Assessment of L2 Oral Proficiency using Speech Large Language Models
This addresses the growing demand for automated spoken language assessment tools for L2 English learners, representing an incremental improvement over existing methods.
The paper tackled automatic grading of L2 English oral proficiency by exploring speech large language models (LLMs), showing they outperform previous baselines with superior performance on two datasets and demonstrating strong generalization in cross-part/task evaluations.
The growing population of L2 English speakers has increased the demand for developing automatic graders for spoken language assessment (SLA). Historically, statistical models, text encoders, and self-supervised speech models have been utilised for this task. However, cascaded systems suffer from the loss of information, while E2E graders also have limitations. With the recent advancements of multi-modal large language models (LLMs), we aim to explore their potential as L2 oral proficiency graders and overcome these issues. In this work, we compare various training strategies using regression and classification targets. Our results show that speech LLMs outperform all previous competitive baselines, achieving superior performance on two datasets. Furthermore, the trained grader demonstrates strong generalisation capabilities in the cross-part or cross-task evaluation, facilitated by the audio understanding knowledge acquired during LLM pre-training.