Feature Extraction of Engine Sound by Fast Fourier Transform Technique.
Mutia Delina (a*), Panji Ahmad Nurhusni (a)

a) Physics Department, Faculty of Mathematics and Natural Sciences, Universitas Negeri Jakarta, Jl Rawamangun Muka No 1, Jakarta Timur 13220, Indonesia
*mutia_delina[at]unj.ac.id

Abstract

This research aimed to study the feature of engine sound from various exhausts by the Fast Fourier Transform technique. The engine sound was extracted into three different features; energy, entropy, and zero-crossing rate. The energy and entropy were applied to recognize if the sound came from the exhaust or not. The zero-crossing rate was used to distinguish the exhausts into the ordinary exhaust or the racing exhaust. The result showed that the crossing rate of ordinary exhaust was 0.05 to 0.1 while the racing exhaust was below 0.05.

Keywords: Fast Fourier Transform, Feature Extraction, Engine Sound

Topic: Physics