Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

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

Sort by

Same author

Bionic Technology in Prosthetics: Multi-Objective Optimization of a Bioinspired Shoulder-Elbow Prosthesis with Embedded Actuation.

Biomimetics (Basel, Switzerland)·2026

Same author

Multimodal Field-Driven Actuation in Bioinspired Robots: An Emerging Taxonomy and Roadmap Towards Hybrid Intelligence.

Biomimetics (Basel, Switzerland)·2025

Same author

Bio-Inspired Compliant Joints and Economic MPC Co-Design for Energy-Efficient, High-Speed Locomotion in Snake-like Robots.

Biomimetics (Basel, Switzerland)·2025

Same author

Design, modeling and validation of a low-cost linkage-spring telescopic rod-slide underactuated adaptive robotic hand.

Bioinspiration & biomimetics·2024

Same author

Dormancy break, sprouting and later tuber reproduction in response to different tuber sizes of tiger nut (<i>Cyperus esculentus</i> L.).

Royal Society open science·2024

Same author

Analog spatiotemporal feature extraction for cognitive radio-frequency sensing with integrated photonics.

Light, science & applications·2024

Search research articles

Related Experiment Video

Updated: May 14, 2026

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

Decoding Mandarin Action Verbs from EEG Using a Dual-LSTM Network: Towards Practical Assistive Brain-Computer

Binshuo Liu¹, Gengbiao Chen², Lairong Yin²

¹International College of Engineering, Changsha University of Science and Technology, Changsha 410114, China.

Sensors (Basel, Switzerland)

|May 13, 2026

Summary

This summary is machine-generated.

This study demonstrates that a Long Short-Term Memory (LSTM) network can decode six Mandarin verbs from electroencephalogram (EEG) signals with 70% accuracy. This brain-computer interface (BCI) technology shows promise for assistive communication, especially for tonal languages.

Keywords:

EEG signal processing Mandarin verb decoding assistive technology brain–computer interfaces (BCI)recurrent neural networks (RNN)

More Related Videos

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Published on: November 24, 2015

STFEEG-Tool: A Spatial-Temporal-Frequency EEG Analysis Tool for Motor Imagery Brain-Computer Interfaces

STFEEG-Tool: A Spatial-Temporal-Frequency EEG Analysis Tool for Motor Imagery Brain-Computer Interfaces

Published on: March 10, 2026

Related Experiment Videos

Last Updated: May 14, 2026

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Published on: November 24, 2015

STFEEG-Tool: A Spatial-Temporal-Frequency EEG Analysis Tool for Motor Imagery Brain-Computer Interfaces

STFEEG-Tool: A Spatial-Temporal-Frequency EEG Analysis Tool for Motor Imagery Brain-Computer Interfaces

Published on: March 10, 2026

Area of Science:

Neuroscience
Computer Science
Linguistics

Background:

Electroencephalogram (EEG)-based brain-computer interfaces (BCIs) are crucial for restoring communication.
Decoding Mandarin, a tonal language, from EEG is difficult due to homophones and complex temporal dynamics.

Purpose of the Study:

To investigate the decoding of six high-frequency Mandarin action verbs from EEG signals.
To evaluate the effectiveness of a recurrent neural network (RNN) framework with dual Long Short-Term Memory (LSTM) layers for this task.

Main Methods:

A visual-cue-based overt speech production experiment was conducted with 30 participants.
EEG data were collected during visually guided reading aloud of six Mandarin verbs: Chi (eat), He (drink), Chuan (wear), Na (take), Kan (look), and Dai (put on).
A dual LSTM network was implemented and compared against a Common Spatial Pattern combined with Support Vector Machine (CSP-SVM) baseline.

Main Results:

The LSTM model achieved an average classification accuracy of 69.93% ± 3.07%, significantly outperforming the CSP-SVM baseline (36.53% ± 3.17%).
Accuracy surpassed 75% with over 15 training repetitions and 38% training data.
The LSTM model demonstrated data efficiency, achieving high performance with only 38% of trial data for training.

Conclusions:

The LSTM architecture effectively captures neural signatures for Mandarin verb processing.
This research provides a foundation for developing practical EEG-based assistive communication technologies for tonal languages.
The model's inference latency under 2 seconds supports near-real-time applications.