This paper has reported on the comparative evaluation of gender recognition algorithms. The work that we described here is the two pitch detection algorithms and the related techniques including preprocessing, post-processing and extraction of pitch pattern. The gender based differences in human speech are partially due to physiological differences such as vocal fold thickness or vocal tract length and partially due to differences in speaking style. As a result of changes in shape of human vocal tract during generation of different words, resonance frequencies of vocal tract, formants, also changes. Using this phenomenon, we extract voice features of each command and we have implemented a gender recognition system. In this work we have demonstrated the importance of information in the excitation component of speech (pitch) for gender recognition task. Vowels and Words which are combination of vowels and consonants as well as group of voiced and unvoiced sounds, are chosen as database. The recognition performance depends on the training speech length selected for training to capture the speaker-specific excitation information. Larger the training length, the better is the performance, although smaller number reduces computational complexity. Since it’s obvious that the voice signal tends to have different temporal rate, the alignment is important to produce the better performance. This paper presents the viability of Cepstral, autocorrelation coefficients and linear predictive coding to extract gender biased features such as pitch (fundamental frequency) and formant frequencies. In feature matching step, Euclidean distance method is implemented to compare the test patterns.