Computer readable storage medium and bone voiceprint detection headphone applying same

Abstract

The invention relates to the technical field of headphones, in particular to a computer readable storage medium and a bone voiceprint detection headphone applying the same. A computer program is stored in the medium, and can be executed by a processor of the bone voiceprint detection headphone. A bone voiceprint signal is acquired by a bone conduction sensor, which has very high sensitivity in detecting the vibration caused by skull muscles when people speak. Three-dimensional coordinates are determined by taking a direction perpendicular to human ears, i.e., a direction perpendicular to the receiving end surface of the bone conduction sensor as a Z axis, and whether the headphone is worn properly or not can be judged by the data of an X / Y / Z triaxial space and the size of the headphone soas to ensure that the acquired bone voiceprint signal is valid. Then, the acquired bone voiceprint signal is analyzed in frequency domain for a limited time, and the voiceprint features of the bone voiceprint signal are extracted. The voiceprint features are matched with a voice matching library to determine whether the bone voiceprint signal is a target signal of the voice matching library or not, i.e., database feature matching is adopted to identify whether a current speaker is the speaker or not.

Description

technical field [0001] The present invention relates to the technical field of earphones, in particular to a computer-readable storage medium and a bone voiceprint detection earphone using the medium. A computer program is stored in the medium, and the computer program can be executed by a processor of the bone voiceprint detection earphone. Background technique [0002] With the rapid development of artificial intelligence, various new types of recognition and detection technologies such as face recognition and fingerprint recognition emerge in an endless stream, while voiceprint recognition technology is not a very new recognition technology. As early as the 1960s and 1970s, public security organs began to use voice Voiceprint recognition technology is used to investigate criminal cases. Because the collection of identification materials is more natural and it is more difficult to falsify, voiceprint recognition is still an indispensable method in modern recognition technol...

AI Technical Summary

Problems solved by technology

However, relying solely on voiceprint recognition technology, there are still relatively large risks in occasions that only require simple instructions such as mobile phone unlocking and mobile payment.

Different from the traditional voiceprint recognition technology that conducts sound through the air, bone conduction transmits sound through the skull and lymphatic system. Conduction wearable products, such as bone conduction earphones, hearing aids, etc., but limited by the recognition rate, there is still a large risk of misidentification

Images