Speech is a signal which consisting of various complex information, like the information contain in a signal which to be communicate, language, speaker, region, emotions etc. Speech Processing is one of the important parts of signal processing and finds applications in Audio mining, Human computer interfaces, Assistive technologies, Telecommunication, Security etc. For the speaker identification and emotion analysis Gender Identification is a pre requites. By analyzing the features of different voice signals of male & female it can be determined that which parameters are creating a difference in between the male voice & female voice. For this a comparison between the different parameters present in a voice signal like energy, mean, median and standard deviation is done. It observed that in the graph of mean and standard deviation one can differentiate between a male voice and female as the statistical results are shown. In this paper an input voice signal taken as .wav file and added some noise or unwanted signal; the unwanted noise signals are removed from the original signal by reducing the intensities of high frequencies by averaging and Gaussian window for the smoothing and conv function to removing the pseudo numbers that added as a noise to the original signal