Improving Security Accesibility Using Speech Recognition System Based on Pitch, Cepstrum and Resonance Frequency

Abstract

Access to the system becomes very important to note. Voice character becomes a significant differentiator to detect one person with another. The introduction discusses ways to use a userfriendly GUI, where each sample can be used and matched based on pitch, cepstrum and formant comparisons. The results of experiments using Internal and External software show stable results on the system being designed. The system can identify every user recorded in the data by bringing up photos that the user detected. Delay in decision-making by the system can be compared with the application of appropriate algorithm. The results show that the system can work optimally and accurately.