Sound masking using genetic algorithm & artificial neural network (SMUGAANN)

• Main Authors: Culibrina, Francisco B., Dadios, Elmer P.
• Format: text
• Published: Animo Repository 2014
• Subjects: Soundproofing; Computational auditory scene analysis; Genetic algorithms; Sound pressure; Neural networks (Computer science); Manufacturing
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/3916
• Institution: De La Salle University

One of the significant factor of privacy is the source of sound. It creates the level that determines the effectiveness of the other factor. To experience maximum privacy, cancelling of unwanted sound is necessary. To automate the design of sound masking, this proposal use Genetic Algorithm and Artificial Neural Network (SMUGAAN). By evolutionary method two stages are use: a) to determine the target sound being mask evaluation of the parameters of functional elements and b) analysis of the target sound to get the fitness value to be mask and test signals with the help of Sound Synthesis Algorithm (SSA) technique. This stage gives audible sound to mask the target sound. © 2014 IEEE.