Academic Journal

Investigation of Speech Landmark Patterns for Depression Detection

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Investigation of Speech Landmark Patterns for Depression Detection
Συγγραφείς: Huang, Z, Epps, J, Joachim, D
Πηγή: IEEE Transactions on Affective Computing. 13:666-679
Στοιχεία εκδότη: Institute of Electrical and Electronics Engineers (IEEE), 2022.
Έτος έκδοσης: 2022
Θεματικοί όροι: anzsrc-for: 1702 Cognitive Sciences, 0301 basic medicine, 4608 Human-Centred Computing, Depression, anzsrc-for: 46 Information and Computing Sciences, anzsrc-for: 4603 Computer vision and multimedia computation, 02 engineering and technology, Mental Illness, Brain Disorders, anzsrc-for: 0806 Information Systems, anzsrc-for: 4608 Human-Centred Computing, 03 medical and health sciences, Mental Health, 46 Information and Computing Sciences, anzsrc-for: 4602 Artificial intelligence, 0202 electrical engineering, electronic engineering, information engineering, anzsrc-for: 0801 Artificial Intelligence and Image Processing
Περιγραφή: The massive and growing burden imposed on modern society by depression has motivated investigations into early detection through automated, scalable and non-invasive methods, including those based on speech. However, speech-based methods that capture articulatory information effectively across different recording devices and in naturalistic environments are still needed. This article proposes two feature sets associated with speech articulation events based on counts and durations of sequential landmark groups or n-grams. Statistical analysis of the duration-based features reveals that durations from several consecutive landmark bigrams and onset-offset landmark pairs are significant in discriminating depressed from non-depressed speakers. In addition to investigating different normalization approaches and values of n for landmark n-gram features, experiments across different elicitation tasks suggest that the features can be tailored to capture different articulatory aspects of depressed voices. Evaluations of both landmark duration features and landmark n-gram features on the DAIC-WOZ and SH2 datasets show that they are highly effective, either alone or fused, relative to existing approaches.
Τύπος εγγράφου: Article
Περιγραφή αρχείου: application/pdf
ISSN: 2371-9850
DOI: 10.1109/taffc.2019.2944380
Σύνδεσμος πρόσβασης: https://ieeexplore.ieee.org/document/8861018
Rights: IEEE Copyright
CC BY NC ND
Αριθμός Καταχώρησης: edsair.doi.dedup.....8aa659c1a35640a85a8045f3aaf16699
Βάση Δεδομένων: OpenAIRE
Περιγραφή
ISSN:23719850
DOI:10.1109/taffc.2019.2944380