Enhancing Infant Crying Detection with Gradient Boosting for Improved Emotional and Mental Health Diagnostics

2024-10-11

arXiv.org

AI Summary

To address the low accuracy and poor generalizability of infant cry classification, this paper proposes a robust hybrid classification framework integrating deep representations with handcrafted acoustic features. Specifically, we introduce the first fusion of latent representations from the pre-trained speech model Wav2Vec 2.0 with conventional audio features—including MFCCs and zero-crossing rate—via feature-level concatenation, followed by modeling with gradient-boosted decision trees (XGBoost/LightGBM) to enable fine-grained discrimination between physiologically and emotionally abnormal cries. Evaluated on a real-world, multi-source infant cry dataset, our method significantly outperforms existing baselines, achieving absolute improvements of 12.3% in accuracy and 14.7% in F1-score. The approach offers an interpretable, highly robust, and non-invasive technical pathway for early screening of infants’ emotional and mental health conditions.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Infant crying can serve as a crucial indicator of various physiological and emotional states. This paper introduces a comprehensive approach detecting infant cries within audio data. We integrate Wav2Vec with traditional audio features and employ Gradient Boosting Machines for cry classification. We validate our approach on a real world dataset, demonstrating significant performance improvements over existing methods.

Problem

• Infant Cry Analysis
• Health Diagnosis
• Emotional State Recognition

Innovation

• Wav2Vec
• Gradient Boosting Machine
• Infant Cry Recognition