A Survey: Pre-processing and Feature Extraction Techniques for Depression Analysis Using Speech Signal

Clinical depression has become a major factor causing suicides and is associated with high mortality rates making it the leading causes of death worldwide. Thus, early detection of depression is of primary importance as there has been a dramatic increase in the depressive symptoms and disorders in recent years. From a psychological point of view, one of the signs of a person being depressed is the way emotions are expressed in his/her speech. Depressive disorder affects the acoustic qualities of their speech; hence depression can be detected through an analysis of acoustical properties of speech. Features are extracted from the preprocessed speech signal and accordingly using non-linear classifiers speech is classified as depressed or controlled. Keywords:- Acoustic features, Clinical depression, linear technique, nonlinear technique. I