Implementation of Neural Network Backpropagation Using Audio Feature Extraction For Classification Of Gamelan Notes
Fahri Firdausillah, Dwiky Gilang Mahendra, Junta Zeniarja, Ardytha Luthfiarta, Heru Agus Santoso, Adhitya Nugraha, Erwin Yudi Hidayat, Abdul Syukur · 2018 International Seminar on Application for Technology of Information and Communication · 2018
Gamelan consists of several musical instruments including kendang, saron, bonang, panerus, kempul, gender, gambang, kethuk, flute, siter, clempung, slenthem, demung, japan, kempyang, peking, and gong. Many people do not know the name of each gamelan instrument and how it sounds. In an effort to increase the popularity and introduction of gamelan as a traditional musical instrument to the community, this research proposes an analysis based on audio classification of gamelan musical instruments. Classification is done by data mining technique using backpropagation Neural Network (BPNN) method. Audio or sound data from gamelan recordings is conducted by preprocessing using Zero Crossing Rate and Short Time Energy before processed to Backpropagation Neural Network. This research only classifies 4 gamelan musical instruments namely gong, kenong, saron, and gambang. The experimental result shows accuracy at 82.5 %.