The task of recognizing people from their voices is termed as speaker recognition. Such systems extract features from speech, model them and use them to recognize the person from his/her voice.