Spoken Grammar Assessment Using LLM

TL;DR

This paper introduces an innovative spoken language assessment system that evaluates grammar directly from spoken utterances using large language models, reducing reliance on written assessments and enhancing test variability.

Contribution

It presents a novel end-to-end spoken grammar assessment system leveraging LLMs and a hybrid ASR, outperforming existing methods and making grammar evaluation more flexible.

Findings

Hybrid ASR with custom language model outperforms state-of-the-art ASR.

The system reduces predictability in grammar testing.

LLMs enable varied and unteachable assessments.

Abstract

Spoken language assessment (SLA) systems restrict themselves to evaluating the pronunciation and oral fluency of a speaker by analysing the read and spontaneous spoken utterances respectively. The assessment of language grammar or vocabulary is relegated to written language assessment (WLA) systems. Most WLA systems present a set of sentences from a curated finite-size database of sentences thereby making it possible to anticipate the test questions and train oneself. In this paper, we propose a novel end-to-end SLA system to assess language grammar from spoken utterances thus making WLA systems redundant; additionally, we make the assessment largely unteachable by employing a large language model (LLM) to bring in variations in the test. We further demonstrate that a hybrid automatic speech recognition (ASR) with a custom-built language model outperforms the state-of-the-art ASR engine for spoken grammar assessment.