L1-Aware Phoneme-Level Pronunciation Assessment of Uzbek L1– English L2 Speech Using wav2vec 2.0 Acoustic-Phonetic Representations
Article Sidebar
Main Article Content
Abstract
Automatic pronunciation assessment is especially important for Uzbek learners of English because differences between the Uzbek and English phonological systems can produce systematic L1-induced deviations in segmental and prosodic realization. This paper proposes an interpretable pronunciationassessment model in which wav2vec 2.0 is used as a self-supervised acousticphonetic representation extractor, while the assessment stage is formulated as a phoneme-level scoring problem rather than a conventional speech-recognition problem. For each target phoneme, the model combines four complementary indicators: posterior evidence for the canonical phoneme, distance between learner and reference embeddings, duration/prosodic conformity, and an Uzbek-L1-specific confusion penalty. These components are fused by a bounded logistic scoring function and then aggregated hierarchically from phoneme to word and utterance levels. The mathematical formulation is designed to preserve interpretability: every final score can be decomposed into acoustic, temporal, prosodic, and L1interference contributions. The proposed formulation is suitable for low-resource Uzbek L1–English L2 corpora because most representation learning is transferred from wav2vec 2.0, while only the calibration and scoring layers require taskspecific labeled data. The model provides a technical basis for future computerassisted pronunciation training systems that can produce both a numerical score and error-oriented feedback.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
How to Cite
References
A. Baevski, Y. Zhou, A. Mohamed, and M. Auli, “wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations,” Advances in Neural Information Processing Systems, vol. 33, pp. 12449–12460, 2020.
S. M. Witt and S. J. Young, “Phone-level pronunciation scoring and assessment for interactive language learning,” Speech Communication, vol. 30, no. 2–3, pp. 95–108, 2000. doi: 10.1016/S0167-6393(99)00044-8.
S. Bannò and M. Matassoni, “Proficiency Assessment of L2 Spoken English Using Wav2Vec 2.0,” in 2022 IEEE Spoken Language Technology Workshop (SLT), 2023, pp. 1088–1095. doi: 10.1109/SLT54892.2023.10023019.
R. C. M. C. Shekar, M. Yang, K. Hirschi, S. Looney, O. Kang, and J. H. L. Hansen, “Assessment of Non-Native Speech Intelligibility using Wav2vec2-based Mispronunciation Detection and Multi-level Goodness of Pronunciation Transformer,” in Proc. Interspeech 2023, pp. 984–988, 2023. doi: 10.21437/Interspeech.2023-2371.
S. Gupta, A. Unnam, K. Yadav, and V. Aggarwal, “Towards Building a Language-Independent Speech Scoring Assessment,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 21, pp. 23200–23206, 2024. doi: 10.1609/aaai.v38i21.30366.
D. Cai, B. Naismith, M. Kostromitina, Z. Teng, K. P. Yancey, and G. T. LaFlair, “Developing an Automatic Pronunciation Scorer: Aligning Speech Evaluation Models and Applied Linguistics Constructs,” Language Learning, vol. 75, Suppl. 1, pp. 170–203, 2025. doi: 10.1111/lang.70000.
M. Shahin, J. Epps, and B. Ahmed, “Phonological level wav2vec2-based Mispronunciation Detection and Diagnosis method,” Speech Communication, vol. 173, Art. no. 103249, 2025. doi: 10.1016/j.specom.2025.103249.
D. A. Yuldasheva, “Phonetic interference in Uzbek learners’ production of English sounds,” Advances in Science and Humanities, vol. 2, no. 4, pp. 63–66, 2026. doi: 10.70728/human.v02.i04.016.
N.Raximov, J.Kuvandikov, D.Khasanov, “The importance of loss function in artificial intelligence”, International Conference on Information Science and Communications Technologies (ICISCT 20222), DOI:
1109/ICISCT55600.2022.10146883 [10] Rahimov Nodir, Khasanov Dilmurod. (2022). The Mathematical Essence Of Logistic Regression For Machine Learning. https://doi.org/10.5281/zenodo.7239169
Khasanov Dilmurod, Tojiyev Ma’ruf, Primqulov Oybek., “Gradient Descent In Machine”. International Conference on Information Science and Communications Technologies (ICISCT), https://ieeexplore.ieee.org/document/9670169
Kuvandikov, J., o‘g‘li, R. M. E. (2026). An explainable neuro-symbolic speech matrix model for personalized foreign language pronunciation feedback. Raqamli Transformatsiya va Sun’iy Intellekt, 4(4), 153-158. https://dtai.uz/articles/an-explainable-neuro-symbolic-speech-matrix-model-forpersonalized-foreign-language-pronunciation-feedback/
o‘g‘li, A. A. A., o‘g‘li, R. M. E. (2026). Development of an interactive aibased tool for automatic pronunciation assessment in foreign language learning. Raqamli Transformatsiya va Sun’iy Intellekt, 4(1), 38-45. https://dtai.uz/articles/development-of-an-interactive-ai-based-tool-for-automaticpronunciation-assessment-in-foreign-language-learning /
Rohmonqulov , M. (2025). Development of an interactive software tool for automatic assessment and improvement of pronunciation in foreign language learning. Scientific Practical Conference, 1(1), 344-349. https://dpressa.com/index.php/spc/article/view/258