Deep Learning-Based Feature Fusion for Emotion Analysis and Suicide Risk Differentiation in Chinese Psychological Support Hotlines

2025-01-15

AI Summary

This study addresses suicide risk prediction in Chinese psychological helpline scenarios by modeling dynamic vocal emotion patterns. Recognizing that existing methods overlook temporal emotional evolution, we propose novel clinical assessment metrics—emotional fluctuation intensity and frequency—and design an acoustic–deep feature collaborative fusion framework. Specifically, we integrate pitch-based acoustic features with a hybrid CNN-LSTM model, incorporating multi-scale feature fusion and joint binary/multi-class optimization. Evaluated on real-world helpline recordings, our approach achieves a 79.13% F1-score for negative-emotion binary classification and outperforms current state-of-the-art methods in multi-class emotion recognition. Critically, we provide the first empirical evidence of statistically significant associations between dynamic emotional features and suicidal behavior. This work establishes an interpretable, deployable speech analytics paradigm for real-time crisis intervention.

Technology Category

• Natural Language Processing: Speech
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Humans and AI: Emotional Intelligence

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Mental health is a critical global public health issue, and psychological support hotlines play a pivotal role in providing mental health assistance and identifying suicide risks at an early stage. However, the emotional expressions conveyed during these calls remain underexplored in current research. This study introduces a method that combines pitch acoustic features with deep learning-based features to analyze and understand emotions expressed during hotline interactions. Using data from China's largest psychological support hotline, our method achieved an F1-score of 79.13% for negative binary emotion classification.Additionally, the proposed approach was validated on an open dataset for multi-class emotion classification,where it demonstrated better performance compared to the state-of-the-art methods. To explore its clinical relevance, we applied the model to analysis the frequency of negative emotions and the rate of emotional change in the conversation, comparing 46 subjects with suicidal behavior to those without. While the suicidal group exhibited more frequent emotional changes than the non-suicidal group, the difference was not statistically significant.Importantly, our findings suggest that emotional fluctuation intensity and frequency could serve as novel features for psychological assessment scales and suicide risk prediction.The proposed method provides valuable insights into emotional dynamics and has the potential to advance early intervention and improve suicide prevention strategies through integration with clinical tools and assessments The source code is publicly available at https://github.com/Sco-field/Speechemotionrecognition/tree/main.

Problem

• Emotion Analysis
• Suicide Risk Prediction
• Mental Health Assessment

Innovation

• Deep Learning
• Voice Pitch Analysis
• Suicide Risk Prediction