A silent communication method, system and storage medium

Abstract

The invention discloses a silent communication method, system and storage medium to meet information communication requirements in some specific scenarios. The silent communication method includes: during the speaking process of the first user, respectively collecting the first IMU signal, the first EMG signal and the first voice signal; determining the first shape position change information of joint muscles according to the first IMU signal; according to the first The EMG signal determines the first myoelectric change information; extracts the first acoustic representation feature information from the first speech signal; performs training based on the first shape position change information, the first myoelectric change information and the first acoustic representation feature information, Obtaining a speech prediction model; using the speech prediction model, during the process of speech expression by the second user, identifying speech information expressed by the second user.

Description

technical field [0001] The present invention relates to the technical field of voice signal processing, in particular to a silent communication method, system and storage medium. Background technique [0002] In some specific scenarios, such as in a noisy environment, because the surrounding environment is too noisy, the sound will be affected, resulting in the inconvenience of voice communication, and even affect the transmission of information; and in some environments that require extremely quiet Communication will affect other people. [0003] At present, there are already some high-precision electronic larynxes. In the medical field, there are also many electronic devices that assist people with vocal impairment. The vibration signal when speaking achieves the purpose of reconstructing the voice signal. However, these devices have the following problems. First, it is inconvenient to use the breath to operate. It relies on the hose put into the throat to work, which is...

AI Technical Summary

Problems solved by technology

However, these devices have the following problems. First, it is inconvenient to use the breath to operate. It relies on the hose put into the throat to work, which is very unfriendly to the human body. Second, the electronic throat is converted into an analog signal through a vibration signal and then uses a digital signal for audio. For signal conversion, you need to press the sensor with your hand to make it close to your throat, and in some cases there will be problems such as distortion

Images