Silent speech recognition method based on face and neck surface myoelectricity

Abstract

The invention discloses a silent speech recognition method based on face and neck surface electromyography, and the method comprises the steps: carrying out the data preprocessing and feature extraction of surface electromyography signals collected by a high-density electrode array and discrete electrodes, obtaining a high-density sEMG image set and a channel sparse sEMG image set, and constructing a source domain database and a target domain database; then training the word classification deep neural network by using the source domain database, and completing the calibration of the network in the target domain database by using transfer learning; and if the test user expresses the words silently under the input of the discrete electrodes, wherein the calibrated network can complete word classification and realize silent speech recognition. According to the invention, the capability of capturing abundant muscle activation mode information of a high-density electrode array and the portability and easiness in wearing of discrete electrodes are considered, certain robustness is provided for slight electrode offset and cross-user conditions, and the performance of silent speech recognition under discrete electrode input is improved; and a new thought is provided for the silent speech recognition method.

Description

technical field [0001] The invention belongs to the fields of biological signal processing, machine learning and intelligent control, and specifically relates to a silent voice recognition method based on facial and neck surface electromyography. Background technique [0002] Voice interaction is one of the most natural and direct ways for people to interact, because the voice signal contains information such as the emotion and intention that the speaker wants to express. Automatic speech recognition (automatic speech recognition, ASR) refers to the computer to analyze and understand the collected speech signal, and convert it into text or other forms of information. ASR plays a vital role in human-computer interaction, but it also has limitations in special scenarios, such as high-noise backgrounds, people with vocal impairments, and private input environments. Therefore, how to overcome these difficulties in practical applications has always been a hot topic in the resear...

AI Technical Summary

Problems solved by technology

Transfer learning methods based on deep neural networks can adapt EMG pattern classifiers to current electrode positions or new user domains, but also complicate the use of EMG control and increase the training burden for users

Moreover, most of these studies focus on different tasks under the same measurement electrode conditions, and the migration methods across electrode domains are hardly considered and discussed.

Images