Examination of the sensitivity of acoustic-phonetic parameters of speech to depression

Abstract

Examination of the sensitivity of acoustic-phonetic parameters of speech to depression was started in the Laboratory of Speech Acoustics. In the presented research work some segmental parameters of the depressed speech were examined, and compared with the parameters of normal speech. It was clear from the experiments that there are significant differences between the parameters of depressed and normal speech. The jitter and shimmer values of the examined vowels were significantly higher, the first and second formant frequencies were lower in depressed speech. Because it is clear that for a good examination, much more depressed patient voices are necessary, so we are going to continue the collection of the depressed speech database. Moreover the extension of the examination to the supra-segmental parameters of speech, like speaking rate, vocalization/pause ratio etc. are planned.