Automatic Assessment of Oral Reading Accuracy for Reading Diagnostics

TL;DR

This paper evaluates six advanced ASR systems for automatically assessing Dutch oral reading accuracy, demonstrating promising agreement with human evaluations and highlighting the importance of language model design for reading diagnostics.

Contribution

It compares multiple ASR systems for Dutch reading assessment and shows how language model design impacts accuracy and reliability.

Findings

The best system achieved MCC = .63 with human evaluations.

Language model design including reading errors improves assessment.

Forced decoding confidence correlates with word correctness (r = .45).

Abstract

Automatic assessment of reading fluency using automatic speech recognition (ASR) holds great potential for early detection of reading difficulties and subsequent timely intervention. Precise assessment tools are required, especially for languages other than English. In this study, we evaluate six state-of-the-art ASR-based systems for automatically assessing Dutch oral reading accuracy using Kaldi and Whisper. Results show our most successful system reached substantial agreement with human evaluations (MCC = .63). The same system reached the highest correlation between forced decoding confidence scores and word correctness (r = .45). This system's language model (LM) consisted of manual orthographic transcriptions and reading prompts of the test data, which shows that including reading errors in the LM improves assessment performance. We discuss the implications for developing automatic assessment systems and identify possible avenues of future research.