The paper describes the possibilities of modern emotion recognition which can be also helpful for speaker verification systems. A proper choice of feature vectors is one of the most important tasks. Two important vectors have been chosen as the background for efficient system of emotion recognition: fundamental speaking frequency and time-energy distribution vector. The proposed method permits to compare specific features of the current speaker voice with the computer base containing stored vectors of different emotions, recorded during learning sessions. As the results, the distances between these vectors have been obtained. The possibility of hardware realizations, including qualities and disadvantages of some voice processing components, have been also discussed