Filtering by topic speech-technology(1)Clear all filters

• Paper·arXiv — Language & NLP (cs.CL)·4 Jun 2026

  Domain-Aware Mispronunciation Detection and Diagnosis Using Language-Specific Statistical Graphs

  Huu Tuong Tu, Hanh Nguyen, Thien Van Luong, Nguyen Tien Cuong et al.

  Proposes a method using language-specific statistical graphs to represent phoneme confusion patterns for mispronunciation detection and diagnosis (MDD). The approach leverages directed graphs to capture systematic pronunciation differences across native language (L1) backgrounds. On the L2-ARCTIC benchmark, it achieves an F1-score of 59.52%, outperforming several competitive baselines.

  Original abstract

  Mispronunciation Detection and Diagnosis (MDD) has gained increasing importance in computer-assisted language learning and speech technology in recent years. In this paper, we propose a method for constructing statistical graphs that enable models to learn phoneme confusion patterns represented as directed graphs. Furthermore, we introduce a language-specific strategy to capture systematic pronunciation differences across various native language (L1) backgrounds. The effectiveness of our approach is demonstrated through extensive experiments on the L2-ARCTIC benchmark, where it achieves an F1-score of 59.52%, outperforming several competitive baselines.

  mispronunciation-detectioncomputer-assisted-language-learningspeech-technologyphoneme-confusion★ 8/10