Exploring ASR-Based Wav2Vec2 for Automated Speech Disorder Assessment: Insights and Analysis

TL;DR

This paper analyzes the use of Wav2Vec2 ASR-based models for automated speech disorder assessment, focusing on interpretability, key model layers, and comparison of different pre-trained models, setting a new baseline in clinical speech evaluation.

Contribution

It provides the first detailed analysis of Wav2Vec2 models for speech disorder assessment, including layer-wise insights and interpretability methods like CCA and visualization.

Findings

Identification of key layers influencing assessment tasks

Comparison of SSL and ASR Wav2Vec2 models based on pre-training data

Enhanced interpretability through post-hoc XAI methods

Abstract

With the rise of SSL and ASR technologies, the Wav2Vec2 ASR-based model has been fine-tuned for automated speech disorder quality assessment tasks, yielding impressive results and setting a new baseline for Head and Neck Cancer speech contexts. This demonstrates that the ASR dimension from Wav2Vec2 closely aligns with assessment dimensions. Despite its effectiveness, this system remains a black box with no clear interpretation of the connection between the model ASR dimension and clinical assessments. This paper presents the first analysis of this baseline model for speech quality assessment, focusing on intelligibility and severity tasks. We conduct a layer-wise analysis to identify key layers and compare different SSL and ASR Wav2Vec2 models based on pre-trained data. Additionally, post-hoc XAI methods, including Canonical Correlation Analysis (CCA) and visualization techniques, are used to track model evolution and visualize embeddings for enhanced interpretability.