A study on phonemes recognition method for Mandarin pronunciation based on improved Zipformer-RNN-T(Pruned) modeling
PLOS One · 2025
DOI: 10.1371/journal.pone.0324048 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Zhaohui Du und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Information Science and Technology, Shihezi University, Shihezi, China. Hauptthemen laut MEDLINE: Phonetics, Language, Speech Recognition Software. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: A Data-driven Regional Smart Education Service Key Technology Research and Application Demonstration.
Abstract (Originalsprache)
In recent years, empowered by artificial intelligence technologies, computer-assisted language learning systems have gradually become a hot topic of research. Currently, the mainstream pronunciation assessment models rely on advanced speech recognition technology, converting speech into phoneme sequences, and then determining mispronounced phonemes through sequence comparison. To optimize the phoneme recognition task in pronunciation evaluation, this paper proposes a Chinese pronunciation phoneme recognition model based on the improved Zipformer-RNN-T(Pruned) architecture, aiming to improve recognition accuracy and reduce parameter count. First, the AISHELL1-PHONEME and ST-CMDS-PHONEME datasets for Mandarin phoneme recognition through data preprocessing. Then, three layers of the Zipformer Block architecture are introduced into the Zipformer encoder to significantly enhance model performance. In the stateless Pred Network, the GELU activation function is adopted to effectively prevent neuron deactivation. Furthermore, a hybrid Pruned RNN-T/CTC Loss fusion strategy is proposed, further optimizing recognition performance. The experimental results demonstrate that the method performs excellently in the phoneme recognition task, achieving a Word Error Rate (WER) of 1.92% (Dev) and 2.12% (Test) on the AISHELL1-PHONEME dataset, and 4.28% (Dev) and 4.51% (Test) on the ST-CMDS-PHONEME dataset. Moreover, the model requires only 61.1M parameters, striking a balance between performance and efficiency.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Information Science and Technology, Shihezi University, Shihezi, China.
- Xinjiang Uygur Autonomous Region Education Examination Centre, Urumqi, China.
ORCID-Profile: Zhaohui Du, Baohua Yu
Förderung
- A Data-driven Regional Smart Education Service Key Technology Research and Application Demonstration
Zitieren
APA
Du, Z., Zhao, X., Li, L., Yu, B., & Miao, L. (2025). A study on phonemes recognition method for Mandarin pronunciation based on improved Zipformer-RNN-T(Pruned) modeling. PLOS One. https://doi.org/10.1371/journal.pone.0324048
Vancouver
Du Z, Zhao X, Li L, Yu B, Miao L. A study on phonemes recognition method for Mandarin pronunciation based on improved Zipformer-RNN-T(Pruned) modeling. PLOS One. 2025. doi:10.1371/journal.pone.0324048
BibTeX
@article{Du2025,
author = {Du, Zhaohui and Zhao, Xiaofeng and Li, Lin and Yu, Baohua and Miao, Lijiang},
title = {A study on phonemes recognition method for Mandarin pronunciation based on improved Zipformer-RNN-T(Pruned) modeling},
journal = {PLOS One},
year = {2025},
doi = {10.1371/journal.pone.0324048}
}
EUVIMED