Mohd. Mokji, Musa and N., Senan and R., Ibrahim and N. M., Nawi (2010) The ideal data representation for feature extraction of traditional Malay musical instrument sounds classification. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2010, Changsha, China.
Official URL: http://dx.doi.org/10.1007/978-3-642-14922-1_43
In presenting the appropriate data sets, various data representation and feature extraction methods have been discovered previously. However, almost all the existing methods are utilized based on the Western musical instruments. In this study, the data representation and feature extraction methods are applied towards Traditional Malay musical instruments sounds classification. The impact of five factors that might affecting the classification accuracy which are the audio length, segmented frame size, starting point, data distribution and data fraction (for training and testing) are investigated. The perception-based and MFCC features schemes with total of 37 features was used. While, Multi-Layered Perceptrons classifier is employed to evaluate the modified data sets in terms of the classification performance. The results show that the highest accuracy of 97.37% was obtained from the best data sets with the combination of full features.
|Item Type:||Conference or Workshop Item (Paper)|
|Uncontrolled Keywords:||data representation, feature extraction, traditional malay musical instruments, multi-layered perceptrons|
|Subjects:||T Technology > TK Electrical engineering. Electronics Nuclear engineering|
|Deposited By:||Liza Porijo|
|Deposited On:||29 Aug 2012 06:23|
|Last Modified:||29 Aug 2012 06:25|
Repository Staff Only: item control page