Implementation of a Two-Level Security Door Access Using Keypad and Voice Recognition

Abstract

It is possible today to automate a growing number of speaker-recognition tasks with such technologies as voice or speaker verification and voice identification. These systems have been applied to real-world security applications. They have, however, been undergoing modifications due to the poor level of security architecture that they provide to users. Hence there is the need, as has been done in this work, to develop a more efficient, user friendly and complex security system that can curtail illegal intrusion into special facilities. The system described in this paper uses matrix keypad which is interfaced with a microcontroller for first level security validation to monitor and control the execution of desired tasks within the keypad and voice recognition (KVR) system. A voice recognition Integrated Circuit is used for second level security verification. A tristate buffer is also employed to logically isolate the buses of the digital signal processing (DSP) chip and those of the microcontroller. From the research, tests were conducted for both noiseless and noisy environments and the recommended distance between the user and microphone in a noiseless environment using electrets type condenser microphone is 1.0 cm to 16 cm while under noisy condition it was found to be 1.0 cm to 6.0 cm. These results depend on the sensitivity of the microphone and the environmental conditions at the time. It can be deduced from the research that integrating the keypad and voice recognition design can help to optimize the security level of the system and by implication help in controlling unwanted intrusion into the special facilities such as museum, military armory, financial institution, etc.