Automated detection of pronunciation errors in non-native English speech employing deep learning

TL;DR

This paper introduces advanced deep learning techniques for detecting pronunciation errors in non-native English speech, significantly improving accuracy over existing methods by generating synthetic mispronounced data and employing end-to-end detection models.

Contribution

It presents novel deep learning approaches, including synthetic data generation and multi-task end-to-end models, to enhance pronunciation error detection in non-native speech.

Findings

Outperformed state-of-the-art in AUC metric by 41%

Effective detection and reconstruction of dysarthric speech

Applied in Amazon for automatic pronunciation error detection

Abstract

Despite significant advances in recent years, the existing Computer-Assisted Pronunciation Training (CAPT) methods detect pronunciation errors with a relatively low accuracy (precision of 60% at 40%-80% recall). This Ph.D. work proposes novel deep learning methods for detecting pronunciation errors in non-native (L2) English speech, outperforming the state-of-the-art method in AUC metric (Area under the Curve) by 41%, i.e., from 0.528 to 0.749. One of the problems with existing CAPT methods is the low availability of annotated mispronounced speech needed for reliable training of pronunciation error detection models. Therefore, the detection of pronunciation errors is reformulated to the task of generating synthetic mispronounced speech. Intuitively, if we could mimic mispronounced speech and produce any amount of training data, detecting pronunciation errors would be more effective. Furthermore, to eliminate the need to align canonical and recognized phonemes, a novel end-to-end multi-task technique to directly detect pronunciation errors was proposed. The pronunciation error detection models have been used at Amazon to automatically detect pronunciation errors in synthetic speech to accelerate the research into new speech synthesis methods. It was demonstrated that the proposed deep learning methods are applicable in the tasks of detecting and reconstructing dysarthric speech.