International Journal of Artificial Intelligence and Neural Networks
Author(s) : AMOL C.WANI , PRABHAKAR V. MHADSE
This paper discusses the methodology for project named “Automation System Using Speech Recognition for Speaker Dependency through MFCC”. Automation system is growing to improve the quality; the goal of design of speaker dependent automation is to provide an efficient integration and higher security among appliances. Such automation systems must comply with the household equipments. This paper discusses the overall design of Automation System Using Speech Recognition for Speaker Dependency through MFCC which is to be implemented. This system will have hardware for appliance interface using Zigbee wireless modules and speech processor. Software section includes development of Mel Frequency cepstral coefficient (MFCC) algorithm using MATLAB. The system focus on recognition of speech commands stored in data base of MATLAB and it is matched with incoming speech command of speaker. Using MFCC system will be speaker dependent, for speaker independent systems speech processor of will be used. It uses low-power RF ZigBee transceiver wireless communication modules which are relatively cheap. Such Automation system is developed to control lights and electrical appliances in a home or office with secure use. Wireless transmitter module is interfaced through serial communication with computer. Appliances to be controlled will be connected to zigbee receiver module.