Table 6.

Performance comparison of SVM using different feature sets for multiclass experiments. Bold font indicates the best result obtained against each feature set.

Feature Set	Classifier	Accuracy(%)	Sensitivity(%)	Specificity(%)	PPV(%)	NPV(%)	Error(%)
MFCC	SVM-L	83.93	92.86	85.71	68.42	97.3	16.07
	SVM-Q	86.9	90.48	90.48	76	96.61	13.1
	SVM-C	88.69	90.48	94.44	84.44	96.75	11.31
	SVM-G	83.33	97.62	81.75	64.06	99.04	16.67
1D-LTP	SVM-L	94.64	97.62	93.65	83.67	99.16	5.36
	SVM-Q	94.05	90.48	95.24	86.36	96.77	5.95
	SVM-C	94.64	90.48	96.03	88.37	96.8	5.36
	SVM-G	93.45	92.86	93.65	82.98	97.52	6.55
MFCC+1D-LTP	SVM-L	93.45	97.62	92.06	80.39	99.15	6.55
	SVM-Q	94.43	95.05	94.41	85.06	98.28	5.57
	SVM-C	95.24	95.24	95.24	86.96	98.36	4.76
	SVM-G	93.45	100	91.27	79.25	100	6.55