In this paper we present a comparison of several parameters that are commonly used in the fields of speech and speaker recognition. Analysis of the variance was the evaluation method employed. The parameters that were tested were the Mel CEPSTRAL [1] parameters, the LPC parameters derived from a PLP2] analysis, the LPC and Mel CEPSTRAL parameters derived from RASTA-PLP [3] analysis and all the differential parameters of the above (A-Mel Cepstral, e.t.c). The testing procedure used a 500-word speech database of the Greek language. We concluded with several results which show the performance of the different parameters for speech and speaker recognition, for each phoneme category and for each sex.