This paper provides a concise exploration of speaker recognition using DL, focusing on the analysis of speech signals and their transformation through Fourier analysis. The study employs Fast Fourier Transform to quantify dissimilarities between speakers and validates its efficacy in speaker recognition. As human voices possess distinct characteristics, voice-based recognition emerges as a valuable biometric technique. This research delves into the realm of Speaker Recognition technology. Spectrum analysis involves converting time-domain signals into frequency-domain representations using Fourier Transform and Convolutional neural network (CNN). Fast Fourier Transform (FFT) is harnessed to extract frequency content from analogue signals. DNN algorithms play a pivotal role in speaker classification. Through these approaches, the paper advances our understanding of speaker recognition within the domain of DL.