Abstract: Acoustic complements is an important methodology to perceive the sounds from environment. Significantly machines in different conditions can have the hearings capability like smartphones, different software or security systems. This kind of work can be implemented through conventional or deep learning machine models that contain revolutionized speech identification to understand general environment sounds. This work focuses on the acoustic classification and improves the performance of deep neural networks by using hybrid feature extraction methods. This study improves the efficiency of classification to extract features and make prediction of cost graph. We have adopted the hybrid feature extraction scheme consisting of DNN and CNN. The results have 12% improvement from the previous results by using mix feature extraction scheme.

Muhammad Ahsan Aslam, Muhammad Umer Sarwar, Muhammad Kashif Hanif, Ramzan Talib and Usama Khalid, “Acoustic Classification using Deep Learning” International Journal of Advanced Computer Science and Applications(IJACSA), 9(8), 2018. http://dx.doi.org/10.14569/IJACSA.2018.090820

@article{Aslam2018,
title = {Acoustic Classification using Deep Learning},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2018.090820},
url = {http://dx.doi.org/10.14569/IJACSA.2018.090820},
year = {2018},
publisher = {The Science and Information Organization},
volume = {9},
number = {8},
author = {Muhammad Ahsan Aslam and Muhammad Umer Sarwar and Muhammad Kashif Hanif and Ramzan Talib and Usama Khalid}
}