Year: 2000 Source: IEEE Transactions on Biomedical Engineering, v.47, no.7, (July 2000), p.829-837 SIEC No: 20010800

Studies were performed to analyze & compare the speech acoustics of separate male & female samples comprised of normal individuals & individuals with diagnoses of depression & high-risk, near-term suicidality. Acoustic analysis of voice fundamental frequency (Fo), amplitude modulation (AM), formants, & power distribution were performed on speech samples from audio recordings. Multivariate feature & discriminant analyses were performed. Features derived from the formant & power spectral density measurements were found to be the best discriminators of class membership in both the male & female studies. Results support theories which identify psychomotor disturbances as central elements in depression & suicidality. (46 refs.)