Snoring sound signal identification method

Abstract

The invention provides a snoring sound signal identification method. The method comprises the following steps of acquiring audio information of a preset sleep time period, and extracting Mel-frequencycepstral coefficients of the audio information as training samples; classifying the training samples into snoring sounds and non snoring sounds by utilizing a k-means clustering algorithm, and then removing the non snoring sounds according to a clustering result, so that a snoring sound training set is obtained; performing training by a Gaussian mixture model to obtain an identification model; calculating the generation probabilities of the snoring sounds in the training set by utilizing the identification model, arranging the generation probabilities in an ascending order, and taking front preset percentage data as a snoring sound generation probability threshold of an experiment object; and acquiring an audio segment by an audio acquisition device, extracting a Mel-frequency cepstral coefficient of the audio segment, then using the model to calculate the generation probability of the audio segment, and judging whether the audio segment is the snoring sound of the experiment object or not. According to the method, preprocessing operation is carried out on an original snoring sound signal, so that the processing amount of the data can be effectively reduced, and the distinguishingdegree of the snoring sound and non-snoring sound signals can be increased.

Description

technical field [0001] The invention relates to the technical field of snoring signal processing, in particular to a snoring signal recognition method. Background technique [0002] Sleep-disordered breathing is abnormal breathing that occurs during sleep, including sleep apnea syndrome, hypoventilation syndrome, and related sleep-disordered breathing caused by chronic pulmonary and neuromuscular diseases, among which obstructive eye apnea syndrome is in ( OSAS) mainly. [0003] Studies have shown that OSAS can cause symptoms such as daytime drowsiness, dizziness, headaches, memory loss, fatigue, unresponsiveness, and abnormal sleep behaviors. Long-term OSAS can cause hypertension, coronary heart disease, heart failure, stroke and other diseases. The medical community attaches great importance to the study of this disease, and has achieved significant results. However, most of the instrument detection costs are expensive, which is not conducive to popularization. Some ex...

AI Technical Summary

Problems solved by technology

However, most of the instrument detection costs are expensive, which is not conducive to popularization

Some existing snoring detection algorithms also have shortcomings such as large errors and slow processing speed.

Images