Automatic Pronunciation Assessment -- A Review

TL;DR

This paper reviews recent advances in automatic pronunciation assessment, covering phonemic and prosodic evaluation, highlighting challenges, limitations, resources, and future research directions in the field.

Contribution

It provides a comprehensive update on methods, challenges, and resources in pronunciation assessment, integrating recent deep learning approaches and categorizing research trends.

Findings

Identification of key challenges in pronunciation assessment

Analysis of existing limitations and resources

Discussion of future research directions

Abstract

Pronunciation assessment and its application in computer-aided pronunciation training (CAPT) have seen impressive progress in recent years. With the rapid growth in language processing and deep learning over the past few years, there is a need for an updated review. In this paper, we review methods employed in pronunciation assessment for both phonemic and prosodic. We categorize the main challenges observed in prominent research trends, and highlight existing limitations, and available resources. This is followed by a discussion of the remaining challenges and possible directions for future work.