# An Audio Language Model-Based Voice Concept Bottleneck Framework for Interpretable Health Assessment

arXiv: https://arxiv.org/abs/2607.16967 · HTML: https://mlsanity.com/paper/2607.16967/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Jul 2026.

## Built on by (most cited first, 1 of 1)

- [Uncovering shortcut learning in audio classifiers by discovering recurring concepts in temporal explanations](https://mlsanity.com/paper/2609.34030/) (2026)

## Abstract (arXiv)

Interpretability is critical in clinical decision support. Concept bottleneck frameworks improve it by representing inputs as human-understandable concepts and restricting predictions solely on them. However, research on their use for voice-based health assessment remains limited. In this study, we propose a voice concept bottleneck framework for interpretable health assessment using an audio language model (ALM). The ALM is fine-tuned on a voice quality assessment dataset to enhance its understanding of voice concepts and serves as an independent concept extractor, producing discrete, interpretable scores for a lightweight downstream classifier. The discrete concept scores provide intuitive interpretation, while the lightweight classifier facilitates post-hoc interpretability analyses. Results on depression and dysarthria assessment tasks demonstrate that the proposed framework can flexibly adapt voice concepts to different health conditions and consistently outperforms openSMILE-based and self-supervised speech model-based baselines.