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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Sequence-to-Sequence Voice Reconstruction for Silent Speech in a Tonal Language.

Brain sciences·2022
Same author

Qualitative Recognition of Primary Taste Sensation Based on Surface Electromyography.

Sensors (Basel, Switzerland)·2021
Same author

Inductive conformal prediction for silent speech recognition.

Journal of neural engineering·2020
Same author

Transcriptomic Profiling of the Developing Cardiac Conduction System at Single-Cell Resolution.

Circulation research·2019
Same author

Deconvolution and network analysis of IDH-mutant lower grade glioma predict recurrence and indicate therapeutic targets.

Epigenomics·2019
Same author

CT Super-Resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE).

IEEE transactions on medical imaging·2019

Related Experiment Video

Updated: Dec 14, 2025

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

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

Published on: July 26, 2013

43.9K

Silent Speech Decoding Using Spectrogram Features Based on Neuromuscular Activities.

You Wang1, Ming Zhang1, RuMeng Wu1

  • 1State Key Laboratory of Industrial Control Technology, Institute of Cyber Systems and Control, Zhejiang University, Hangzhou 310027, China.

Brain Sciences
|July 16, 2020
PubMed
Summary
This summary is machine-generated.

This study decodes silent speech using Brain-Computer Interface (BCI) technology and surface electromyography (sEMG) signals. Deep learning models achieved 90% accuracy in recognizing articulatory muscle movements.

Keywords:
Xceptionbidirectional long short-term memoryneuromuscular signalsilent speech decodingspectrogram features

More Related Videos

A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS
12:43

A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS

Published on: February 21, 2011

35.7K
Functional Near Infrared Spectroscopy of the Sensory and Motor Brain Regions with Simultaneous Kinematic and EMG Monitoring During Motor Tasks
11:31

Functional Near Infrared Spectroscopy of the Sensory and Motor Brain Regions with Simultaneous Kinematic and EMG Monitoring During Motor Tasks

Published on: December 5, 2014

15.5K

Related Experiment Videos

Last Updated: Dec 14, 2025

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

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

Published on: July 26, 2013

43.9K
A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS
12:43

A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS

Published on: February 21, 2011

35.7K
Functional Near Infrared Spectroscopy of the Sensory and Motor Brain Regions with Simultaneous Kinematic and EMG Monitoring During Motor Tasks
11:31

Functional Near Infrared Spectroscopy of the Sensory and Motor Brain Regions with Simultaneous Kinematic and EMG Monitoring During Motor Tasks

Published on: December 5, 2014

15.5K

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Signal Processing

Background:

  • Silent speech decoding is an emerging Brain-Computer Interface (BCI) application.
  • It utilizes articulatory neuromuscular activities, simplifying data acquisition and processing.
  • Surface electromyography (sEMG) is a viable method for capturing these signals during mimed speech.

Purpose of the Study:

  • To investigate spatial features and decoders for recognizing neuromuscular signals in silent speech.
  • To evaluate the effectiveness of transfer learning and deep learning methods for this task.
  • To compare the performance of different deep learning architectures in silent speech recognition.

Main Methods:

  • sEMG data were collected from subjects performing mimed speech.
  • sEMG data were transformed into spectrograms for rich time-frequency information.
  • Transfer learning (Xception model) was used for feature extraction.
  • Deep learning models (Multi-Layer Perception, Convolutional Neural Network, bidirectional Long Short-Term Memory) were trained and evaluated.

Main Results:

  • The proposed decoders successfully recognized silent speech patterns.
  • Bidirectional Long Short-Term Memory achieved the highest accuracy at 90%.
  • The results validate the use of spectrogram features and deep learning for silent speech decoding.

Conclusions:

  • Spectrogram features effectively capture information for silent speech recognition.
  • Deep learning algorithms, particularly bidirectional Long Short-Term Memory, are highly effective for decoding silent speech from sEMG.
  • This research demonstrates a promising approach for BCI-based silent speech communication.