Search papers, labs, and topics across Lattice.
This paper introduces a phonological feature-based framework for automatic mispronunciation detection and diagnosis (MDD) in L2 Mandarin learners, enhancing the traditional end-to-end Wav2Vec2-CTC architecture. By explicitly modeling both segmental and tonal attributes, the method significantly improves diagnostic feedback, achieving a 10.1% reduction in False Acceptance Rate (FAR) and a 23.6% decrease in Diagnostic Error Rate (DER) compared to existing phoneme-only systems. The approach allows for more granular and interpretable feedback, addressing a critical gap in current MDD methodologies for language learning.
Enhanced diagnostic feedback for L2 Mandarin learners reduces mispronunciation errors by over 23%, transforming how pronunciation is taught and assessed.
Automatic mispronunciation detection and diagnosis (MDD) plays a crucial role in L2 Mandarin pronunciation learning. While end-to-end (E2E) based MDD methods have substantially improved phoneme-level detection accuracy, diagnostic feedback remains limited, as segmental and tonal errors are not explicitly separated. In this paper, we propose a phonological feature-based MDD framework that models both segmental and tonal attributes within a unified Wav2Vec2-CTC architecture. Experimental results show that the proposed method reduces the False Acceptance Rate (FAR) by 10.1% and the Diagnostic Error Rate (DER) by 23.6% compared with the phoneme-only baseline system. By decomposing phonemes into low-level phonological components, the proposed approach enables more detailed and interpretable diagnostic feedback for L2 learners.
