Silent voice acquisition processing method and equipment based on ear cavity vibration

Abstract

The invention discloses a silent voice acquisition processing method and equipment based on ear cavity vibration. The method comprises the following steps that: 1) a silent voice acquisition unit continuously monitors a vibration signal of an ear cavity of a wearer, and when the vibration amplitude or frequency change of the vibration signal is higher than a set threshold value, an air vibration sensor and a bone vibration sensor in the silent voice acquisition unit start to acquire signals at the same time and send the acquired signals to a data processing end; 2) after receiving the signals, the data processing end separately divides the collected continuous voice signals and vibration signals into a plurality of frames by taking a set time length as a basic unit, calculates the average energy and the average frequency of each frame, and then compares a calculation result with a set energy threshold to determine whether a user inputs traditional voice or silent voice; 3) an end-to-end voice recognition model is called to perform voice content recognition for traditional voice input, for silent voice input, a convolutional neural network model is used to perform feature extraction, fusion and content recognition on collected vibration signals.

Description

technical field [0001] The invention belongs to the field of human-computer interaction, and relates to a silent voice input suitable for various noise environments, in particular to a silent voice collection and processing method and equipment based on ear cavity vibration. Background technique [0002] Speech input has always been one of the key issues in the field of human-computer interaction research. Traditional voice input has obvious disadvantages because of its high requirements on the environment. For example, it is inconvenient to use normal volume of voice for text input in public places due to privacy issues and use traditional voice for input in noisy environments. When the text recognition effect is not good, etc. Therefore, traditional voice input is only suitable for use in a relatively quiet (no noise) and private environment, which completely cannot meet the needs of users for voice input in various noise environments. At present, in order to avoid colle...

AI Technical Summary

Problems solved by technology

Traditional voice input has obvious disadvantages because of its high requirements on the environment. For example, it is inconvenient to use normal volume of voice for text input in public places due to privacy issues and use traditional voice for input in noisy environments. When the text recognition effect is not good, etc.

Therefore, the traditional voice input is only suitable for use in a relatively quiet (no noise) and private environment, which cannot meet the needs of users for voice input in a variety of different noise environments.

At present, in order to avoid collecting various environmental noises around the user's voice input, voice collection methods based on bone conduction have emerged. Although these methods do not collect environmental noise to a certain extent, because the source of the collected signal is still high-decibel traditional voice input, so the voice accuracy collected by these methods is greatly affected by the accuracy of the user's pronunciation

However, there is still no relevant technology and research on low-decibel silent speech acquisition and recognition based on the ear cavity.

Images