BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Context Aware Deep Learning for Multi Modal Depression Detection -
  Genevieve Lam (Independent Researcher)
DTSTART:20250718T160000Z
DTEND:20250718T163000Z
UID:TALK234433@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:In this talk\, we focus on automated approaches to detect depr
 ession from clinical interviews using multi-modal machine learning (ML). O
 ur approach differentiates from other successful ML methods such as contex
 t-aware analysis through feature engineering and end-to-end deep neural ne
 tworks for depression detection utilizing the Distress Analysis Interview 
 Corpus. We propose a novel method that incorporates: (1) pre-trained Trans
 former combined with data augmentation based on topic modeling for textual
  data\; and (2) deep 1D convolutional neural network (CNN) for acoustic fe
 ature modeling. The simulation results demonstrate the effectiveness of th
 e proposed method for training multi-modal deep learning models. Our deep 
 1D CNN and Transformer models achieved state-of-the-art performance for au
 dio and text modalities respectively. Combining them in a multi-modal fram
 ework also outperforms state-of-the-art for the combined setting.\n\nThis 
 work was previously presented as an Oral at International Conference on Ac
 oustics\, Speech and Signal Processing 2019\, United Kingdom
LOCATION:Computer Laboratory\, William Gates Building\, Room FW26.
END:VEVENT
END:VCALENDAR