Oleksandr Pogorilyi, Mohammad Fard, David Traylor, Joan Day
This article investigates whether the robust landmark-based audio fingerprinting method created for recognizing music can be applied to identify squeak and rattle (S&R) types of sounds. The identification is performed by matching a query audio sample to the perceptually closest audio sample that is stored in a pre-developed database of S&R audio sounds. The aim of the application of the method in the automotive industry is to facilitate the process S&R experts go through during sound identification. The experimental results show that the algorithm can be used for identification of different types of S&R sounds when the audio database contains a limited number of reference samples.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados