SpeechT-RAG: Reliable Depression Detection in LLMs with Retrieval-Augmented Generation Using Speech Timing Information

TL;DR

SpeechT-RAG introduces a novel method that leverages speech timing features with retrieval-augmented generation to improve depression detection accuracy and confidence estimation without additional training.

Contribution

The paper presents SpeechT-RAG, a system that uses speech timing information for depression detection, addressing limitations of text-only approaches and enhancing trustworthiness in mental health assessment.

Findings

Outperforms traditional text-based RAG in depression detection accuracy

Provides reliable confidence scores based on speech timing features

Achieves results comparable to fine-tuned LLMs without extra training

Abstract

Large Language Models (LLMs) have been increasingly adopted for health-related tasks, yet their performance in depression detection remains limited when relying solely on text input. While Retrieval-Augmented Generation (RAG) typically enhances LLM capabilities, our experiments indicate that traditional text-based RAG systems struggle to significantly improve depression detection accuracy. This challenge stems partly from the rich depression-relevant information encoded in acoustic speech patterns information that current text-only approaches fail to capture effectively. To address this limitation, we conduct a systematic analysis of temporal speech patterns, comparing healthy individuals with those experiencing depression. Based on our findings, we introduce Speech Timing-based Retrieval-Augmented Generation, SpeechT-RAG, a novel system that leverages speech timing features for both accurate depression detection and reliable confidence estimation. This integrated approach not only outperforms traditional text-based RAG systems in detection accuracy but also enhances uncertainty quantification through a confidence scoring mechanism that naturally extends from the same temporal features. Our unified framework achieves comparable results to fine-tuned LLMs without additional training while simultaneously addressing the fundamental requirements for both accuracy and trustworthiness in mental health assessment.