An SVM-based classification of oral and nasal snoring sounds with Kullback-Leibler kernel

Recently, numerous investigations have shown that loud habitual snoring is due to nasal obstruction and loud oral snoring is found in many sleep apnea patients. So, it is important to detect oral snoring in the earlier stage, but unfortunately we cannot know our own sleep condition. For such purpose, we adopt a Support Vector Machine (SVM) classifier with Kullback-Leibler (KL) kernel so as to classify oral and nasal snoring sounds based on the spectral properties, and compare it with the other kernel functions.