Automatic Proficiency Assessment in L2 English Learners

TL;DR

This paper investigates deep learning methods, including CNNs, ResNet, wav2vec 2.0, and BERT, for automated assessment of L2 English proficiency through speech, transcription, and dialogue analysis, showing promising results on multiple datasets.

Contribution

It introduces a comprehensive deep learning framework combining speech and text analysis for automated L2 proficiency evaluation, leveraging pretrained models for improved accuracy.

Findings

wav2vec 2.0 outperforms other models in speech proficiency prediction

BERT-based text assessment achieves competitive results

Deep learning models demonstrate potential for robust automated evaluation

Abstract

Second language proficiency (L2) in English is usually perceptually evaluated by English teachers or expert evaluators, with the inherent intra- and inter-rater variability. This paper explores deep learning techniques for comprehensive L2 proficiency assessment, addressing both the speech signal and its correspondent transcription. We analyze spoken proficiency classification prediction using diverse architectures, including 2D CNN, frequency-based CNN, ResNet, and a pretrained wav2vec 2.0 model. Additionally, we examine text-based proficiency assessment by fine-tuning a BERT language model within resource constraints. Finally, we tackle the complex task of spontaneous dialogue assessment, managing long-form audio and speaker interactions through separate applications of wav2vec 2.0 and BERT models. Results from experiments on EFCamDat and ANGLISH datasets and a private dataset highlight the potential of deep learning, especially the pretrained wav2vec 2.0 model, for robust automated L2 proficiency evaluation.