Nighttime sleeping sound signal analyzing method based on multiple characteristics

Abstract

The invention discloses a nighttime sleeping sound signal analyzing method based on multiple characteristics. The method comprises the steps that firstly, nighttime sleeping sound signal data, recorded by a microphone, of a person to be detected are read and endpoint detection is conducted on a nighttime sleeping sound signal through a short-time energy threshold method; secondly, a plurality of acoustic characteristics of the nighttime sleeping sound signal are extracted through a series of signal processing algorithms; finally, a database used for following analysis and research is established according to the nighttime sleeping sound signal characteristics of the person to be detected through a basic statistics method. The method has the advantages that being easy to realize, simple, low in hardware cost after a product is developed and wide in application prospect.

Description

technical field [0001] The invention belongs to the technology of realizing the database establishment of human night sleep sound signal by means of sound signal processing, in particular to a night sleep sound signal analysis method based on multi-features. Background technique [0002] The analysis of the human body's nighttime sleep sound signal is extremely important for the development of a health monitoring system in a smart home. Nighttime sleep sound signals that are often paid attention to usually include abnormal speech acoustic signals such as snoring, coughing, and moaning. These nocturnal sleep sound signals are of great significance for early monitoring and detection of abnormal snoring, judgment of health status, and monitoring of sleepwalking. [0003] The use of acoustic signal analysis means to extract features of the subject's sleep sound signal at night is the basis for subsequent large-scale data analysis, mathematical modeling and pattern recognition. ...

AI Technical Summary

Problems solved by technology

Scholars in Taiwan have proposed a method for data analysis of sleep sound signals throughout the night (1. Liao Wenhung, Su Yisyuan. Classification of audio signals in all-night sleep studies [C]. Proc. 18th Int. Conf. on Pattern Recognition, Hong Kong Kong, China, August2006, 302-305), but they did not involve the establishment of a database of the night sleep sound signals of the subjects, and the selected features were single, and the physical meaning was not clear

Therefore, it cannot reflect the characteristic distinction of sound source signals with specific physical meanings, so it is not suitable for subsequent analysis and modeling

[0004] It can be seen from the above that the existing technology has defects such as incomplete methods, single function, and unsuitability for follow-up research.

Images