Computer-aided analysis of ketamine-influenced speech
MetadataShow full item record
This is a computer-aided analysis of the speech of healthy volunteers under the influence of ketamine as contrasted with their normal speech. The project was undertaken since ketamine has been shown to simulate some of the symptoms of schizophrenia and ketamine-influenced speech shares much similarity with schizophrenic speech. Thus, the study of ketamine-influenced speech provides information relevant to the diagnosis and treatment of schizophrenia. I contrastively studied 20 pairs of normal and ketamine-influenced speech samples from 10 healthy volunteers at various linguistic levels: phonetic, syntactic, semantic and pragmatic. I conclude that ketamine-influenced speech exhibits several distinctive properties: a smaller percentage of verbs, a higher percentage of nouns, an increase in the use of fillers, and greater variability of vowel height. I implemented software that recognizes the distinctive properties of ketamine-influenced speech. The software was tested on 37 pairs of testing data and achieved high accuracy in identifying ketamine-influenced speech.