EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

STCCA: Spatial–Temporal Coupled Cross-Attention Through Hierarchical Network for EEG-Based Speech Recognition

Liang Dong iD, Hengyi Shao, Lin Zhang, Lei Li iD

Sensors · 2025

DOI: 10.3390/s25216541 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Liang Dong und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China. Hauptthemen laut MEDLINE: Electroencephalography, Speech, Speech Recognition Software. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Speech recognition based on Electroencephalogram (EEG) has attracted considerable attention due to its potential in communication and rehabilitation. Existing methods typically process spatial and temporal features with sequential, parallel, or constrained feature fusion strategies. However, the intricate cross-relationships between spatial and temporal features remain underexplored. To address these limitations, we propose a spatial–temporal coupled cross-attention mechanism through a hierarchical network, named STCCA. The proposed STCCA consists of three key components: local feature extraction module (LFEM), coupled cross-attention (CCA) fusion module, and global feature extraction module (GFEM). The LFEM employs CNNs to extract local temporal and spatial features, while the CCA fusion module leverages a dual-directional attention mechanism to establish deep interactions between temporal and spatial features. The GFEM uses multi-head self-attention layers to model long-range dependencies and extract global features comprehensively. STCCA is validated on three EEG-based speech datasets, achieving accuracies of 45.45%, 25.91%, and 29.07%, corresponding to improvements of 1.95%, 3.98%, and 1.98% over the comparison models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ElectroencephalographySpeechSpeech Recognition SoftwareHumansNeural Networks, ComputerAlgorithmsSignal Processing, Computer-Assisted

Stichwörter der Autor:innen

EEG-based speech recognitioncross-attention fusiondeep learningmulti-head self-attention

Autor:innen und Institutionen

  • School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China.
  • Beijing BigData Center, Beijing 101100, China.

ORCID-Profile: Liang Dong, Lei Li

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Dong, L., Shao, H., Zhang, L., & Li, L. (2025). STCCA: Spatial–Temporal Coupled Cross-Attention Through Hierarchical Network for EEG-Based Speech Recognition. Sensors. https://doi.org/10.3390/s25216541

Vancouver

Dong L, Shao H, Zhang L, Li L. STCCA: Spatial–Temporal Coupled Cross-Attention Through Hierarchical Network for EEG-Based Speech Recognition. Sensors. 2025. doi:10.3390/s25216541

BibTeX
@article{Dong2025,
  author = {Dong, Liang and Shao, Hengyi and Zhang, Lin and Li, Lei},
  title = {STCCA: Spatial–Temporal Coupled Cross-Attention Through Hierarchical Network for EEG-Based Speech Recognition},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s25216541}
}
Kostenlos starten →