Saved in:
Bibliographic Details
Main Authors: Yu, Hangbin, Yang, Yudong, Su, Rongfeng, Yan, Nan, Wang, Lan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.10181
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913022504599552
author Yu, Hangbin
Yang, Yudong
Su, Rongfeng
Yan, Nan
Wang, Lan
author_facet Yu, Hangbin
Yang, Yudong
Su, Rongfeng
Yan, Nan
Wang, Lan
contents Automatic depression detection using speech signals with acoustic and textual modalities is a promising approach for early diagnosis. Depression-related patterns exhibit sparsity in speech: diagnostically relevant features occur in specific segments rather than being uniformly distributed. However, most existing methods treat all frames equally, assuming depression-related information is uniformly distributed and thus overlooking this sparsity. To address this issue, we proposes a depression detection network based on Adaptive Cross-Modal Gating (ACMG) that adaptively reassigns frame-level weights across both modalities, enabling selective attention to depression-related segments. Experimental results show that the depression detection system with ACMG outperforms baselines without it. Visualization analyses further confirm that ACMG automatically attends to clinically meaningful patterns, including low-energy acoustic segments and textual segments containing negative sentiments.
format Preprint
id arxiv_https___arxiv_org_abs_2604_10181
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning to Attend to Depression-Related Patterns: An Adaptive Cross-Modal Gating Network for Depression Detection
Yu, Hangbin
Yang, Yudong
Su, Rongfeng
Yan, Nan
Wang, Lan
Sound
Automatic depression detection using speech signals with acoustic and textual modalities is a promising approach for early diagnosis. Depression-related patterns exhibit sparsity in speech: diagnostically relevant features occur in specific segments rather than being uniformly distributed. However, most existing methods treat all frames equally, assuming depression-related information is uniformly distributed and thus overlooking this sparsity. To address this issue, we proposes a depression detection network based on Adaptive Cross-Modal Gating (ACMG) that adaptively reassigns frame-level weights across both modalities, enabling selective attention to depression-related segments. Experimental results show that the depression detection system with ACMG outperforms baselines without it. Visualization analyses further confirm that ACMG automatically attends to clinically meaningful patterns, including low-energy acoustic segments and textual segments containing negative sentiments.
title Learning to Attend to Depression-Related Patterns: An Adaptive Cross-Modal Gating Network for Depression Detection
topic Sound
url https://arxiv.org/abs/2604.10181