Speech Intelligibility Assessment of Dysarthric Speech by using Goodness of Pronunciation with Uncertainty Quantification
This work addresses automatic speech intelligibility assessment for dysarthric speech, which is an incremental improvement over existing methods.
The paper tackled the problem of assessing speech intelligibility in dysarthric speech by improving Goodness of Pronunciation with uncertainty quantification, resulting in relative performance increases of 5.66% for English, 3.91% for Korean, and 23.65% for Tamil compared to baseline methods.
This paper proposes an improved Goodness of Pronunciation (GoP) that utilizes Uncertainty Quantification (UQ) for automatic speech intelligibility assessment for dysarthric speech. Current GoP methods rely heavily on neural network-driven overconfident predictions, which is unsuitable for assessing dysarthric speech due to its significant acoustic differences from healthy speech. To alleviate the problem, UQ techniques were used on GoP by 1) normalizing the phoneme prediction (entropy, margin, maxlogit, logit-margin) and 2) modifying the scoring function (scaling, prior normalization). As a result, prior-normalized maxlogit GoP achieves the best performance, with a relative increase of 5.66%, 3.91%, and 23.65% compared to the baseline GoP for English, Korean, and Tamil, respectively. Furthermore, phoneme analysis is conducted to identify which phoneme scores significantly correlate with intelligibility scores in each language.