Predicting Psychological Well-Being from Spontaneous Speech using LLMs

TL;DR

This study explores how large language models can predict psychological well-being scores from spontaneous speech with high accuracy, using domain-informed prompts and linguistic analysis.

Contribution

It demonstrates the effectiveness of instruction-tuned LLMs in zero-shot psychological assessment from speech and provides insights into linguistic features influencing predictions.

Findings

LLMs achieved up to 0.8 Spearman correlation with PWB scores.

Models can extract meaningful semantic cues from spontaneous speech.

Analysis reveals linguistic features driving model predictions.

Abstract

We investigate the use of Large Language Models (LLMs) for zero-shot prediction of Ryff Psychological Well-Being (PWB) scores from spontaneous speech. Using a few minutes of voice recordings from 111 participants in the PsyVoiD database, we evaluated 12 instruction-tuned LLMs, including Llama-3 (8B, 70B), Ministral, Mistral, Gemma-2-9B, Gemma-3 (1B, 4B, 27B), Phi-4, DeepSeek (Qwen and Llama), and QwQ-Preview. A domain-informed prompt was developed in collaboration with experts in clinical psychology and linguistics. Results show that LLMs can extract semantically meaningful cues from spontaneous speech, achieving Spearman correlations of up to 0.8 on 80\% of the data. Additionally, to enhance explainability, we conducted statistical analyses to characterise prediction variability and systematic biases, alongside keyword-based word cloud analyses to highlight the linguistic features driving the models' predictions.