In this paper, the authors show the development of a system able to recognize five separate words in a low noise environment. The created system allows electronic control of lighting, using the computational power of a personal computer to run the algorithms involved in the processing and recognition of the voice signal. For this process, an algorithm called Fast Fourier Transform and artificial neural networks are used. Implementing the electronic control includes using the zigbee protocol to communicate between the user’s voice acquisition device and the computer. The result is a portable voice recognition system for lighting control, which is only limited by the coverage of the transmission-reception modules used. This work is being
developed by the industrial automation research group of the Universidad Cooperativa de Colombia as part of the project “Design and construction of a generic voice activated control module for people with disabilities”, approved in 2009.
Keywords:
voice signal processing, zigbee protocol, voice recognition, Artificial Neural Networks (rna), Fast Fourier Transform (fft)