Disentangled Latent Speech Representation for Automatic Pathological Intelligibility Assessment

TL;DR

This paper introduces a novel automatic method for assessing speech intelligibility in patients with speech disorders by analyzing disentangled latent speech representations, showing high correlation with subjective measures and robustness with limited data.

Contribution

The study presents a new approach using disentangled latent speech representations for objective pathological speech intelligibility assessment, invariant to reference speaker variations.

Findings

High correlation (R = -0.9) with subjective measures

Method remains robust with fewer utterances (R = -0.89)

Minimal deviation across different reference speakers (+-0.01 to +-0.02)

Abstract

Speech intelligibility assessment plays an important role in the therapy of patients suffering from pathological speech disorders. Automatic and objective measures are desirable to assist therapists in their traditionally subjective and labor-intensive assessments. In this work, we investigate a novel approach for obtaining such a measure using the divergence in disentangled latent speech representations of a parallel utterance pair, obtained from a healthy reference and a pathological speaker. Experiments on an English database of Cerebral Palsy patients, using all available utterances per speaker, show high and significant correlation values (R = -0.9) with subjective intelligibility measures, while having only minimal deviation (+-0.01) across four different reference speaker pairs. We also demonstrate the robustness of the proposed method (R = -0.89 deviating +-0.02 over 1000 iterations) by considering a significantly smaller amount of utterances per speaker. Our results are among the first to show that disentangled speech representations can be used for automatic pathological speech intelligibility assessment, resulting in a reference speaker pair invariant method, applicable in scenarios with only few utterances available.