The Autoregressive Model is used as a tool to design a recognition system that is presented by speaker identification. The main goal of this paper is to design a speaker identification system by using the autoregressive model to identify the identity of speaker according to the voice frequency, the speaker is being asked to say a certain word then it’s matched to the same word stored earlier in the database. This research is based on speech recognized words using the Autoregressive Model based on a limited dictionary.