Title :
Classification of Turkish musical instruments
Author :
M. Erdal Ozbek;F. Acar Savaci
Author_Institution :
Elektrik-Elektronik M?hendisli?i B?l?m?, ?zmir Y?ksek Teknoloji Enstit?s?, Turkey
fDate :
4/1/2009 12:00:00 AM
Abstract :
In this work, Turkish musical instruments are classified with the features of mel frequency cepstral coefficients used for recognition of Western musical instruments. The performance, sensitivity and specificity ratios are calculated by finding the confusion matrices for different kernel parameters following multi-class classification performed by support vector machines. High performance reaching 97% for Ud samples and 90% in average for all samples is obtained.
Keywords :
"Instruments","Model driven engineering","Mel frequency cepstral coefficient","Sensitivity and specificity","Kernel","Support vector machines","Support vector machine classification"
Conference_Titel :
Signal Processing and Communications Applications Conference, 2009. SIU 2009. IEEE 17th
Print_ISBN :
978-1-4244-4435-9
DOI :
10.1109/SIU.2009.5136358
