Stress detection from speech is a less explored eld than Automatic
Emotion Recognition and it is still not clear which features are
better stress discriminants. VOCE aims at doing speech classication
as stressed or not-stressed in real-time, using acoustic-prosodic features
only. We therefore look for the best discriminating feature subsets from
a set of 6285 features { 6125 features extracted with openSMILE toolkit
and 160 Teager Energy Operator (TEO) features. We use a mutual information
lter and a branch and bound wrapper heuristic with an SVM
classier to perform feature selection. Since many feature sets are selected,
we analyse them in terms of chosen features and classier performance
concerning also true positive and false positive rates. The results
show that the best features types for our application case are Audio
Spectral, MFCC, PCM and TEO. We reached results as high as 70.36%
for generalisation accuracy.