Speech voice is created by vibrations of the vocal cords and can be represented as a set of vocoder parameters at some fixed frame rate. The rate of vibration of the cords is named fundamental frequency or F0. Although a number of algorithms for this purpose are described in the literature, none of the them have proven to be highly accurate. The reason may be related to the speech signal varies with time, the acoustic signal is not always voiced, or background noise usually prevents F0 algorithm from getting the correct information. The task of the student is need to study and investigate some of the previous F0 estimation algorithms, evaluate the performance of each of them in noisy environments, and then try to develop a new method for estimating the F0. Programming skill in Python and Matlab, and English knowledge is required.