Abstract / Description of output

People’s lived experience of mental illness often includes symptoms associated with several different conditions. The Brief Symptom Inventory (BSI) is a well validated tool for collecting symptom self-reports that covers nine dimensions: depression, anxiety, phobia, paranoia, psychosis, OCD, hostility, somatisation, and interpersonal. In this paper, we investigate to what extent these dimensions are reflected in prosody. Prosody was characterised using five core principal components (PC) derived from GeMAPS analysis of a data set from 8 studies (14907 sound files, 990 participants). We used the data from a subset of 317
participants (5967 sound files, 4 studies) who had completed the BSI. Each BSI dimension shows a distinct pattern of correlations between the number of symptoms reported and our five PCs, but these patterns differ
when comparing undergraduates (more Caucasians, more females) and a patient sample (more African-Americans, more males). We conclude that speech corpora for mental health studies need better demographic balance.
Original languageEnglish
Title of host publicationProceedings of the 19th International Congress of Phonetic Sciences ICPhS 2019
Subtitle of host publication5-9 August 2019, Melbourne, Australia
EditorsSasha Calhoun, Paola Escudero, Marija Tabain, Paul Warren
Place of PublicationCanberra, Australia: Australasian Speech Science and Technology Association Inc.
PublisherAustralian Speech Science & Technology Association Inc
Number of pages5
ISBN (Print)978-0-646-800069-1
Publication statusPublished - 31 Aug 2019
Event19th International Congress of Phonetic Sciences - Melbourne, Australia
Duration: 5 Aug 20199 Aug 2019


Conference19th International Congress of Phonetic Sciences
Abbreviated titleICPhS 2019
Internet address


Dive into the research topics of 'Distinct Prosodic Correlates for Nine Dimensions of Mental Health Symptoms'. Together they form a unique fingerprint.

Cite this