Work

Tracking depressed mood using speech pause patterns

Year: 2015

Type: article

Abstract: The speech of people with depression often shows clear signs of their condition (e.g., flat intonation, slow speech, long pauses), but it is not clear to what extent these signs covary with diurnal fluctuations in mood. In this paper, we report results from a pilot longitudinal study where 11 people with depression tracked various aspects of their mental health for a month. This included a daily mood tracker and regular completion of speech tasks. Speech tasks were designed to be emotionally neutral and require different levels of automaticity. We found that participants differed in their willingness to complete the speech tasks, and that preliminary analyses show no clear link between mood and prosody. We discuss implications of this study for tracking depressed mood using speech in real-life applications.

Source: Edinburgh Research Explorer (University of Edinburgh)

Authors: Maria Wolters, Luis Ferrini, Elaine Farrow, Aurora Szentágotai Tătar, Christopher Burton

Institution: Babeș-Bolyai University

Language: English

Topic: Emotion and Mood Recognition

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: green

OpenAlex ID: W2741312069

FWCI: 0.59

Cites: 21

Cited by: 4

Related to: 10