech-processing strategy has been based on the presentation of an estimate of F 2 coded by electrode position and F 0 coded by pulse rate. Although providing limited information, this strategy has produced good results with significant hearing-alone performance. This paper describes a number of strategies that provide further speech information in an attempt to increase hearing-alone performance to a level where the cochlear implant is able to operate in its own right rather than as an adjunct to lipreading. The strategies are all based on the addition of F 1 to the existing strategy. Both electrode and temporal coding of F 1 is described, and the performance and percepts produced are discussed. Amplitudes of the two formants must be carefully controlled to avoid masking. The implications of the strategies on the design of hardware are described.