Vibration wave processing method and device, and storage medium

Abstract

The invention provides a vibration wave processing method and device, and a storage medium, and belongs to the technical field of electronic information technology and artificial intelligence, and the method comprises the steps: collecting vibration wave signals of a full frequency domain and / or multiple modes in a preset space range, and obtaining a to-be-processed vibration wave signal; according to a preset feature extraction model and the to-be-processed vibration wave signals, obtaining audio feature data in one-to-one correspondence with the to-be-processed vibration wave signals; determining an interference signal according to the audio feature data; and carrying out reduction or suppression processing on the interference signal. According to the method, the interference source influencing the human body perception can be determined from the actual environment, at the moment, targeted real-time online noise reduction processing can be realized without knowing the condition of the use environment in advance, and the effect of eliminating the environmental noise sensed by the human body is improved.

Description

Technical Field

[0001] The present application relates to, but is not limited to, the fields of electronic information technology and artificial intelligence technology, and in particular to a vibration wave processing method, device, and storage medium. Background Art

[0002] With the development of science and technology, people's requirements for the quality of life are constantly improving, among which the comfort of the sound environment is an important aspect. The sources of sound perceived by the human body include not only sound waves in the conventional auditory frequency range that can usually be heard, but also sound waves in some unconventional auditory frequency ranges, and may also include some electromagnetic waves or waves of other modes. The abnormal sounds may have a certain impact on the human body, such as causing physical discomfort or interfering with normal sleep. Therefore, how to effectively reduce the interference of abnormal sounds in the environment and improve the physical comfort of the human body is an urgent technical problem to be solved. In the existing technology, the noise processing technology for this type of environmental noise is mainly through physical methods, such as using sound insulation materials, sound absorbing materials, etc. to reduce the propagation of noise. In addition, there are some active reduction technologies that generate signals opposite to the noise in the device based on the signal parameters of the known noise, thereby achieving a cancellation effect. However, existing noise processing technologies have the following main problems: First, although traditional physical methods can reduce some noise, they are limited by the materials themselves and the construction methods, resulting in limited noise reduction effects of physical methods. In addition, sound insulation materials may cause secondary pollution to the environment during the production process or discarded sound insulation materials. Secondly, existing active reduction technologies often require prior knowledge of the characteristics of the noise in order to identify the noise and effectively reduce it. However, the above two methods cannot significantly improve the effect of reducing environmental noise perceived by the human body. Summary of the Invention

[0003] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a vibration wave processing method, device, and storage medium that can improve the effect of eliminating ambient noise perceived by the human body.

[0004] In a first aspect, an embodiment of the present application provides a vibration wave processing method, the method comprising: Collecting vibration wave signals in the full frequency domain and / or multiple modes within a preset spatial range to obtain vibration wave signals to be processed; Obtaining audio feature data corresponding to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed; determining an interference signal according to the audio feature data; The interference signal is reduced or suppressed.

[0005] According to some embodiments of the first aspect of the present application, obtaining audio feature data corresponding one-to-one to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed includes: performing digital conversion and audio conversion on the vibration wave signal to be processed in sequence to obtain an audio signal corresponding to the vibration wave signal to be processed; the vibration wave signal to be processed includes at least one of an acoustic wave signal and an electromagnetic wave signal; The audio feature data is obtained by performing feature extraction on the audio signal using the feature extraction model.

[0006] According to some embodiments of the first aspect of the present application, determining the interference signal according to the audio feature data includes: The interference signal is determined according to the frequency feature and the sound feature, wherein the frequency feature and the sound feature are both one of the data in the audio feature data.

[0007] According to some embodiments of the first aspect of the present application, determining the interference signal according to the frequency characteristics and the sound characteristics includes: Grouping the frequency characteristics and sound characteristic data corresponding to the plurality of vibration wave signals to be processed based on a preset clustering algorithm to obtain a plurality of clustering characteristic data; Determining interference clustering feature data from the plurality of clustering feature data according to preset reference feature data; The vibration wave signals corresponding to the interference clustering feature data are all used as the interference signals.

[0008] According to some embodiments of the first aspect of the present application, the reducing or suppressing the interference signal includes: Acquiring signal parameters of the interference signal and generating an adjustment wave based on the signal parameters; Dynamically and in real time locate the vibration wave source position of the interference signal to obtain the vibration wave source position; The adjustment wave is emitted toward the source position of the vibration wave.

[0009] According to some embodiments of the first aspect of the present application, dynamically and in real time locating the vibration wave source position of the interference signal to obtain the vibration wave source position includes: Determining, based on the mutual correlation between the preset collection points at the multiple different locations, a relative time delay for the collection points at the multiple different locations to receive the interference signal at the same location; The position is solved according to the relative time delay and a preset distance equation difference formula to determine the position of the vibration wave source.

[0010] In a second aspect, an embodiment of the present application provides a vibration wave processing device, comprising: A vibration wave signal acquisition module is used to acquire vibration wave signals in the full frequency domain and / or multiple modes within a preset spatial range to obtain vibration wave signals to be processed; a decision analysis and processing module, configured to obtain audio feature data corresponding to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed; and to determine an interference signal based on the audio feature data; The interference processing module is used to reduce or suppress the interference signal.

[0011] In a third aspect, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the vibration wave processing method as described in the first aspect above.

[0012] The vibration wave processing method, device, and storage medium according to the embodiments of the present application have at least the following beneficial effects: by collecting vibration wave signals in the full frequency domain and / or multiple modes to obtain the vibration wave signal to be processed, and by using a feature extraction model to obtain audio feature data representing the characteristics of the vibration wave signal to be processed, and based on the audio feature data, the interference signal is determined, thereby being able to determine the interference source that affects human perception in the actual environment. In this case, targeted real-time online noise reduction processing can be achieved without prior knowledge of the usage environment. Therefore, compared with related technologies, the embodiments of the present application can improve the effect of eliminating environmental noise perceived by the human body. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a schematic diagram of a module of a vibration wave processing device provided by one embodiment of the present application; Figure 2 This is a diagram illustrating the module operation interaction of a vibration wave processing device provided by one embodiment of the present application; Figure 3 This is a schematic diagram of a module for collecting acoustic wave signals in a vibration wave processing device provided in one embodiment of the present application; Figure 4 This is a flow chart of a vibration wave processing method provided by one embodiment of the present application; Figure 5 This is a schematic diagram of the processing process of the vibration wave processing method provided by one embodiment of the present application; Figure 6 This is a hardware structure diagram of a vibration wave processing device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0014] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application.

[0015] In the description of this application, "above," "below," and "within" are understood to be exclusive of the number indicated, while "above," "below," and "within" are understood to be inclusive of the number indicated. The use of "first" and "second" is solely for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly specifying the number or order of the indicated technical features.

[0016] In the description of this application, unless otherwise clearly defined, terms such as setting, installing, and connecting should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in this application based on the specific content of the technical solution.

[0017] With the development of science and technology, people's requirements for the quality of life are constantly improving, among which the comfort of the sound environment is an important aspect. The sources of sound perceived by the human body include not only sound waves in the conventional auditory frequency range that can usually be heard, but also some sound waves in the unconventional auditory frequency range, and may also include some electromagnetic waves. The abnormal sound waves may have a certain impact on the human body, such as causing physical discomfort or interfering with normal sleep. Therefore, how to effectively reduce the interference of abnormal sounds in the environment and improve the physical comfort of the human body is an urgent technical problem to be solved. In the existing technology, the noise processing technology for this type of environmental noise is mainly through physical methods, such as using sound insulation materials, sound absorbing materials, etc. to reduce the spread of noise. In addition, there are some technologies that actively generate signals opposite to the noise to achieve a canceling effect. However, existing noise treatment technologies have the following major problems: First, while traditional physical methods can reduce some noise, they are limited by the inherent limitations of the materials and construction methods, resulting in limited noise reduction effectiveness. Furthermore, the production process or discarded sound insulation materials may cause secondary pollution to the environment. Second, existing active noise reduction technologies often require prior knowledge of noise characteristics in order to identify and effectively reduce noise. Furthermore, their effectiveness is limited to pre-production sample testing and cannot be adapted to individual differences. Furthermore, existing active noise reduction technologies are primarily used in industry and headphones, and do not directly affect the environment or the human body. Therefore, the noise reduction effect on environmental noise perceived by the human body is not significant.

[0018] Based on this, the embodiments of the present application propose a vibration wave processing method, device and storage medium, which can improve the noise reduction effect of environmental noise perceived by the human body.

[0019] Based on this, refer to Figure 1 As shown, an embodiment of the present application provides a vibration wave processing device, the device comprising: The vibration wave signal acquisition module 100 is used to acquire vibration wave signals in the full frequency domain and / or multiple modes within a preset spatial range to obtain vibration wave signals to be processed; the spatial range is determined by the structure of the vibration wave signal acquisition module 100; The decision analysis and processing module 300 is used to obtain audio feature data corresponding to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed; and to determine the interference signal based on the audio feature data; The interference processing module 400 is used to reduce or suppress interference signals.

[0020] Therefore, the vibration wave signal to be processed is obtained by collecting vibration wave signals in the full frequency domain and / or multiple modes, and audio feature data characterizing the characteristics of the vibration wave signal to be processed is obtained through a feature extraction model, and the interference signal is determined based on the audio feature data, so that the interference source that affects human perception can be determined from the actual environment. At this time, targeted real-time online noise reduction processing can be achieved without knowing the usage environment in advance.

[0021] In some embodiments, the decision-analysis and processing module 300 and the vibration wave signal acquisition module 100 are integrated into a single terminal device. In other embodiments, the decision-analysis and processing module 300 is located on a remote server (such as a cloud server), while the interference processing module 400 and the vibration wave signal acquisition module 100 can be integrated into the same terminal device. In either embodiment, the terminal device can be moved with the target object or fixed in place.

[0022] The spatial range corresponding to the vibration wave signal acquisition module 100 can be set according to the application scenario. The spatial range of the vibration wave signal acquisition module 100 can be set dynamically or pre-configured. In this regard, the embodiment of the present application does not limit how the spatial range of the vibration wave signal acquisition module 100 is set.

[0023] By collecting vibration wave signals across the entire frequency domain, vibration wave signals in all frequency bands in the environment can be monitored. By collecting vibration wave signals in multiple modes, signals of all modes that cause interference in the environment can be collected. Therefore, the signal collection of the embodiments of the present application can further reduce the probability of missing interference sources from different dimensions. Furthermore, by collecting vibration wave signals in real time and performing feature extraction based on a feature extraction model, real-time online interference source identification can be achieved. Even when the source of environmental interference is unclear, interference signals can be accurately identified, making noise cancellation applicable to a wider range of environments.

[0024] The embodiments of the present application do not limit the modes or number of modes of vibration wave signals that can be collected. Each mode has a different propagation mode, and a combination of modes can be configured based on the propagation mode of interference sources in the environment. For example, if the environment contains sound waves, electromagnetic waves, and other waves, sound waves and electromagnetic waves can be used as the modes to be collected in the embodiments of the present application.

[0025] In some embodiments, sound waves of the entire frequency domain can be collected. In other embodiments, electromagnetic waves of the entire frequency domain can be collected. In other embodiments, both sound waves and electromagnetic waves of the entire frequency domain can be collected. In other embodiments, electromagnetic waves of a specific frequency domain and sound waves of the entire frequency domain can also be collected. The embodiments of the present application do not go into detail about this. In some embodiments, dynamic configuration can be performed on the vibration wave processing device to select the supported modes and frequency domains of the collected vibration waves. In other embodiments, fixed settings can also be used. The embodiments of the present application do not impose too many restrictions on this. Those skilled in the art can selectively set it according to actual conditions.

[0026] The feature extraction model is used to extract audio features. The audio feature data includes frequency and sound features, where the sound features represent the category of the audio. The feature extraction model is a pre-trained network model. The feature extraction model can be trained using a machine learning algorithm or a deep learning algorithm.

[0027] In some embodiments, reference Figure 2 As shown, the vibration wave processing device also includes an acoustic signal conversion module 211, an electromagnetic signal demodulation and conversion module 212, and a digital signal conversion module 220. The acoustic signal conversion module 211 is used to convert acoustic signals into digital signals, the electromagnetic signal demodulation and conversion module 212 is used to convert electromagnetic signals into digital signals, and the digital signal conversion module 220 is used to convert digital signals into audio signals. The acoustic signal conversion module 211, the electromagnetic signal demodulation and conversion module 212, and the digital signal conversion module 220 enable normalization processing of the vibration wave signal acquisition module 100, allowing the decision analysis and processing module 300 to directly extract features and determine interference signals based on the audio signal, resulting in higher adaptability and a simpler processing flow.

[0028] In some embodiments, reference Figure 2As shown, the interference processing module 400 includes an adjustment sound wave generation module 410 and an adjustment electromagnetic wave generation module 420. The adjustment sound wave generation module 410 is used to generate a cancellation signal for the sound wave signal that reduces or suppresses interference when the interference signal is a sound wave signal. The adjustment electromagnetic wave generation module 420 is used to generate a cancellation signal for the electromagnetic wave signal that reduces or suppresses interference when the interference signal is an electromagnetic wave signal.

[0029] In some embodiments, reference Figure 2 As shown, the vibration wave processing device also includes a dynamic wave source positioning module 430 to dynamically and accurately locate the interference source in real time when the noise source and the vibration wave signal acquisition module 100 move relative to each other, thereby improving the effect of mutual cancellation between the generated cancellation signal and the noise.

[0030] In some embodiments, the vibration wave processing device further includes an evaluation and optimization module configured to determine the noise cancellation effect of the interference signal based on the brainwave signal and, based on the noise cancellation effect, to select whether to optimize the interference condition, thereby re-determining and processing the interference signal based on the optimized interference condition. In some embodiments, the brainwave signal may be acquired by the brainwave signal acquisition module.

[0031] It is understandable that referring to Figure 4 As shown, according to an embodiment of the present application, a vibration wave processing method is provided, the method comprising: Step S100: collecting vibration wave signals of the entire frequency domain and / or multiple modes within a preset spatial range to obtain vibration wave signals to be processed; Step S200: obtaining audio feature data corresponding to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed; Step S300: determining an interference signal based on audio feature data; Step S400: Eliminate or suppress the interference signal.

[0032] Therefore, by collecting vibration wave signals across the entire frequency domain and / or multiple modes to obtain the vibration wave signals to be processed, and using a feature extraction model to obtain audio feature data representing the characteristics of the vibration wave signals to be processed, and based on the interference signals determined based on the audio feature data, it is possible to determine the interference sources that affect human perception in the actual environment. In this case, targeted real-time online noise reduction processing can be achieved without prior knowledge of the usage environment. Therefore, compared with related technologies, the embodiments of the present application can improve the effectiveness of eliminating environmental noise perceived by the human body.

[0033] The full frequency domain in step S100 means that there will be no frequency restriction on the collected signal. As long as the signal is generated within the spatial range and meets the mode defined by the collected signal, it will be collected. The multiple modes in step S100 mean that a set number of different types of propagating vibration wave signals are collected. In some embodiments, the device used in the method of the embodiment of the present application supports the collection of vibration wave signals in the full frequency domain and multiple modes. In actual applications, the frequency domain and mode of the collected vibration wave signal can be determined by configuring the corresponding collection items, thereby achieving efficient and low-power processing in personalized scenarios. The spatial range supported in step S100 is determined by the vibration wave signal collection module 100 provided on the device to which the vibration wave processing method is applied.

[0034] Each vibration wave signal to be processed has a corresponding signal parameter (e.g., frequency, amplitude, phase, and wavelength for acoustic signals, and frequency, amplitude, phase, and wavelength for electromagnetic signals). The present embodiment of the application does not limit how the interference signal is determined based on the audio feature data in step S300. For example, in some embodiments, the audio feature data is compared with feature data of a known noise signal. In other embodiments, the audio feature data may be classified, and the classification results may be compared with preconfigured interference conditions to determine the interference signal.

[0035] The embodiments of this application do not restrict how steps S100-S400 are deployed. For example, in some embodiments, the vibration wave processing method is applied to an independent vibration wave processing device, and steps S100-400 are all executed on the vibration wave processing device. In other embodiments, the vibration wave processing device includes an intelligent terminal and a server, and steps S100-S400 are deployed on the server. In this regard, the embodiments of this application can select one or both configuration methods based on actual circumstances, thereby enabling real-time calculations to be performed by the server when a network is available, and by the intelligent terminal itself when an network is not available.

[0036] For example, taking the acquisition of full-frequency vibration wave signals of multiple modes (i.e., propagation modes) as an example, the specific steps are as follows: Step 1: collecting vibration wave signals of the full frequency domain and multiple modes within the target range to obtain multiple vibration wave signals to be processed; each vibration wave signal to be processed has different signal parameters; Step 2: converting each vibration wave signal to be processed into a digital signal in the manner of a digital signal corresponding to each mode, and converting the digital signal into an audio signal; the audio signal corresponds to the vibration wave signal to be processed one by one; Step 3: extracting features from the signal parameters of the audio signal, classifying, numbering, and clustering them to form at least one space vector group, wherein the same space vector group corresponds to at least one vibration wave signal, and the audio feature data corresponding to the vibration wave signal are similar (or the similarity meets the clustering requirements); In some embodiments, the audio feature data includes a frequency feature f and a sound feature s, wherein the classification and numbering means performing supervised classification on the frequency feature f and the sound feature s to associate the frequency feature f and the sound feature s of the same vibration wave signal, and numbering each group (f, s) after classification and then marking them in the initial feature matrix. The initial feature matrix is as follows: ; Among them, (f1,s1) represents a set of categories.

[0037] Among them, clustering means using a clustering algorithm (such as K-Means) to group the two-dimensional vectors formed by the classification of each row of the initial feature matrix, and using the cluster center of each cluster group as the representative vector group of the corresponding cluster group, then a spatial vector group can be obtained: ; Among them, (Cf1, Cs1) represents the representative vector group of a cluster group.

[0038] Step 4: Determine whether each vibration wave signal needs to be attenuated based on the space vector group; If so, it means that the vibration wave signal is an interference signal and the intelligent control program needs to be triggered. At this time, an adjustment wave can be generated according to the signal parameters corresponding to the interference signal; when there are multiple interference signals, multiple adjustment waves are generated.

[0039] In some embodiments, interference signals can be determined by setting preset baseline characteristic data. In practical applications, the preset baseline characteristic data or interference conditions can be selected and changed by the intelligent controller based on actual application conditions, including individual differences, preferences, and user experience, so that the determined interference signals are more individualized.

[0040] Step 5: The adjustment wave is accurately transmitted according to the preset dynamic wave source positioning algorithm.

[0041] In summary, the embodiments of the present application can collect full-frequency domain and multi-modal vibration wave signals within a preset range, not only to monitor sound waves within the conventional auditory frequency range, but also to effectively monitor sound waves within the unconventional auditory frequency range, and also to effectively monitor some electromagnetic waves that can be converted into audio signals, thereby achieving comprehensive monitoring of environmental noise, more comprehensively reducing the interference of abnormal noise perceived by the human body, and improving physical comfort.

[0042] The embodiment of the present application can realize online real-time detection of environmental noise through steps S100~S400, and automatically identify interference signals. In addition, the embodiment of the present application does not require prior knowledge of the characteristics of the interference source in the environment, which greatly improves the processing integrity, efficiency and accuracy.

[0043] In some embodiments, embodiments of the present application may also incorporate brainwave signals. Vibration wave signals that cause physical discomfort can be accurately identified through the correlation between brainwave signals and ambient noise, and then adjustment waves can be generated to actively reduce or suppress them, thereby achieving targeted noise reduction. In some embodiments, the user can input a configuration item to select whether to perform noise reduction optimization, thereby optimizing processing in conjunction with brainwave signals when noise reduction optimization is supported.

[0044] Therefore, the embodiment of the present application can effectively reduce the all-round sound interference perceived by the human body and improve the physical comfort. Therefore, it can be widely used in various occasions where sound interference needs to be reduced.

[0045] When collecting full-frequency and / or multi-modal vibration wave signals, the collected vibration wave signals to be processed are signals of unfiltered frequency bands, so that vibration wave signals of the full frequency band can be collected.

[0046] It is understandable that the vibration wave signals of the full frequency domain and / or multiple modes within the preset spatial range are collected to obtain the vibration wave signals to be processed, including: The vibration wave signal to be processed is sequentially digitally converted and audio-converted to obtain an audio signal corresponding one-to-one to the vibration wave signal to be processed; the vibration wave signal to be processed includes at least one of an acoustic wave signal and an electromagnetic wave signal; and feature extraction is performed on the audio signal through a feature extraction model to obtain audio feature data.

[0047] By collecting vibration wave signals of multiple modes, it is possible to collect vibration waves perceived by the human hearing mechanism, vibration waves perceived by the body surface that cannot be converted into hearing mechanism, and vibration waves perceived by the body surface that can be converted into hearing mechanism.

[0048] The embodiments of this application do not limit the specific methods of digital conversion and audio conversion. For example, the conversion can be performed using electronic components or by configuring algorithms in combination with hardware. Different types of signals can also use different conversion methods. In this case, the collected vibration wave signal can be converted into a digital signal according to the conversion method corresponding to the different types; and the digital signal can be converted into an audio signal.

[0049] For example, taking a set of acoustic wave signals S as an example, the acoustic wave signal set S is converted into digital signals and then into audio signals, resulting in S' = (S1, S2, ...Si), (n=1, 2...n), where Si represents the audio signal corresponding to the i-th collected vibration wave signal. Each audio signal Si is component-wise resolved to obtain the corresponding audio component. Audio features can then be extracted based on the audio signal components. The extracted audio features include at least one of time domain features, frequency domain features, and time-frequency domain features. In some embodiments, time domain features include short-time energy, zero crossing rate (ZCR), and autocorrelation coefficient function; frequency domain features include frequency, Mel-frequency cepstral coefficients (MFCCs), linear prediction coefficients (LPCs), spectral centroid / bandwidth, and chrominance features; and time-frequency domain features include short-time Fourier transforms (SFTs), Mel-frequency spectrograms (MSPs), and wavelet transforms. The time, frequency, and time-frequency domain features extracted from the acoustic wave signals can be used to determine sound characteristics, thereby identifying the sound category. The sound characteristics serve as audio feature data for interference signal identification.

[0050] For example, taking electromagnetic wave signal set D as an example, D is demodulated into digital signals and converted into audio signals, resulting in D' = (D1, D2…Dj), (j=1, 2…m), where Dj represents the audio signal corresponding to the jth collected vibration wave signal. Each audio signal Di is component-wise resolved to obtain the corresponding audio component. Audio features can then be extracted based on the components of the audio signal, yielding features such as time domain features (including at least one of short-time energy, zero crossing rate (ZCR), and autocorrelation coefficient function); frequency domain features (including at least one of frequency, Mel-frequency cepstral coefficients (MFCCs), linear prediction coefficients (LPCs), spectral centroid / bandwidth, and chrominance); or time-frequency domain features (including at least one of short-time Fourier transform (SFT), Mel-frequency spectrogram, and wavelet transform). The time, frequency, and time-frequency domain features extracted from the electromagnetic waves can be used to determine sound characteristics, thereby identifying the sound category. The sound features are then used as audio feature data for interference signal identification.

[0051] By digitally converting the vibration wave signal to be processed into a digital signal, and then converting the digital signal into an audio signal, it is possible to extract features from non-sound wave signals through a unified audio feature model.

[0052] It is understandable that determining the interference signal based on the audio feature data includes: An interference signal is determined according to the frequency feature and the sound feature, wherein the frequency feature and the sound feature are both one of the data in the audio feature data.

[0053] By determining interference signals based on frequency characteristics and sound characteristics, refined signal management can be achieved.

[0054] In some embodiments, the frequency classification corresponding to the frequency feature can be pre-configured or dynamically configured, so that the frequency feature can be first classified based on the configured frequency classification, so that the pre-set baseline sound feature data can be compared with the real-time extracted sound feature under different frequency classifications to determine the interference signal. In some embodiments, the frequency classification is set to at least three categories. The first classification represents a noise signal in the normal hearing range, with a frequency band within 20HZ-20kHZ; that is, the frequency f1Si satisfies 20HZ < f1Si < 20kHZ; the second classification represents a noise signal in the abnormal hearing range, with a frequency band outside 20HZ-20kHZ; that is, the frequency f2Si satisfies f2Si < 20HZ or f2Si > 20kHZ; the third classification represents an electromagnetic wave noise signal that can be converted into an interfering sound signal. At this time, the frequency classification information is determined by matching the extracted frequency features with the above three categories respectively. For example, for a sound wave signal with a frequency of 40HZ, the classification information is the first classification.

[0055] Sound features characterize sound categories, such as human voices, animal sounds, music, noise, natural sounds, mechanical operation sounds, and so on. By identifying interference signals based on frequency features and sound features, it is possible to identify interference sources from multiple dimensions and improve the accuracy of interference signal identification. In some embodiments, different baseline sound feature data can be set under different frequency classification information as the judgment of interference signals, or baseline sound feature data under different frequency classification information can be set according to individual differences, preferences, and experiences, so that interference signals can be determined based on the matching results of sound features and baseline feature data under preset sound categories, such as correlation. For example, under the first classification information, noise can be set as the main basis for judging interference signals; under the second and third classification information, human voices can be set as the main basis for judging interference signals.

[0056] Among them, the sound characteristics of the vibration wave signal can be dynamically determined by the extracted audio features and the classifier model, wherein the audio features are obtained by extracting audio features based on the components of the audio signal to obtain at least one of time domain features, frequency domain features and time-frequency domain features, wherein the time domain features include at least one of short-time energy, zero crossing rate ZCR, and autocorrelation coefficient function; the frequency domain features include at least one of frequency, Mel frequency cepstral coefficient MFCC, linear prediction coefficient LPC, spectrum centroid / bandwidth, and chroma features; the time-frequency domain features include at least one of short-time Fourier transform, Mel frequency spectrum map, and wavelet transform.

[0057] Among them, the classifier model is based on machine learning and deep learning algorithms for classification. The classifier model can be one of the following: MFCC+SVM / Random Forest, Mel-spectrogram+CNN / transformer, LSTM / CRNN, lightweight CNN, etc.; the audio features extracted above are classified by the classifier model, so that the category to which the vibration wave signal belongs can be obtained. Since the sound categories included in the audio signal obtained from the vibration wave signal in the target range are not fixed, in some embodiments, different classifier models and extracted feature combinations can be dynamically configured to achieve accurate classification and improve the effect of interference signal prediction and classification. For example, multiple classifier models are pre-set, and the weights of individual audio features and combined audio features in each classifier model are different. Therefore, accurate classification can be achieved by selecting a classifier model that is adapted to the current application environment and selecting a suitable audio feature combination as the input of the classifier model.

[0058] Among them, the sound categories under different frequency categories may have at least partial overlap, and the same sound category may cause different effects under different frequency categories. Therefore, more refined recognition of vibration wave signals can be achieved, thereby improving recognition accuracy.

[0059] It is understandable that the interference signal is determined based on the frequency characteristics and sound characteristics, including: The frequency characteristics and sound characteristic data corresponding to the plurality of vibration wave signals to be processed are grouped based on a preset clustering algorithm to obtain a plurality of clustering characteristic data; Determining interference clustering feature data from a plurality of clustering feature data according to preset benchmark feature data; The vibration wave signals corresponding to the interference clustering feature data are all regarded as interference signals.

[0060] The embodiment of the present application does not limit the clustering algorithm used for clustering processing, such as the GMM clustering algorithm or the K-means clustering algorithm.

[0061] The cluster feature data is audio feature data corresponding to the cluster center of the corresponding cluster group.

[0062] The clustering algorithm can be used to divide audio feature data with similar characteristics into the same group. At this time, when the cluster feature data is matched based on the personalized interference type, the vibration wave signal corresponding to the matching result covers more comprehensively, thereby realizing personalized noise cancellation processing.

[0063] The reference feature data includes reference frequency feature data and reference sound feature data. Based on the frequency feature, the corresponding frequency classification is found from the reference frequency feature data in the reference feature data. By comparing the sound feature with the reference sound feature, the interference signal can be determined.

[0064] It is understood that the interference signal reduction or suppression processing includes: obtaining signal parameters of the interference signal and generating an adjustment wave based on the signal parameters; Perform dynamic real-time positioning of the vibration wave source of the interference signal to obtain the vibration wave source position; The adjustment wave is emitted toward the vibration wave source. By dynamically updating the position of the interference signal in real time, the noise cancellation effect can be ensured even in scenarios where there is relative displacement between the vibration wave processing device and the interference signal. If the vibration wave processing device is portable, the perceived comfort of the human body can still be ensured even when the human body is moving. If the interference signal is mobile, real-time positioning can ensure that the interference signal can be offset or suppressed.

[0065] It is understandable that the vibration wave source position of the interference signal is dynamically located in real time to obtain the vibration wave source position, including: Based on the mutual correlation between the preset multiple collection points at different locations, determining the relative time delay of the multiple collection points at different locations receiving the interference signal at the same location; The position is solved based on the relative time delay and the preset distance equation difference formula to determine the position of the vibration wave source.

[0066] There are multiple collection points for collecting the same modal signal. The embodiments of the present application do not limit how the collection points are set. For example, for sound wave signals, microphones are used for collection, and the collection points are each microphone in the microphone array. For example, if an antenna is used for signal collection of electromagnetic wave signals, an antenna array can be set up for multi-point collection. At this time, since the interference signal is in the same position for each collection point at the same time, and the signal is also the same, the time difference of the same characteristic signal collected by each collection point can be used, so that the real-time position of the interference signal can be determined based on the time difference. In some embodiments, a vibration wave source dynamic positioning algorithm based on time arrival difference TDOA can be used to estimate the vibration wave source position by analyzing the time delay difference of the vibration wave signal arriving at different collection points (such as sensors); if there is relative motion between the collection point and the signal source, a dynamic model is introduced to calculate the vibration wave source position.

[0067] The principle of TODA is as follows: Assuming that the collection point is a sensor array, and the sensor array includes N sensors, the positions of N sensors can be expressed as: ,in, 3....,N. The position of the signal source s is expressed as: ;in, Represent the coordinates of the signal source s on the X-axis, Y-axis, and Z-axis respectively, Represent the coordinates of the i-th sensor on the X-axis, Y-axis, and Z-axis respectively. At this time, the arrival time difference between sensor i and reference sensor j is: ; The distance difference equation is constructed based on the arrival time difference: ; Where c is the propagation speed of the vibration wave in the medium (including the speed of sound waves, the speed of light waves, etc.). In some embodiments, when there is and / or , represents the relative displacement between sensor i and the signal source, represents the relative displacement between the reference sensor j and the signal source, the position of sensor i can be corrected. At this time, the position of sensor i is: ,or ,or ; At this time, the corrected distance difference equation is At this time, the position of the vibration wave source can be calculated based on the distance difference equation.

[0068] For example, the calculation of the vibration wave source position based on the TODA principle is as follows: Step 1: Time Delay Estimation (TDE): Using the cross-correlation method, calculate the cross-correlation function of the sensor i and j signals , determine the relative delay by the peak position of the cross-correlation function ; t is the time, T is the time delay, dt is the time difference between the time ti and tj when sensors i and j receive the same vibration signal; the specific formula is as follows: ; ; in, Indicates that the corresponding cross-correlation function is maximized. (T) represents the cross-correlation function between sensor i and reference sensor j. The above formula reflects the temporal similarity of the two signals. Ideally, the cross-correlation function will have a peak at the actual delay position. Therefore, the relative delay can be determined based on the position corresponding to the peak. .

[0069] Then, based on the generalized cross-correlation GCC-PHAT, the signal is weighted to improve the delay resolution. The specific formula is as follows: = ; in, represents the time domain signal of sensor i and the time domain signal of reference sensor j The generalized cross-correlation function of is the inverse Fourier transform, and is a time domain signal and The Fourier transform of yes The complex conjugate of represents the cross power spectrum, represents the amplitude of the cross-power spectrum, which is used to normalize the phase information, and T represents the delay variable.

[0070] At this time, through The peak value of the delay is determined, and the weighting function can suppress the influence of noise and reverberation, thereby improving the accuracy of the delay estimation.

[0071] Step 2: Position solution: Convert the distance difference equation into an optimization problem and solve it to minimize the residual error. The formula is as follows: ; in, is the wave source position coordinate, is the known position coordinate of the i-th sensor at time t, is the known position coordinate of the j-th sensor at time t, and ||·|| represents the Euclidean distance.

[0072] Step 3: Relative displacement compensation: If the sensor position is known, such as through IMU or motion encoder measurement, the displacement can be obtained , directly brought in If the sensor position is unknown, such as when relative displacement occurs, it is necessary to jointly estimate the source position and displacement. For example, an extended Kalman filter (EKF) can be used to determine the sensor position and substitute it into the above-mentioned residual minimization formula to solve the position.

[0073] It is understandable that, after the interference signal is reduced or suppressed, the method further includes: Acquire brain wave signals within a preset spatial distance; The brainwave signals are provided in plurality and the plurality of brainwave signals are brainwave signals collected within a preset first time period; Optimize the preset interference conditions based on brain wave signals; The optimization of the interference condition includes the optimization of at least one classification condition of frequency classification and sound classification, and the interference condition is used for determining the interference signal.

[0074] The embodiments of the present application do not restrict the method for acquiring brainwave signals, nor do they restrict how to optimize interference conditions based on brainwave signals. In some embodiments, whether to optimize interference conditions can be determined directly based on the similarity between the brainwave signal and the reference brainwave signal. In other embodiments, features can be extracted from the brainwave signal, and based on the extracted features, it can be determined whether negative emotions caused by interference sources exist. When negative emotions exist, the interference conditions can be optimized. In this case, the problem of inaccurate interference condition settings caused by individual differences and / or dynamic changes in the collected vibration wave signals can be further eliminated, thereby improving the accuracy of interference signal identification.

[0075] The optimization of frequency classification may include, for example, optimizing the threshold range of frequency classification, the number of frequency classification categories, etc., to achieve noise elimination based on individual differences. For example, in the above embodiment, if the audio feature data is classified based on three frequency classifications, the category range of each frequency classification can be adjusted (such as the 20HZ~20kHZ of the first frequency classification can be adjusted to 10HZ~20kHZ), so that more accurate classification can be achieved during classification, thereby ensuring that the progress of interference signal identification under the corresponding classification meets the requirements. This application does not impose too many restrictions on the specific adjustment, and can provide relevant algorithms for predicting threshold steps, or can repeatedly make small step adjustments. In some embodiments, the optimization of sound classification may include the type optimization of the classifier model and the optimization of the audio feature combination. For example, based on the brain wave signal, it is judged that the vibration wave signal that will currently cause negative effects is more prominent in the time domain (the prominent audio features can be identified by pre-setting the characteristic tendencies of different sound categories). Then, other time domain features can be added to the original time domain features. For example, the original feature that only includes short-time energy can be modified to include short-time energy and zero-crossing rate ZCR or autocorrelation coefficient function to determine the sound category.

[0076] After the interference conditions are optimized, the vibration wave signal will be collected again and the interference signal will be determined again, thereby reducing the impact of interference sources in the environment on human perception.

[0077] The embodiments of the present application do not impose any restrictions on spatial distance, and can be dynamically configured according to the usage scenario of the device. For example, if the vibration wave processing device is carried with you, the spatial distance is set according to the carrying position and the position of the brain. If the vibration wave processing device is placed independently, the spatial distance can be set according to the size of the placement space.

[0078] It is understandable that there are multiple brainwave signals and the multiple brainwave signals are brainwave signals collected within a preset first time period; according to the brainwave signals, the preset interference conditions are optimized, including: Acquire a vibration wave signal to be processed that is collected in the same time period as the brain wave signal; The interference conditions are optimized according to the correlation between the brain wave signal and the vibration wave signal to be processed.

[0079] The present embodiment does not limit how the first duration is set. Those skilled in the art can determine the value of the first duration based on actual experimental data. The first duration can cover the time period before the adjustment wave is transmitted, or it can only cover the time period after the adjustment wave is transmitted. The present embodiment does not impose any additional restrictions on this.

[0080] The embodiments of the present application do not limit how to determine the correlation between the brain wave signal and the vibration wave signal, and those skilled in the art can selectively set it according to actual needs.

[0081] When the correlation represents the correlation between the changes in negative emotions of the target object (such as a person carrying a vibration wave processing device) affected by the noise cancellation effect as the vibration wave signal changes, and when the correlation is greater than the preset correlation threshold, it means that the vibration wave signal has a greater impact on the negative emotions of the target object, the noise cancellation effect of the adjustment wave signal has not met expectations, and the interference conditions need to be optimized.

[0082] In some embodiments, the correlation may be calculated based on a CCA / PCA algorithm, as follows: Acquiring audio feature data corresponding to a plurality of vibration wave signals that are time-aligned with the brain wave signal; Extract features from brainwave signals to obtain brainwave feature data corresponding to the brainwave signals; The audio feature data and brain wave feature data are used as variables and linear fitting is performed separately; Compare the results of linear fits of two variables over the same time period to determine correlation.

[0083] In some embodiments, when the correlation characterization shows that there are still multiple first target vibration wave signals after the adjustment wave is emitted, it will cause a negative impact. The first target vibration wave signal is a vibration wave signal among multiple vibration wave signals that are time-aligned with the brain wave signal; and the sound feature corresponding to the first target vibration wave signal does not match all the preset baseline sound features (that is, even if the frequency classification category is adjusted, the probability of determining it as an interference signal will not be increased). In some embodiments, the combined audio feature data of the multiple first target vibration wave signals in different combinations can be calculated separately, and the combined sound features and combined audio features in each combined audio feature data are matched with the preset baseline feature data to obtain target combined audio feature data, and at least one first target vibration wave signal corresponding to the target combined audio feature data is used as an interference signal for signal cancellation or suppression processing. In some embodiments, when there are multiple target combination audio feature data, the vibration wave signal combination that matches the baseline feature data can be first screened out from the multiple target combination audio data, and then the vibration wave signal with the largest number of repetitions in the screened vibration wave signal combination can be used as the first target vibration wave signal. When the vibration wave signals in each vibration wave combination corresponding to the multiple target combination audio feature data are different, at this time, one vibration wave signal in one group of vibration wave signal combinations can be selected as the first target vibration wave signal for attenuation or suppression according to the priority of the frequency classification category.

[0084] In some embodiments, at least two vibration wave signal acquisition modules 100 are provided for collecting vibration wave signals, at least one of which is detachable from the main body of the vibration wave processing device. Different vibration wave signal acquisition modules are placed in different locations, and the characteristics of the vibration waves collected from the same signal source also vary. In practical applications, each vibration wave signal acquisition module 100 can be placed at different locations relative to a target object, with one vibration wave signal acquisition module 100 positioned farther from the target object than the remaining vibration wave signal acquisition modules (e.g., if acquisition module 1 is at distance s1 from the target object, and acquisition modules 2-3 are at distances s2-s4, respectively, then s1 is less than s2-s4). This allows different vibration wave signal acquisition modules 100 to identify non-interference sources designated for the target object based on the frequency and sound characteristics of the signal sources at different locations. Non-interference sources can be sounds desired by the target object, such as sounds that aid sleep. For example, the user actively turns on music with a frequency of f1 to assist sleep, and there is also music with a frequency of f1 at other locations. When the music of f1 from two different sources is superimposed, the sleep assistance effect becomes ineffective, thereby causing negative emotions in the user. At this time, when negative emotions are present in the brain wave signal, the signal source positions and audio feature data determined by different vibration wave signal acquisition modules 100 are compared with each other, and unnecessary non-interference sources are identified and directly suppressed. For example, the signal source with the highest ratio calculated by weighting the distance and sound intensity is used as the unnecessary interference signal source, wherein the weighting coefficients of the distance and sound intensity can be adaptively adjusted according to actual conditions.

[0085] For example, taking the case of supporting electromagnetic wave signals and sound wave signals for noise recognition, combined with Figure 2 、 Figure 3 as well as Figure 5 The vibration wave processing method according to the embodiment of the present application is described with two specific examples as follows: Example 1: S1. Collect vibration wave signals in the full frequency domain. The vibration wave signals include two modes of signals, namely, acoustic wave signals and electromagnetic wave signals. For acoustic wave signals, the microphone acquisition module can be used to collect mainly the somatosensory vibration wave signals around the human head. The microphone sound acquisition module includes a microphone sound acquisition circuit and a signal processing and analysis module (i.e. Figure 3 DSP / MCU shown). Among them, refer to Figure 3As shown, the hardware structure of the microphone sound acquisition module can include a microphone driver circuit, a first-stage amplifier circuit, and a second-stage amplifier circuit. The first-stage amplifier circuit amplifies the differential signal generated during the microphone sound pickup process, while the second-stage differential amplifier circuit integrates the differential signal output by the previous stage (i.e., the first-stage amplifier circuit) into a single-ended signal. The use of the first and second-stage amplifier circuits can reduce the impact of circuit component noise on the collected signal. Since the signal collected by the microphone is not filtered, it can capture signals across the full frequency range from infrasound to ultrasound. The DSP / MCU can use algorithms to extract signal parameters (such as frequency, amplitude, power, and sound pressure) from the collected sound signal. For frequency and amplitude, a Fast Fourier Transform (FFT) can be used to transform the time domain signal into the frequency domain, thereby obtaining the frequency and amplitude of the main frequency band of the sound wave signal and simultaneously calculating the signal power and decibel value. Based on the digital signal conversion into an audio signal, frequency and sound features can be extracted accordingly, particularly by extracting audio features including short-term energy and zero-crossing rate (ZCR) in the time domain; Mel-frequency cepstral coefficients (MFCCs), spectral flux, and chromaticity in the frequency domain; and high-order audio features including fundamental frequency stability and harmonics to noise ratio (HNR) in the frequency domain. These features can then be used to determine the corresponding sound characteristics. In other embodiments, sound wave signals can also be analyzed using filters and output corresponding signal parameters. For electromagnetic wave signals, channel filters and receivers can be used to acquire the signal, demodulate the digital signal through a demodulator, and then convert the digital-to-analog signal into an audio signal. This allows the extraction of corresponding audio features, including frequency and sound features, from the audio signal based on feature extraction requirements. In some other embodiments, electromagnetic wave detectors may also be used for detection.

[0086] S2. Process the vibration wave signal, refer to Figure 5 As shown, the details are as follows: S2.1.1. When the collected vibration wave signal is a sound wave signal, convert the sound wave signal into a digital signal.

[0087] S2.1.2. When the collected vibration wave signal is an electromagnetic wave signal, convert the electromagnetic wave signal into a digital signal.

[0088] S2.2. Convert the digital signals converted in S2.1.1 and S2.1.2 into audio signals.

[0089] S3. Perform intelligent analysis and decision-making on the audio signal, specifically including the following: Perform feature extraction and classification on audio signals to obtain clustered audio feature data.

[0090] The frequency features and sound features in the clustered audio feature data are classified through the classifier model to determine the classification information. The sound features based on the clustered audio feature data are compared with the baseline feature data to determine whether it is noise under the corresponding classification information. At this time, it can be identified whether the vibration wave signal is common noise that can easily cause discomfort to people, including noise from normal hearing mechanisms or noise from abnormal hearing mechanisms.

[0091] S4. Issue adjustment instructions based on the decision result of S3, as follows: S4.1. When the interference signal is determined to be a sound wave signal, generate an adjustment sound wave; S4.2. When it is determined that the interference signal is an electromagnetic wave signal, an adjustment electromagnetic wave is generated.

[0092] In some embodiments, the adjustment sound wave can be Figure 2 The adjustment sound wave generation module 410 is shown as generating the adjustment sound wave. The adjustment sound wave generation module 410 can be composed of a controller and a sound generating unit. The controller determines the frequency and amplitude of the adjustment sound wave based on signal parameters such as the frequency of the interference signal, and then obtains data of the adjustment sound wave with the same frequency and opposite amplitude as the ambient sound. The sound generating unit generates the adjustment sound wave based on the data to suppress the damage caused by the ambient sound to the human body. The adjustment electromagnetic wave can be generated in the same way as the adjustment sound wave.

[0093] S4.3. Dynamically locate the source of the interference signal.

[0094] Since the vibration wave device of the embodiment of the present application can be carried as a portable device and the ambient noise in the environment is uncontrollable, relative motion between the vibration wave processing device and the ambient sound source is inevitable. The noise cancellation effect can be further ensured by dynamically locating the wave source of the interference signal. In some embodiments of the present application, the sound source can be located based on TDOA (time difference of arrival) and extended Kalman filter (EKF). For example, the measured value of the sound source is estimated by the time difference between each microphone in the microphone array receiving the sound source. Then, assuming that the ambient sound source and the target object are in relative uniform motion, a predicted value of the sound source can be obtained. The measured value and the predicted value are weighted and fused using EKF to obtain the final noise position of the interference signal.

[0095] At this time, by accurately transmitting and adjusting sound waves and / or electromagnetic waves to the range of the interference signal positioning, an effective reduction effect is achieved.

[0096] In some embodiments, the vibration wave processing device supports online real-time evaluation and optimization to further perform noise elimination processing based on individual differences, and further includes the following steps: Step 1: Collect brainwave signals. In some embodiments, brainwave signals can be collected by an EEG sensor. In some embodiments, a mature EEG device can be used for collection, considering the quality of the EEG signals.

[0097] After EEG signal acquisition, features can be extracted and classified based on the network model corresponding to the convolutional neural network algorithm (such as EEGNet) to obtain classification results that represent the EEG signal's representation of the human mental state. The network model can be pre-trained using existing massive EEG data and high-performance GPUs to form a deep learning model. This allows for direct feature extraction and classification based on the trained network model after EEG signal acquisition.

[0098] Step 2: Effect evaluation and optimization: At this time, based on the classification results of the EEG signals in step 1, it is judged whether the vibration wave signal after the adjustment wave denoising still causes discomfort to the individual. If it is satisfied, it means that the interference conditions need to be adjusted. Among them, individual correlation analysis is used to determine whether the interference conditions need to be adjusted. Individual correlation can use the CCA / PCA method to calculate the linear correlation coefficient between two groups of variables (the sound time-frequency domain characteristics and the EEG classification results are each a group of variables, that is, the clustered audio feature data and the EEG feature data are each a group of variables), so as to determine which type of signal is not regarded as an interference signal but has a perceptual impact on the human body. At this time, based on the determined signal, the frequency classification or sound classification is adjusted, and the process jumps to S1, so that when executing S3, the probability of identifying it as an interference signal can be improved, and the noise perception caused by individual differences can be reduced.

[0099] Example 2: Before initiating the vibration wave processing method, a vibration wave processing device is deployed. The device includes a smart terminal equipped with a high-precision microphone and electromagnetic wave detector, and a cloud server equipped with an artificial intelligence algorithm. The smart terminal uses its built-in microphone and electromagnetic wave detector to collect environmental noise and electromagnetic wave signals in real time, and transmits the collected vibration wave signals and brain wave signals to the cloud server.

[0100] At this time, refer to Figure 5 As shown in S2 and S3, upon receiving the vibration wave signal transmitted by the smart terminal, the cloud server immediately invokes a built-in artificial intelligence algorithm to intelligently identify the vibration wave signal. For any interference signals identified, the cloud server immediately sends a command to the smart terminal, initiating an active mitigation mechanism. Upon receiving the command, the smart terminal immediately generates sound waves in the opposite direction of the interference signal, effectively reducing noise. Simultaneously, for any detected microwave electromagnetic wave signals, the smart terminal also generates electromagnetic waves in the opposite direction, using a similar method, to offset their impact on brain waves.

[0101] In some embodiments, in step 2 of Example 1 above, during the entire system operation, the system can continuously collect and analyze environmental data through adaptive learning and optimization, and automatically adjust recognition and reduction parameters based on environmental changes to ensure optimal noise reduction. For example, if individual differences result in some vibration wave signals that should be identified as interference signals not being identified, the system will automatically modify the threshold for interference signal identification to improve the accuracy of interference signal identification perceived by the individual.

[0102] In summary, through the coordinated work of the above steps, the problem that traditional physical methods have limited noise treatment effects and may cause secondary pollution can be effectively solved; at the same time, the problem of electromagnetic wave signal influence can be eliminated; and the applicable scenarios are wider.

[0103] like Figure 6 As shown, Figure 6 This is a hardware structure diagram of a vibration wave processing device provided by one embodiment of the present application. It includes: The processor 701 may be implemented as a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is configured to execute relevant programs to implement the technical solutions provided in the embodiments of the present application. The memory 702 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called by the processor 701 to execute the vibration wave processing method of the embodiments of this application. Input / output interface 703, used to implement information input and output; Communication interface 704, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.); Bus 705 , which transmits information between various components of the device (e.g., processor 701 , memory 702 , input / output interface 703 , and communication interface 704 ); The processor 701 , the memory 702 , the input / output interface 703 and the communication interface 704 are connected to each other in communication within the device via a bus 705 .

[0104] An embodiment of the present application further provides an electronic device comprising the vibration wave processing device as described above.

[0105] An embodiment of the present application further provides a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the above-mentioned vibration wave processing method is implemented.

[0106] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory optionally includes a memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of the above-mentioned networks include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0107] Those skilled in the art will appreciate that all or some of the steps and systems disclosed above can be implemented as software, firmware, hardware, or any suitable combination thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is well known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. Furthermore, as is well known to those skilled in the art, communication media typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0108] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the above implementation mode. Technical personnel familiar with the art can also make various equivalent modifications or substitutions under the shared conditions that do not violate the spirit of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A vibration wave processing method, characterized in that: The method comprises: Collecting vibration wave signals in the full frequency domain and / or multiple modes within a preset spatial range to obtain vibration wave signals to be processed; Obtaining audio feature data corresponding to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed; determining an interference signal according to the audio feature data; The interference signal is reduced or suppressed.

2. The vibration wave processing method according to claim 1, characterized in that: The method of obtaining audio feature data corresponding one-to-one to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed includes: performing digital conversion and audio conversion on the vibration wave signal to be processed in sequence to obtain an audio signal corresponding one-to-one to the vibration wave signal to be processed; the vibration wave signal to be processed includes at least one of an acoustic wave signal and an electromagnetic wave signal; and obtaining the audio feature data by performing feature extraction on the audio signal through the feature extraction model.

3. The vibration wave processing method according to claim 1, wherein: The determining of the interference signal according to the audio feature data includes: The interference signal is determined according to the frequency feature and the sound feature, wherein the frequency feature and the sound feature are both one of the data in the audio feature data.

4. The vibration wave processing method according to claim 3, wherein: The determining the interference signal according to the frequency characteristics and the sound characteristics includes: Grouping the frequency characteristics and sound characteristic data corresponding to the plurality of vibration wave signals to be processed based on a preset clustering algorithm to obtain a plurality of clustering characteristic data; Determining interference clustering feature data from the plurality of clustering feature data according to preset reference feature data; The vibration wave signals corresponding to the interference clustering feature data are all used as the interference signals.

5. The vibration wave processing method according to claim 1, wherein: The reducing or suppressing processing of the interference signal includes: Acquiring signal parameters of the interference signal and generating an adjustment wave based on the signal parameters; Dynamically and in real time locate the vibration wave source position of the interference signal to obtain the vibration wave source position; The adjustment wave is emitted toward the source position of the vibration wave.

6. The vibration wave processing method according to claim 5, characterized in that: The dynamically and real-time positioning of the vibration wave source position of the interference signal to obtain the vibration wave source position includes: Determining, based on the mutual correlation between the preset collection points at the multiple different locations, a relative time delay for the collection points at the multiple different locations to receive the interference signal at the same location; The position is solved according to the relative time delay and a preset distance equation difference formula to determine the position of the vibration wave source.

7. A vibration wave processing device, characterized in that: include: A vibration wave signal acquisition module is used to acquire vibration wave signals in the full frequency domain and / or multiple modes within a preset spatial range to obtain vibration wave signals to be processed; a decision analysis and processing module, configured to obtain audio feature data corresponding to the vibration wave signal to be processed based on a preset feature extraction model and the vibration wave signal to be processed; and to determine an interference signal based on the audio feature data; The interference processing module is used to reduce or suppress the interference signal.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the vibration wave processing method according to any one of claims 1 to 6.

Images