59 results about "Breath sound" patented technology

The embodiment of the invention provides a continuous breath sound monitoring application system, which is oriented to double versions of a medical professional version and a civil public version, can continuously collect and analyze breath sound, adopts a multi-level label system to carry out classified labeling, and combines expert labeling and machine learning to improve the recognition accuracy. And meanwhile, a perfect database architecture and a local / cloud mixed storage strategy are provided to ensure reliable management of data, and data security compliance is ensured through encryption and access control. The system is also integrated with an interactive teaching module to meet the requirements of professional training and popular science popularization, and is equipped with a multi-parameter linkage grading alarm mechanism to provide real-time early warning. In addition, the system has good expandability and can be integrated with a hospital information system (HIS) and a remote medical platform through standard interfaces.

The invention discloses a child abnormal breathing early warning and remote cooperation system for a family environment, and relates to the technical field of breathing data processing, and the child abnormal breathing early warning and remote cooperation system comprises a data acquisition module, a first data calculation module, a first data judgment module, a second data judgment module, a second data calculation module, a third data judgment module and a data output module. The system sequentially performs primary risk assessment, hidden risk mining and potential decompensation index calculation on the basis of a multi-modal signal processing technology by synchronously acquiring a breathing sound signal, a non-contact thoracoabdominal movement signal and a contact thoracoabdominal movement signal, so that graded early warning is realized, and a remote cooperation mechanism is automatically triggered according to the early warning level. According to the method, the hidden risk in the early stage of respiratory compensation of children can be effectively identified, and advanced early warning and timely intervention of abnormal respiration can be realized in a family environment.

The invention discloses a breath sound electronic auscultation signal identification method based on multi-mode and deep learning, and relates to the technical field of deep learning and multi-mode learning. Firstly, audio data related to breathing sound classification and text description corresponding to the audio data are obtained; preprocessing the audio data, including de-noising and feature extraction; thirdly, matching an audio sample with the generated text description through a comparative learning method so as to optimize the representation capability of the model; evaluating and verifying the trained model, and performing label prediction on an audio sample by using a classification head; and finally, combining the metadata with the audio features by adopting a multi-modal fusion method, and carrying out classification prediction through a reasoning stage. On the basis of focusing on the classification accuracy of the model, the multi-modal input effect of the model is focused at the same time, the robustness of the model is improved by integrating the metadata, and whether the model can still keep good classification performance or not under the condition that part of the metadata is missing is verified. In addition, the method also reveals the importance of different metadata on the model prediction result and the interaction relationship of the different metadata. In conclusion, the invention can provide reliable support for clinical application.

The invention discloses a method and a system for designing a breath sound classification recognition accelerator based on an FPGA (Field Programmable Gate Array), and relates to the technical field of breath sound recognition. The method specifically comprises the following steps that a breath sound data set is established, and the breath sound data set comprises multiple kinds of breath event sounds such as normal sounds, dry rale sounds, gasp sounds and gasp sounds; extracting breathing sound features based on the breathing sound data set to obtain a breathing sound feature map; taking the breath sound feature map as a learning sample, and training a breath sound classification recognition model based on a neural network; and using the trained breath sound classification recognition model to perform inversion on the breath sound data set to obtain the recognition accuracy of different types of breath sounds. Compared with a stethoscope, the method has the advantages that the human resource cost is reduced, and the difference caused by human subjective judgment is reduced. Compared with other hardware platforms such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), the system has the advantages of lower power consumption, higher response speed and higher portability.

Through the multi-sensor fusion design of the microphone, the millimeter wave radar sensor and the piezoelectric film sensor, compared with an existing single or double-sensor sleep monitoring technology, the monitoring precision, the function dimension, the environmental adaptability and the like are remarkably improved. According to the method, a weighted voting decision fusion mechanism and an improved random forest model are adopted, and after a microphone acoustic monitoring dimension is newly added, the apnea detection sensitivity is improved; acoustic characteristics such as snore, breath sound rhythm, cough and dream can be accurately captured; health risk assessment dimensions are richer, preliminary screening of diseases such as sleep apnea syndrome and periodic limb dyskinesia can be realized, snore indexes and breathing stability are newly increased, and a multi-dimensional sleep quality report is provided for a user; in the whole process, an electrode plate or a wrist strap does not need to be worn, the three sensors all adopt non-intrusive design, no foreign body sensation exists after the piezoelectric film is embedded into the mattress, the problem of wearing discomfort of traditional monitoring is solved, and the user compliance is improved.

The invention relates to the technical field of audio signal processing. The invention provides a method, system and device for predicting pulmonary nodules through sound and a storage medium. The method comprises the following steps: in a quiet test environment, directionally collecting a sound signal of a patient through an audio sensor; wherein the sound signals comprise breath sounds, cough sounds, vowels and sentences; performing noise reduction processing on the collected sound signals to obtain a noise reduction data set; extracting multi-dimensional voiceprint features from the noise reduction data set, and constructing a sound feature matrix; and performing multi-modal fusion analysis on the sound feature matrix and clinical data of the patient, inputting the sound feature matrix and the clinical data into a sound prediction model optimized by a focus loss function, and outputting a pulmonary nodule risk prediction result and a visual diagnosis report. The problems that the misdiagnosis and missed diagnosis rate is high, the radiation risk is large and the equipment dependence is strong when the pulmonary nodules are examined by the existing iconography, and the characteristic analysis is single, the noise immunity is poor and the screening accuracy is insufficient in the sound detection technology are solved.

The invention relates to the technical field of health management systems, in particular to a respiration monitor and a nursing monitoring device.The respiration monitor comprises a sound pickup tube bound to the throat of a patient and used for collecting respiration sound of the patient; the converter is used for sensing the breathing sound, converting the breathing sound into an electric signal and analyzing and processing the electric signal to obtain a real-time monitoring waveform; one end of the monitoring interface module is connected with the converter through a cable, the other end is connected with the monitor, and the monitoring interface module is used for transmitting real-time monitoring waveforms to the monitor; and the monitor is used for displaying the monitoring waveform so as to display the breathing frequency and the breathing mode of the patient, collecting the breathing sound through the throat and processing the breathing sound, so that the breathing state of the patient is effectively monitored, and the nursing efficiency of medical staff on the patient is improved.

The invention provides a breath sound signal processing and evaluating method and system. The method comprises the following steps: firstly, collecting breath sound signals of a plurality of preset parts of the chest of a user through an electronic stethoscope, and collecting clinical symptom data through a user interface; a first feature set and a second feature set are then extracted from the breath sound signal, where the first feature set is extracted from six sites for preliminary analysis and the second feature set is extracted from four sites for in-depth analysis. And performing analysis based on the first feature set and the first machine learning model, and outputting a first indication value. Decision is made in combination with clinical symptom indication values, and deep analysis is started when any indication value reaches a threshold value. The deep analysis outputs a second indication value based on a second feature set and a second machine learning model. And finally, fusing a plurality of indication values to generate a comprehensive evaluation result.

The present application relates to the technical field of signal processing, in particular to a health condition evaluation method based on wavelet packet decomposition, comprising: constructing a classification and recognition network model; recording patient breathing sound audio, and preprocessing the breathing sound audio to obtain an audio sequence; extracting features of the audio sequence through a wavelet packet decomposition algorithm to obtain a feature vector; inputting the feature vector into the classification and recognition network model for training to obtain a state change; and assigning scores after comparing the state change with a health state to obtain an evaluation result, wherein the present application evaluates the health condition through the constructed classification and recognition network model based on sound signals, improves the accuracy of the examination result, and solves the problem of low accuracy of the examination result of the existing evaluation method.

The music automatic selection method comprises the following steps: receiving a pre-recorded background sound and a real-time environment sound through a processor; generating, by a processor, a reduced noise according to the pre-recorded background sound and the real-time ambient sound; detecting, by a processor, a breathing sound in the reduced noise to generate a breathing frequency; and selecting, by the processor, a music according to the respiratory rate, wherein the number of beats per minute of the music corresponds to the respiratory rate.

This invention discloses a multimodal respiratory disease-related data acquisition system and method. The data acquisition system includes a disposable oral-nasal breathing interface, an oral-nasal breath sound acquisition module, a bidirectional flow / differential pressure sensing module, a temperature and humidity auxiliary acquisition module, a chest and back multi-channel auscultation module, a channel auscultation front end, a trimodal respiratory data acquisition box, a mobile terminal annotation module, a clinical label association module, and a local storage module. The data acquisition method is as follows: Step 1, prepare the oral-nasal breathing interface and acoustic coupling patch; Step 2, start the trimodal respiratory data acquisition box; Step 3, the patient performs quiet breathing; Step 4, the patient performs a maximal inhalation followed by a forceful exhalation; Step 5, associate the clinical examination results with the acquired data; Step 6, output risk scores, abnormality prompts, or classification prompts. Beneficial effects: Reduces wind noise interference, allows acoustic signals and lung function data to be acquired at the same time reference, and reduces frictional noise and poor contact.

The invention discloses a breath sound signal-based multi-feature fusion enhanced feature extraction method and system, and relates to the field of machine learning. The problems that according to a breathing sound signal modeling method in the prior art, due to the fact that average processing and dimension reduction splicing are conducted on different features in the time dimension, time sequence information loss is likely to be caused, the complementary advantages of the features cannot be effectively exerted due to the fact that the multiple features are fused in a simple splicing mode, and the recognition capacity of a model for breathing sound dynamic changes is limited are solved. The method comprises the following steps: acquiring original breath sound signal data, and preprocessing; extracting time-frequency characteristics of the preprocessed original breath sound signal data; a dual-channel synchronous random masking mechanism is designed, and random masking of time and frequency dimensions is carried out on the static features and the dynamic features at the same time; and constructing an adaptive feature fusion and enhancement module, and inputting the fused and enhanced features into a backbone network of the pre-trained AST to complete feature extraction of multi-feature fusion and enhancement based on the breath sound signals.

The invention discloses a breathing difficulty degree quantitative evaluation method and system based on artificial intelligence and a medium. The method comprises the steps that breathing sound signals, basic information and medical history information of a wearer are obtained; extracting time domain features and frequency domain features in the breath sound signals based on a preset machine learning model, and analyzing and evaluating in combination with the basic information and the medical history information to obtain a current breath condition level; according to the historical breathing data and the current breathing condition grade of the wearer, whether abnormal breathing exists or not is evaluated; if yes, the current breathing condition grade is corrected according to the motion information of the wearer in the period corresponding to the breathing sound signals, and the current dyspnea grade of the wearer is obtained. The breathing sound signals are collected in real time and analyzed and evaluated by the preset machine learning model, and the historical breathing data and the motion information are combined for analysis and correction, so that the accuracy of the evaluated dyspnea grade is improved, and the breathing condition of a wearer in daily life is accurately monitored in real time from multiple dimensions.

Monitoring module (6) for arrangement on a face mask, wherein the face mask is designed as a face half mask and has one or more paper layers and / or one or more fabric layers for covering a mouth and nose area of ​​a user and the monitoring module (6) is mechanically coupleable with the face mask and comprises the following: - an electronic circuit (61) for monitoring the use of the mouth and nose protection, - at least one first sensor element (681) arranged and configured on the face mask to provide a sensor signal suitable for detecting the breathing processes of a user, wherein the at least one first sensor element (681) - is designed and configured as a temperature sensor element to provide a sensor signal that is representative of the temperature of air located between a facial surface in the mouth and nose area of ​​the user and the mouth and nose protection, or - is designed and configured as a humidity sensor element to provide a sensor signal that is representative of the humidity of the air located between a facial surface in the mouth and nose area of ​​the user and the mouth and nose protection, or - is designed and configured as a pressure sensor element to provide a sensor signal that is representative of the air pressure of the air located between a facial surface in the mouth and nose area of ​​the user and the mouth and nose protection, or - is designed and configured as a carbon dioxide gas sensor element to provide a sensor signal representative of a carbon dioxide concentration in the air located between a facial surface in the mouth and nose area of ​​the user and the protective element of the mouth and nose protection, or - is designed and configured as a sound transducer sensor element to provide a sensor signal that is representative of a breath sound and / or exhalation sound of the user of the mouth and nose protection device or - is designed and configured as a thermal conductivity sensor element to provide a sensor signal representative of the thermal conductivity of the air located between a facial surface in the mouth and nose area of ​​the user and the protective element of the mouth and nose covering, and - a power supply unit (64) for at least the electronic circuit (61), wherein the power supply unit (64) comprises a solid-state accumulator without electrolytes.

A display device for panting-related information comprising: a breath sound detecting unit (111, 112, 115) configured to detect a breath sound of a measurement subject and obtain a breath sound signal in a time series expressing the breath sound; a determination processing unit configured to convert the respiratory sound signal into a frequency domain based on the respiratory sound signal to obtain a frequency spectrum of the respiratory sound, and to determine, for each predetermined processing unit time period, whether or not wheezing is included in the respiratory sound based on the area of ​​a dominant peak having the largest area in a graph of frequency with respect to sound pressure among a plurality of peaks in the frequency spectrum; and a display processing unit configured to display, based on a result of the determination performed by the determination processing unit, as information indicating a temporal change in the frequency of wheezing, a frequency of the processing unit time periods in which wheezing was determined by the determination processing unit to be included, in each addition unit time period (T1, T2, T3, T4) including a plurality of processing unit time periods, on a display screen (10); and further an addition processing unit configured to add the lengths of processing unit periods in which it was determined that the wheezing was included in the addition unit periods (T1, T2, T3, T4) and to obtain the result as a wheezing period (Tw1, Tw2, Tw3, Tw4), wherein the display processing unit displays the information indicating the temporal change in the frequency of wheezing as a bar chart (AT) indicating percentages occupied by the wheezing time periods (Tw1, Tw2, Tw3, Tw4) in bars of a certain length corresponding to the addition unit time periods (T1, T2, T3, T4), wherein the display processing unit displays a plurality of bars (AT1, AT2, AT3, AT4) corresponding to the addition unit time periods (T1, T2, T3, T4) in parallel alignment on the display screen (10), wherein the addition processing unit classifies, in each of the processing unit time periods, the intensity of the wheezing sound into a plurality of stages based on the area of ​​the dominant peak and adds the lengths of the processing unit time periods in which the wheezing was determined to be included in each of the classified stages, and wherein the display processing unit displays, based on the addition performed by the addition processing unit, the panting time periods (Tw1, Tw2, Tw3, Tw4) divided into the stages in the bars (AT1, AT2, AT3, AT4) corresponding to the addition unit time periods (T1, T2, T3, T4).

The invention discloses a children breath sound multi-level classification system based on AST, and the system comprises a data collection and preprocessing module which collects and preprocesses children breath sound data, and obtains children breath sound features; the coarse classification module is used for extracting global features based on a pre-trained AST encoder and outputting a coarse classification result; the fine-grained feature detection module is used for obtaining fine-grained features by utilizing AttnPool1d which can be learned and scaled based on the global features, outputting logit values of specific pathological features through MLP, executing a hierarchical gating mechanism according to a coarse classification result, and keeping two levels of logic consistent through balanced-medal Focus Loss and hierarchical consistency constraint; and the diagnosis output module is used for comparing the preset threshold value of the specific pathological feature with the logit value and outputting a result indicating whether the specific pathological feature exists or not. The children breath sound multi-level classification precision is effectively improved.

The invention provides a biological audio data processing method, a biological audio data processing device and a medium, and relates to the technical field of data processing, the method can process data in a biological audio data set corresponding to respiration audio of a target organism to obtain an RGB value corresponding to a signal value of each frame of respiration audio signal in the respiration audio; constructing a target image corresponding to the breathing audio based on the RGB numerical value corresponding to the signal value of each frame of breathing audio signal, inputting the target image into the target convolutional neural network model to obtain a filtering threshold combination matrix, and combining the filtering thresholds based on the filtering threshold combination in the filtering threshold combination matrix, and obtaining a target filtering threshold combination corresponding to each frame of breathing audio signal, and carrying out noise reduction processing on the breathing audio based on the dynamic filter and the target filtering threshold combination corresponding to each frame of breathing audio signal to obtain a target audio corresponding to the breathing audio, so that the noise reduction precision can be improved. Therefore, effective signal loss or noise residue can be avoided.

Disclosed is a method for analyzing a heart sound by an electronic device for measuring a preload of a general anesthesia surgery patient including splitting a heart sound and a breathing sound from heart-lung sound data of a patient obtained externally, specifying a breathing section from the split breathing sound, obtaining a heart sound envelope for obtaining a peak signal by converting a heart sound signal, obtaining a heart sound index from the heart sound envelope, and calculating an acoustic variability index (AVI) by analyzing the heart sound index.

The invention discloses an audio deep forgery detection method and device, equipment, a storage medium and a product, and the method comprises the steps: carrying out the detection of to-be-processed audio data through a pre-trained audio deep forgery detection model, and carrying out the detection of the to-be-processed audio data in the detection process of the audio deep forgery detection model, extracting audio frequency features and audio time features of the audio data based on a time-frequency double-branch structure, and enhancing breathing sound clues in the audio time features through a breathing sound feature linear modulation mechanism; performing cross attention fusion on the audio time features and the audio frequency features, and predicting the fused features to obtain a forged audio detection result; according to the method, the breathing sound feature linear modulation mechanism and the cross attention fusion mechanism are utilized, breathing sound clue enhancement and fusion of the breathing sound clues and the time-frequency features can be achieved, natural voice fine clues such as breathing sound are effectively recognized, and therefore the accuracy of audio deep forgery detection is improved.

A breathing sound measurement device includes a measurement section configured to be brought into contact with a skin of a front neck of a subject and measure breathing sound of the subject and a wearing section configured to have a first end connected to the measurement section and have elasticity to extend in a circular arc shape along an outer circumference of the subject's neck. The wearing section has a second end provided with a pressing portion configured to be pressed against a skin of a rear neck opposite a right front neck with respect to a midpoint between right and left halves of the subject's neck.

1. The name of the design product: breathing sound host. 2. The use of the design product: the design product is a breathing sound collecting device, which can realize the collection of breathing sound. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

A system and associated method for automatically identifying a mask (9) used in a pressure support system (2) for delivering a flow of breathing gas to the airway of a patient. The system includes a controller (20) implementing a trained machine learning model (22). The controller is structured and configured to receive a sound signal, the sound signal being indicative of breathing sounds (e.g., exhalation and / or inhalation sounds) captured from the patient during use of the mask in the pressure support system, generate acoustic spectrum data indicative of an acoustic spectrum of the exhalation sounds based on the sound signal, provide the acoustic spectrum data to the trained machine learning model (22), and determine a brand, type and / or size of the mask (9) in the trained machine learning model based on the provided acoustic spectrum data.

The invention provides an intelligent early warning intervention system for chronic obstructive pulmonary diseases, and relates to the technical field of medical instruments. Comprising a core host, the core host is integrated with an active multi-modal sensor arm and an integrated drug delivery arm, a patient can be actively guided to enter a standardized test posture, and high-quality multi-modal physiological data such as lung function, blood oxygen and breath sound are synchronously collected; an intelligent medicine bin and an automatic delivery mechanism are arranged in the intelligent medicine box, and accurate taking, delivery and auxiliary inhalation of medicine can be completed after early warning is triggered. The standardized seat assists in fixing the sitting posture of a patient through a physical guiding structure, and the wearable monitoring vest achieves respiratory rhythm monitoring and auxiliary sputum excretion. According to the system, all the modules are cooperated through the central controller, and a complete entity closed loop from standard monitoring to intelligent early warning to automatic intervention is constructed. The system effectively solves the problems that household lung function monitoring data is unreliable and intervention is difficult to execute after early warning, and the timeliness, safety and patient compliance of chronic disease management are remarkably improved.

The invention discloses an asthma sound denoising method and device based on a dynamic threshold value, and belongs to the technical field of medical signal processing. The method comprises the following steps: performing wavelet decomposition on breathing sound of a user to obtain a plurality of frequency bands; for each frequency band, respectively estimating a noise standard deviation by adopting a median absolute deviation, determining a wavelet coefficient threshold value based on the noise standard deviation, and processing the corresponding frequency band signal according to the wavelet coefficient threshold value; and reconstructing the processed wavelet coefficient to complete asthma noise reduction. According to the method, the threshold intensity is dynamically adjusted in combination with the wavelet coefficient variance, so that over-processing or under-processing caused by a single threshold is avoided, and the accuracy and efficiency of asthma sound denoising are remarkably improved.

To provide a snoring determination system capable of suitably determining the existence of snoring from a voice signal by a person to be measured.SOLUTION: A snoring determination system includes: measurement means (S201) for measuring transition of a noise level or sound pressure level of respiratory sound generated from a sleeping person to be measured; frequency spectrum acquisition means (S202) for acquiring a frequency spectrum of the transition of the noise level or sound pressure level; peak identification means (S203) for identifying a plurality of frequency peaks on the basis of the frequency spectrum; and determination means (S205) for determining the existence of snoring by the person to be measured on the basis of an intensity ratio of a first peak having the highest intensity to a second peak having the next highest intensity of the plurality of frequency peaks.SELECTED DRAWING: Figure 2

The application discloses a kind of asthma detection method and portable asthma detection equipment, the method includes obtaining user breathing sound;Segmentation after de-noising, and into frequency spectrum by short-time Fourier transform and logarithmic transformation conversion;And frequency spectrum is input to neural network model to obtain wheezing and wheeze sound.Detection equipment includes shell, main control unit is built in shell, bottom is embedded with contact microphone connected with main control unit;Wherein, main control unit is used to control contact microphone to collect breathing sound and utilize the asthma detection method described above to identify wheezing and wheeze sound.The detection method of the application is fast in operation, good in real-time, can realize the timely detection to the asthma condition of patient, and has good distinguishing effect;While detection equipment is portable, simple to operate, easy to carry.

The invention discloses a tuberculosis auxiliary diagnosis system fusing chest radiography and breath sound characteristics. The tuberculosis auxiliary diagnosis system comprises a breath sound acquisition module, a breath sound dynamic analysis module, a chest radiography acquisition module, a chest radiography detection module, an activity discrimination fusion module and a result output module. Breathing sound signals are collected at a plurality of standard point positions on the body surface of a patient, dynamic abnormal indexes are extracted, meanwhile, chest radiography images are obtained, focus attribute information is recognized, an activity judgment fusion module conducts consistency analysis on the chest radiography images and the dynamic abnormal indexes, standardized risk scores are generated, and high-risk, low-risk or uncertain grading results are output. According to the method, the misdiagnosis and missed diagnosis risks can be effectively reduced under the complex conditions of old focus combined activities, early hidden cases, combined chronic obstructive pulmonary disease and the like, and a visual diagnosis report containing a chest radiograph thermodynamic diagram and a breath sound frequency spectrum is provided. Compared with single chest radiography or single breath sound analysis, the sensitivity and specificity of active tuberculosis diagnosis are improved, and the method has high clinical application value.

The invention belongs to the field of aviation technology design, and particularly relates to a pilot fast-wearing type oxygen mask microphone communication noise reduction method and a system thereof.The pilot fast-wearing type oxygen mask microphone communication noise reduction method comprises the steps of sound source monitoring, specifically, sound in a closed cavity of a fast-wearing type oxygen mask is continuously collected through a microphone; a voice processing step: judging whether the sound in the closed cavity of the quick-wearing oxygen mask is breathing sound or not in real time; the amplitude, the frequency, the period and the unvoiced and voiced sound distribution characteristics are synthesized to judge whether the sound in the closed cavity of the fast-wearing oxygen mask is breath sound or not; a logic control step: when the sound in the closed cavity of the quick-wearing oxygen mask is judged to be breathing sound, switching off microphone communication; and when the sound in the closed cavity of the quick-wearing oxygen mask is judged not to be breathing sound, microphone communication is turned on.

The embodiment of the invention discloses an owner identification method and terminal equipment, which are suitable for the technical field of computers, and the method comprises the following steps: the terminal equipment responds to an unlocking operation of a to-be-tested user on target content, and collects user data of the to-be-tested user, the target content is in a locked state, and the target content is in a non-locked state; the user data comprises at least one of the following items: a to-be-detected face image and to-be-detected breathing audio data; according to the user data and a pre-recorded owner template, owner identification is carried out, and the owner template comprises at least one of the following items: a face template and a breath sound feature template; and under the condition that the owner is identified, unlocking the target content. According to the embodiment of the invention, the owner identification can be carried out based on the data of various modes in the process that the terminal equipment is used, so that the security of private information of the owner is protected.

The invention provides a vehicle driving risk determination method and device, electronic equipment and a storage medium, and the method comprises the steps: monitoring a sound signal of a driver in a current vehicle state, carrying out the feature extraction of a breath sound signal and a sound ray signal, obtaining the breath feature of the breath sound signal and the acoustic feature of the sound ray signal, and determining the driving risk of the driver; judging the current driving state of the driver by adopting the breathing characteristics and the acoustic characteristics, determining whether the driver is in an abnormal driving state, and when the driver is determined to be in the abnormal driving state, determining the driving risk level of the driver according to the abnormal driving state, or determining the driving risk level of the driver according to the abnormal driving state and the current vehicle state. And determining a driving risk level of the driver. According to the invention, non-contact monitoring and multi-dimensional identification are carried out on the sound of the driver in the current vehicle state, the driving state of the driver is judged, and the accuracy of determining the vehicle driving risk is improved through the association of the abnormal driving state and the vehicle state and the independent triggering of the abnormal driving state.