Tracking psychiatric symptoms with daily voice diaries and generative AI – UROP Spring Symposium 2025

Tracking psychiatric symptoms with daily voice diaries and generative AI

Savannah Sneyd

Research Mentor(s): Sekhar Sripada
Mentor Department: Psychiatry/Philosophy
Authors:
Session: Session 6 (3:00pm – 3:50pm)
Presentation Type: Poster 40

Abstract

This research project addresses a critical issue in clinical psychiatric therapy: patients often struggle to accurately recall significant moods and stressors from days or weeks leading up to their therapy session. This memory distortion or forgetting can inhibit clinicians’ ability to understand a patient’s feelings, making it extremely difficult to provide adequate care. As a result, valuable time in therapy sessions is often spent trying to recall past experiences, rather than focusing on solutions to the patient’s problems. To address this issue, the DoPS lab proposes a solution involving an app where patients record daily voice diaries about their moods and stressors of the day. Then, AI transcribes the voice diaries and generates summaries that highlight key information that a clinician would need to know. This method aims to improve the efficiency and accuracy of therapy by providing clinicians with a clear picture of the patient’s emotional state.This project builds on recent work in the realm of AI, but specifically focuses on how AI can improve the clinical therapy experience. By providing therapists with reliable information about patients’ experiences, this project aims to transform how therapy runs, in turn allowing for more effective face-to-face interactions between patient and clinician. Achieving this goal will not only improve effectiveness of therapy but will also change the standard for how AI can benefit healthcare.

lsa logoum logo