Beyond Acoustic Sparsity and Linguistic Bias: A Prompt-Free Paradigm for Mispronunciation Detection and Diagnosis
This work addresses the problem of improving mispronunciation detection for second-language learners, offering a robust method that avoids biases from explicit canonical priors.
The paper tackles mispronunciation detection and diagnosis (MDD) by proposing a prompt-free framework that decouples acoustic fidelity from canonical guidance. The proposed CROTTC-IF model achieves 71.77% F1-score on L2-ARCTIC and 71.70% F1-score on the Iqra'Eval2 leaderboard.
Mispronunciation Detection and Diagnosis (MDD) requires modeling fine-grained acoustic deviations. However, current ASR-derived MDD systems often face inherent limitations. In particular, CTC-based models favor sequence-level alignments that neglect transient mispronunciation cues, while explicit canonical priors bias predictions toward intended targets. To address these bottlenecks, we propose a prompt-free framework decoupling acoustic fidelity from canonical guidance. First, we introduce CROTTC, an acoustic model enforcing monotonic, frame-level alignment to accurately capture pronunciation deviations. Second, we implicitly inject mispronunciation information via the IF strategy under the knowledge transfer principle. Experiments show CROTTC-IF achieves a 71.77% F1-score on L2-ARCTIC and 71.70% F1-score on the Iqra'Eval2 leaderboard. With empirical analysis, we demonstrate that decoupling acoustics from explicit priors provides highly robust MDD.