Gender information is a distinctive and the most important property in a speech. Determination of this information from a speech signal is a substantial subject. Gender information used for various purposes in many applications, provides the less error rate by defining the gender-dependent speech/speaker models. In this study, a system determining the gender of a speaker with no dependency from a text is proposed. In the proposed system, speech records represented by Mel Frequency Cepstral Coefficients (MFCC), Linear Prediction Cepstral Coefficients (LPCC) and Perceptual Linear Prediction Coefficients (PLP) features are classified according to the genders by using GMM model. In the study, the effect of feature type and its dimension and the number of GMM components on the success are comparatively investigated. In the experiments, the best result is obtained as 99.37% with 16-coefficient MFCC and 8-component GMM.