Automated machine-learning speech analysis software was used to analyze speech transcripts obtained from interviews with at-risk youths, to predict which youths would later develop psychosis within two years, with an accuracy of up to 83%.

See the source article by following the link below:
Automated machine-learning speech analysis software was used to analyze speech transcripts obtained from interviews with at-risk youths, to predict which youths would later develop psychosis within two years, with an accuracy of up to 83%.