TY - JOUR
T1 - Detecting Depression from Speech through an Attentive LSTM Network
AU - Zhao, Yan
AU - Xie, Yue
AU - Liang, Ruiyu
AU - Zhang, Li
AU - Zhao, Li
AU - Liu, Chengyu
PY - 2021/11/1
Y1 - 2021/11/1
N2 - Depression endangers people's health conditions and affects the social order as a mental disorder. As an efficient diagnosis of depression, automatic depression detection has attracted lots of researcher's interest. This study presents an attention-based Long Short-Term Memory (LSTM) model for depression detection to make full use of the difference between depression and non-depression between timeframes. The proposed model uses frame-level features, which capture the temporal information of depressive speech, to replace traditional statistical features as an input of the LSTM layers. To achieve more multi-dimensional deep feature representations, the LSTM output is then passed on attention layers on both time and feature dimensions. Then, we concat the output of the attention layers and put the fused feature representation into the fully connected layer. At last, the fully connected layer's output is passed on to softmax layer. Experiments conducted on the DAIC-WOZ database demonstrate that the proposed attentive LSTM model achieves an average accuracy rate of 90.2% and outperforms the traditional LSTM network and LSTM with local attention by 0.7% and 2.3%, respectively, which indicates its feasibility.
AB - Depression endangers people's health conditions and affects the social order as a mental disorder. As an efficient diagnosis of depression, automatic depression detection has attracted lots of researcher's interest. This study presents an attention-based Long Short-Term Memory (LSTM) model for depression detection to make full use of the difference between depression and non-depression between timeframes. The proposed model uses frame-level features, which capture the temporal information of depressive speech, to replace traditional statistical features as an input of the LSTM layers. To achieve more multi-dimensional deep feature representations, the LSTM output is then passed on attention layers on both time and feature dimensions. Then, we concat the output of the attention layers and put the fused feature representation into the fully connected layer. At last, the fully connected layer's output is passed on to softmax layer. Experiments conducted on the DAIC-WOZ database demonstrate that the proposed attentive LSTM model achieves an average accuracy rate of 90.2% and outperforms the traditional LSTM network and LSTM with local attention by 0.7% and 2.3%, respectively, which indicates its feasibility.
KW - Artificial Intelligence
KW - Electrical and Electronic Engineering
KW - Computer Vision and Pattern Recognition
KW - Hardware and Architecture
KW - Software
U2 - 10.1587/transinf.2020edl8132
DO - 10.1587/transinf.2020edl8132
M3 - Letter
VL - E104.D
SP - 2019
EP - 2023
JO - IEICE Transactions on Information and Systems
JF - IEICE Transactions on Information and Systems
SN - 0916-8532
IS - 11
ER -