The cochlear implant (CI) is an electronic device implanted under the skin with electrodes positioned in the cochlea to stimulate the auditory nerve. Electrical currents induce action potentials in the auditory nerve fibers and these are transmitted to the brain. In this master thesis, a special attention will be done to the speech processing strategy, which is considered the key module for an efficient performance of such implants. The goal of the speech processing unit is to transform the input signal from the microphone into a set of stimuli that can be interpreted by the brain. One approach for this transformation, and a very successful one, is to reproduce the damaged or missing physiological functions that have to be bypassed by the neuroprosthesis. The best way to achieve an excellent performance would be to mimic the exact physiological mechanism. However some important details of the hearing mechanism are still being discussed.
In this thesis different speech processing strategies are presented in a chronological way, as developed and used by the CI. Their enhancements respect to the previous strategies and their handicaps and weakness are discussed as well. Finally, a novel speech strategy developed by Chen et al. 2009, the Simulated Phase-Locking Stimulation (SPLS), is also be outlined. Results of the simulations done with MATLAB of the Continuous Interleaved Sampling (CIS) and the SPLS strategies using different important parameters, such as frequency range, number of channels, type of filter and kind of bandwidth, are discussed. In order to compare both strategies, test data from the H-LAD (Heidelberg Lautdifferenzierungstest) are acoustically simulated with both strategies (CIS and SPLS). The obtained simulations were used to realize the test in 5 different normal listeners. The results of the H-LAD (Heidelberg Lautdifferenzierungstest) using the simulations are presented in the last part of the thesis. Further improvements are presented in the conclusion.