Silent speech recognition method and system

A technology of speech recognition and vocal cord vibration, which is applied in character and pattern recognition, pattern recognition in signals, instruments, etc. It can solve problems such as unusable and system recognition preparation rate decline, and achieve high recognition accuracy, application fields and applications promising effect

Inactive Publication Date: 2020-09-29
TIANJIN UNIV
View PDF8 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] The rapid development of speech recognition technology provides a very attractive mode for human-computer interaction. Currently, the commonly used automatic speech recognition (Automatic Speech Recognition, ASR) system recognizes acoustic signals, such as using a microphone as a speech sensor, receiving The sound signal is conducted through the air, so when used in an environment with a lot of background noise, the readiness rate of system recognition will be significantly reduced, or it will not be usable in an environment without a physical medium for sound transmission, such as in the vacuum environment of outer space, etc.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Silent speech recognition method and system
  • Silent speech recognition method and system
  • Silent speech recognition method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] A silent speech recognition method, see figure 1 , the method includes the following steps:

[0038] 101: collecting facial electromyographic signal data and vocal cord vibration signal data when speaking;

[0039] 102: Preprocess the two collected data separately, and after feature extraction and fusion, use deep learning to perform training and recognition in sequence;

[0040] 103: Send the recognized command result to the receiving device or the controlled device.

Embodiment 2

[0042] A silent speech recognition system, see figure 2 , the system mainly includes: data acquisition module, signal processing module and communication interaction module.

[0043] Among them, the data acquisition module includes two information acquisition units: a facial electromyography signal acquisition unit and a vocal cord vibration signal acquisition unit, which respectively synchronously collect electrical signal data generated by facial muscle movement and vibration signal data during vocal cord movement when speaking in a silent manner.

[0044] Further, the signal processing module includes: a preprocessing unit, a feature extraction unit and a recognition unit.

[0045] Wherein, the preprocessing unit is used for processing the facial myoelectric signal data and the vocal cord vibration signal data received by the data acquisition module. The two kinds of data signals are preprocessed, feature extracted, and feature fused, and machine learning algorithms or de...

Embodiment 3

[0054] Combine below Figure 3-Figure 5 The scheme in embodiment 2 is further introduced, see the following description for details:

[0055] like image 3 As shown, it is a schematic diagram of the data acquisition equipment of the system, including: myoelectric sensor and throat microphone. Since the movement of facial muscles corresponds to different nerve electrical activities when people speak, the surface electrode is used as the guide electrode and placed on the On the facial skin around the mouth, the facial myoelectric signal is obtained by measuring the comprehensive potential of the muscle electrical activity at the detection electrode by close contact with the skin surface of the area where the active muscles are located. The sampling rate is 1000Hz. The collected original myoelectricity The signal is a one-dimensional signal of 4 channels. Through the laryngeal microphone close to the larynx, the vocal cord vibration of the larynx when the user speaks will cause...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a silent speech recognition method and system. The method comprises the following steps: acquiring facial electromyographic signal data and vocal cord vibration signal data during speaking; respectively preprocessing two types of acquired data, extracting and fusing features, and sequentially performing training and identifying by using deep learning; and sending the identified command result to the receiving equipment or the controlled equipment. The system comprises a data acquisition module, a signal processing module and a communication interaction module. Comparedwith a silent speech recognition method based on a single signal, the silent speech recognition method integrates two characteristic signals used in a silent speech recognition technology, including afacial electromyographic signal and a vocal cord vibration signal, so that higher recognition accuracy can be obtained.

Description

technical field [0001] The invention relates to the field of voice recognition, in particular to a silent voice recognition method and system. Background technique [0002] The rapid development of speech recognition technology provides a very attractive mode for human-computer interaction. Currently, the commonly used automatic speech recognition (Automatic Speech Recognition, ASR) system recognizes acoustic signals, such as using a microphone as a speech sensor, receiving The sound signal is conducted through the air, so when used in an environment with a lot of background noise, the readiness rate of system recognition will be significantly reduced, or it will not be usable in an environment without a physical medium for sound transmission, such as in the vacuum environment of outer space, etc. . [0003] Due to the technical limitations of ASR and other problems, the silent speech recognition system will be able to effectively overcome the above limitations, and has bro...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04
CPCG06N3/045G06N3/044G06F2218/04G06F2218/08G06F18/253
Inventor 赵涛陶文源闫野印二威马权智刘璇恒谢良
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap