Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Decoding imagined Chinese speech: a capsule neural network based on bidirectional knowledge transfer for hierarchical

Jingyu Gu1, Qian Cai2, Haixian Wang1

  • 1Key Laboratory of Child Development and Learning Science of Ministry of Education, School of Biological Science & Medical Engineering, Southeast University, Nanjing 211189 Jiangsu, People's Republic of China.

Journal of Neural Engineering
|June 22, 2026
PubMed
Summary

Related Concept Videos

Classification of Systems-I01:26

Classification of Systems-I

Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
Classification of Systems-II01:31

Classification of Systems-II

Continuous-time systems have continuous input and output signals, with time measured continuously. These systems are generally defined by differential or algebraic equations. For instance, in an RC circuit, the relationship between input and output voltage is expressed through a differential equation derived from Ohm's law and the capacitor relation,

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Deep learning for EEG-based sleep stage classification: a review.

Medical & biological engineering & computing·2026
Same author

Aligning machines and minds: neural encoding for high-level visual cortices based on image captioning task.

Journal of neural engineering·2025
Same author

Impact of fatigue levels on EEG-based personal recognition.

Medical & biological engineering & computing·2025
Same author

The hotspots and publication trends in glioblastoma and CAR-T immunotherapy: A bibliometric analysis.

Human vaccines & immunotherapeutics·2025
Same author

SynBioNanoDesign: pioneering targeted drug delivery with engineered nanomaterials.

Journal of nanobiotechnology·2025
Same author

Lasmiditan and Different Triptans in Menstrual Migraine: A Bayesian Network Meta-analysis.

Pain and therapy·2025
This summary is machine-generated.

This study introduces a novel hierarchical capsule network for Chinese silent brain-computer interfaces (BCIs). The method effectively decodes hierarchical speech imagery using linguistic structure, achieving high recognition rates.

Area of Science:

  • Neuroscience
  • Artificial Intelligence
  • Linguistics

Background:

  • Brain-computer interfaces (BCIs) research is advancing, with speech imagination being a key area.
  • Silent BCIs for Chinese languages are under-researched, especially those leveraging linguistic structures.

Purpose of the Study:

  • To design an experimental paradigm for Chinese speech imagery utilizing its initial-and-final structure.
  • To develop a method for effectively utilizing hierarchical structural information in multi-granularity classification tasks for BCIs.

Main Methods:

  • A hierarchical capsule network with bidirectional knowledge transfer was proposed.
  • The network utilizes a dynamic routing mechanism tailored for Mandarin Chinese phonology.
  • Bidirectional knowledge transfer (forward and reverse) was employed to enhance feature dependency modeling and mitigate error propagation.
Keywords:
bidirectional knowledge transfercapsule neural networkclass hierarchical structuremulti-label imagined Chinese speech decodingsilent brain–computer interfaces

Related Experiment Videos

Main Results:

  • The proposed algorithm demonstrated effectiveness in hierarchical classification tasks.
  • Highest recognition rates achieved were 90.86% (layer 1), 73.69% (layer 2), and 69.45% (layer 3).

Conclusions:

  • The study offers a novel approach to decoding hierarchical Chinese silent BCI paradigms.
  • Linguistic domain knowledge can guide neural network design for specific BCI applications.
  • The findings provide a foundation for future individual phoneme classification in BCIs.