Abstract— This paper presents Authentication of Speaker’s identity by an User’s voice. In order to authenticate speaker’s identity real time speech signal is compared with the pre-recorded voice signals in the database. Here the sampled signal’s power spectrum density is compared with power spectrum density of pre-recorded voice and on successful match the name of the particular speaker is displayed. The possible application of this project could to operate personalized digital gadgets with voice commands. Speaker recognition, in computer science, the ability of the computer to understand the spoken words for the purpose receiving commands and data input from speaker. Some systems have been developed that can recognize limited vocabularies are spoken by specific individual but developing a systems that deals with a variety of speech and accent, as well as with the various ways in which a request or a statement can be made has so proved a daunting task for system designer. It is based on either text dependent speech. Key words: Speech recognition, Speaker recognition, FFT, DCT, Feature extraction