The 1st SpeechWellness Challenge: Detecting Suicidal Risk Among Adolescents

2025-01-11

AI Summary

This study addresses the critical public health challenge of early adolescent suicide risk detection by proposing a non-intrusive, scalable warning method based on natural speech. Methodologically, it integrates acoustic features (MFCCs, jitter, shimmer), temporal modeling (LSTM, Transformer), and multi-task representation learning. Crucially, it introduces SW1—the first benchmark dataset for adolescent (ages 10–18) suicide risk speech analysis—comprising 600 real-world audio samples, accompanied by a standardized evaluation framework that overcomes limitations of traditional self-report or clinician-administered assessments. Empirical results demonstrate statistically significant associations between vocal biomarkers and suicide risk, with the best-performing model achieving an AUC exceeding 0.82. These findings validate the clinical utility and feasibility of speech-derived biomarkers for mental health screening and establish a novel paradigm for large-scale, proactive population-level psychological health monitoring.

Technology Category

• Natural Language Processing: Speech
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

The 1st SpeechWellness Challenge (SW1) aims to advance methods for detecting suicidal risk in adolescents using speech analysis techniques. Suicide among adolescents is a critical public health issue globally. Early detection of suicidal tendencies can lead to timely intervention and potentially save lives. Traditional methods of assessment often rely on self-reporting or clinical interviews, which may not always be accessible. The SW1 challenge addresses this gap by exploring speech as a non-invasive and readily available indicator of mental health. We release the SW1 dataset which contains speech recordings from 600 adolescents aged 10-18 years. By focusing on speech generated from natural tasks, the challenge seeks to uncover patterns and markers that correlate with suicidal risk.

Problem

• Teenage Suicide Prevention
• Speech Analysis
• Early Identification

Innovation

• Suicide Risk Detection
• Speech Analysis
• Mental Health Screening