Evaluating Logit-Based GOP Scores for Mispronunciation Detection
Not on the map. Why?
Abstract (arXiv)
Pronunciation assessment relies on goodness of pronunciation (GOP) scores, traditionally derived from softmax-based posterior probabilities. However, posterior probabilities may suffer from overconfidence and poor phoneme separation, limiting their effectiveness. This study compares logit-based GOP scores with probability-based GOP scores for mispronunciation detection. We conducted our experiment on two L2 English speech datasets spoken by Dutch and Mandarin speakers, assessing classification performance and correlation with human ratings. Logit-based methods outperform probability-based GOP in classification, but their effectiveness depends on dataset characteristics. The maximum logit GOP shows the strongest alignment with human perception, while a combination of different GOP scores balances probability and logit features. The findings suggest that hybrid GOP methods incorporating uncertainty modeling and phoneme-specific weighting improve pronunciation assessment.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 14 of its 33 references that are in the map. First posted on arXiv 2 Jun 2025.
Builds on
14 of 33 references.
References · 14
- 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,342
- 2021XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scalecited by 601
- 2017Regularizing Neural Networks by Penalizing Confident Output Distributionscited by 464
- 2022Mitigating Neural Network Overconfidence with Logit Normalizationcited by 252
- 2021Simple and Effective Zero-shot Cross-lingual Phoneme Recognitioncited by 79
- 2020CTC-Segmentation of Large Corpora for German End-to-end Speech Recognitioncited by 76
- 2021speechocean762: An Open-Source Non-native English Speech Corpus For Pronunciation Assessmentcited by 58
- 2022Transformer-Based Multi-Aspect Multi-Granularity Non-Native English Speaker Pronunciation Assessmentcited by 37
- 2020Context-aware Goodness of Pronunciation for Computer-Assisted Pronunciation Trainingcited by 26
- 2023Logit Margin Matters: Improving Transferable Targeted Adversarial Attack by Logit Calibrationcited by 18
- 2023A Hierarchical Context-aware Modeling Approach for Multi-aspect and Multi-granular Pronunciation Assessmentcited by 13
- 2023Speech Intelligibility Assessment of Dysarthric Speech by using Goodness of Pronunciation with Uncertainty Quantificationcited by 10
Built on by
Citing papers per year: 2026: 1.
1 paper.