Speaker recognition refers to the task of recognizing a person from a spoken utterance with the aid of a machine. The distinctive features extracted from the speech characteristics of various individuals are utilized to create a unique speaker model for each person. These models are then employed to identify an enrolled speaker from a pool of available speakers by comparing the features of the test speaker with the stored models. This paper explores various feature extraction techniques, including MFCC, LPC, RASTA-PLP, ZCR, and PSMFCC, among others. The paper also aims to explore various techniques for speaker modeling like GMM, I-Vector, ANN, DNN, SVM with an emphasis on Deep Neural Network (DNN) techniques. Additionally, this paper also presents a detailed comparative performance analysis of the techniques used in the domain of speech processing.