EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CAs-Net: A Channel-Aware Speech Network for Uyghur Speech Recognition

Jiang Zhang iD, Miaomiao Xu, Lianghui Xu, Yajing Ma

Sensors · 2025

DOI: 10.3390/s25123783 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Jiang Zhang und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Xinjiang University, Urumqi 830017, China. Hauptthemen laut MEDLINE: Speech Recognition Software, Speech. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Graduate Research Innovation Project of Xinjiang Uygur Autonomous Region.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper proposes a Channel-Aware Speech Network (CAs-Net) for low-resource speech recognition tasks, aiming to improve recognition performance for languages such as Uyghur under complex noisy conditions. The proposed model consists of two key components: (1) the Channel Rotation Module (CIM), which reconstructs each frame’s channel vector into a spatial structure and applies a rotation operation to explicitly model the local structural relationships within the channel dimension, thereby enhancing the encoder’s contextual modeling capability; and (2) the Multi-Scale Depthwise Convolution Module (MSDCM), integrated within the Transformer framework, which leverages multi-branch depthwise separable convolutions and a lightweight self-attention mechanism to jointly capture multi-scale temporal patterns, thus improving the model’s perception of compact articulation and complex rhythmic structures. Experiments conducted on a real Uyghur speech recognition dataset demonstrate that CAs-Net achieves the best performance across multiple subsets, with an average Word Error Rate (WER) of 5.72%, significantly outperforming existing approaches. These results validate the robustness and effectiveness of the proposed model under low-resource and noisy conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Speech Recognition SoftwareSpeechHumansLanguageAlgorithmsChinaNeural Networks, Computer

Stichwörter der Autor:innen

channel modelinglow-resource speech recognitionmulti-scale convolution

Autor:innen und Institutionen

  • School of Computer Science and Technology, Xinjiang University, Urumqi 830017, China.
  • Medical Engineering and Technology, Xinjiang Medical University, Urumqi 830017, China.
  • School of Computer Science and Technology, Xinjiang Normal University, Urumqi 830054, China.

ORCID-Profile: Jiang Zhang

Förderung

  • Graduate Research Innovation Project of Xinjiang Uygur Autonomous Region

Zitieren

APA

Zhang, J., Xu, M., Xu, L., & Ma, Y. (2025). CAs-Net: A Channel-Aware Speech Network for Uyghur Speech Recognition. Sensors. https://doi.org/10.3390/s25123783

Vancouver

Zhang J, Xu M, Xu L, Ma Y. CAs-Net: A Channel-Aware Speech Network for Uyghur Speech Recognition. Sensors. 2025. doi:10.3390/s25123783

BibTeX
@article{Zhang2025,
  author = {Zhang, Jiang and Xu, Miaomiao and Xu, Lianghui and Ma, Yajing},
  title = {CAs-Net: A Channel-Aware Speech Network for Uyghur Speech Recognition},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s25123783}
}
Kostenlos starten →