This paper shows the design and implementation of an algorithm for users voice identification, including considerations on physiologic issues affecting the speech so that when users manifest these problems, lower rates on fake rejections decrease. For purposes of managing the contemplated physiologic problems, algorithm design also takes a standard algorithm that uses cepstral coefficients which include additional characteristics determined by voice acoustic analysis. A test including several records from people with a healthy voice and those with voice affections is carried out in order to observe the performance of the algorithm; thus, observing that when applying those characteristics in voice analysis a better result is achieved regarding the case when cepstral coefficients are implemented.