Third International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2003)

Florence, Italy
December 10-12, 2003

Investigation of Glottal Flow Spectral Slope as Possible Cue for Depression and Near-Term Suicidal Risk

Asli Ozdas (1), Hasmila Omar (3), Richard G. Shiavi (2), Stephen E. Silverman (4), Marilyn K. Silverman (4), D. Mitchell Wilkes (3)

(1) Department of Biomedical Informatics; (2) Department of Biomedical Engineering; (3) Department of Electrical and Computer Engineering; Vanderbilt University, Nashville, TN, USA
(4) Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA

When reviewing his clinical experience in treating suicidal patients, one of the authors observed that successful predictions of suicidality were often based on the patient's voice independent of content. In this study we investigated the discriminating power of an excitation-based speech parameter, the glottal flow spectrum. There were two sets of subjects, male and female. Each set consisted of 10 high-risk near-term suicidal patients, 10 major depressed patients, and 10 non-depressed control subjects. As a result of two sample statistical analyses, the slope of the glottal flow spectrum, was a significant discriminator in five of six comparisons (p<0.05). A maximum likelihood classifier, developed by combining the a posteriori probabilities of two features, yielded correct classification scores between 60 and 95%.

Index Terms. Speech, glottal flow spectrum, suicide, depression, classification

Full Paper (reprinted with permission from Firenze University Press)

Bibliographic reference.  Ozdas, Asli / Omar, Hasmila / Shiavi, Richard G. / Silverman, Stephen E. / Silverman, Marilyn K. / Wilkes, D. Mitchell (2003): "Investigation of glottal flow spectral slope as possible cue for depression and near-term suicidal risk", In MAVEBA-2003, 13-16.