Mobile sound source signal processing method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of sound signal analysis, and discloses a mobile sound source signal processing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring audio sampling signals and audio sampling frequencies of a target mobile sound source at all sampling moments in a preset sampling range; performing difference analysis processing on the audio sampling frequencies, and obtaining a first sampling moment speed of each audio sampling signal based on the audio turning frequencies after the difference analysis; performing feature analysis processing on the audio sampling signals through a preset sound recognition model, and obtaining a second sampling moment speed of each audio sampling signal based on the sound source feature frequency after feature analysis; determining a target sampling moment speed according to the first sampling moment speed and the second sampling moment speed; and performing reduction processing on each audio sampling signal according to the target sampling moment speed and the audio sampling frequency to obtain an audio reduction signal. According to the invention, the accuracy and effect of sound restoration of the mobile sound source are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sound signal analysis, and particularly to a method and device for processing mobile sound source signals, a computer device, and a storage medium. Background Art

[0002] In daily life, when a sound source is in a moving state, the sound emitted by the sound source will enter the human ear and cause frequency changes, resulting in sound distortion. In order to achieve sound restoration during the movement of the sound source, it is necessary to remove the influence of frequency offset based on the speed of the moving sound source. However, in the prior art, the measurement of the speed of the moving sound source mostly relies on ultrasonic waves or complex lidar technology, which has problems such as complex data processing, poor environmental adaptability, and limited measurement accuracy. It is difficult to achieve clear restoration of the sound during the movement of the sound source, especially with a larger deviation in the case of high-speed movement of the sound source, thus affecting the audio analysis and sound recognition effects. Summary of the Invention

[0003] Based on this, it is necessary to provide a method and device for processing mobile sound source signals, a computer device, and a storage medium for the above technical problems, so as to solve the problems of low accuracy in processing existing mobile sound source signals and poor sound restoration effect.

[0004] A method for processing mobile sound source signals includes: Obtaining audio sampling signals at all sampling moments within a preset sampling range of a target mobile sound source, and the audio sampling frequencies of each of the audio sampling signals; Performing difference analysis processing on the audio sampling frequencies, finding out the audio transition frequencies from all the audio sampling frequencies, and performing speed conversion processing on all the audio sampling frequencies based on the audio transition frequencies to obtain the speeds at the first sampling moments of each of the audio sampling signals; Performing feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source feature frequencies of the target mobile sound source, and performing speed conversion processing on all the audio sampling frequencies based on the sound source feature frequencies to obtain the speeds at the second sampling moments of each of the audio sampling signals; Determining the target sampling moment speeds of each of the audio sampling signals according to the speeds at the first sampling moments and the speeds at the second sampling moments; Performing restoration processing on each of the audio sampling signals according to the target sampling moment speeds and the audio sampling frequencies to obtain audio restoration signals corresponding to each of the audio sampling signals.

[0005] A device for processing mobile sound source signals includes: A sampling data acquisition module, configured to acquire audio sampling signals of a target moving sound source at all sampling moments within a preset sampling range, and the audio sampling frequencies of each of the audio sampling signals; An audio transition frequency conversion module, configured to perform difference analysis processing on the audio sampling frequencies, find out the audio transition frequencies from all the audio sampling frequencies, and perform speed conversion processing on all the audio sampling frequencies based on the audio transition frequencies to obtain the first sampling moment speeds of each of the audio sampling signals; An audio characteristic frequency conversion module, configured to perform characteristic analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequencies of the target moving sound source, and perform speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequencies to obtain the second sampling moment speeds of each of the audio sampling signals; A target speed determination module, configured to determine the target sampling moment speeds of each of the audio sampling signals according to the first sampling moment speeds and the second sampling moment speeds; An audio signal restoration module, configured to perform restoration processing on each of the audio sampling signals according to the target sampling moment speeds and the audio sampling frequencies to obtain audio restoration signals corresponding to each of the audio sampling signals.

[0006] A computer device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, where when the processor executes the computer-readable instructions, the above-mentioned moving sound source signal processing method is implemented.

[0007] A computer-readable storage medium, storing computer-readable instructions, where when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to execute the moving sound source signal processing method as described above.

[0008] In the above-mentioned moving sound source signal processing method, device, computer equipment and storage medium, the moving sound source signal processing method obtains audio sampling signals at all sampling moments within a preset sampling range of the target moving sound source, as well as the audio sampling frequencies of each audio sampling signal; performs difference analysis processing on the audio sampling frequencies, finds out the audio transition frequencies from all the audio sampling frequencies, and performs speed conversion processing on all the audio sampling frequencies based on the audio transition frequencies to obtain the speed at the first sampling moment of each audio sampling signal; performs feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequencies of the target moving sound source, and performs speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequencies to obtain the speed at the second sampling moment of each audio sampling signal; determines the target sampling moment speed of each audio sampling signal according to the speed at the first sampling moment and the speed at the second sampling moment; and performs restoration processing on each audio sampling signal according to the target sampling moment speed and the audio sampling frequency to obtain an audio restoration signal corresponding to each audio sampling signal. Based on accurate frequency sampling frequency analysis and in-depth analysis of the sound source characteristics by combining a preset sound recognition model, the present invention can accurately measure the speed of a moving sound source, ensuring the credibility of speed measurement. At the same time, the present invention realizes audio restoration based on a more accurate moving speed of the sound source, can effectively remove the interference of frequency offset, improves the accuracy and effect of sound source sound restoration, and greatly improves the audio quality of the moving sound source. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for the description of the embodiments of the present invention. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0010] Figure 1 is a flowchart of a moving sound source signal processing method according to an embodiment of the present invention; Figure 2 is a schematic diagram of the moving state of a sound source in a moving sound source signal processing method according to an embodiment of the present invention; Figure 3 is a schematic structural diagram of a moving sound source signal processing device according to an embodiment of the present invention; Figure 4 is a schematic diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0012] The mobile sound source signal processing method provided in this embodiment can be applied to an application scenario of sound restoration of the sound signal of a mobile sound source to avoid distortion. The application scenario includes a mobile sound source in a moving state and a sound collection component in a stationary state, and the client of the sound collection component communicates with the server. Among them, the client includes but is not limited to various personal computers, laptop computers, smart phones, tablet computers, and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers. The mobile sound source emits a sound signal, the client of the sound collection component collects the sound signal and sends it to the server, and the server obtains the mobile sound source sound signal and performs signal processing to finally obtain a sound restoration signal corresponding to the mobile sound source sound signal.

[0013] For example, the mobile sound source signal processing method of this embodiment can be applied to a drone communication scenario. The drone, as the target mobile sound source, emits a sound signal while moving. During the movement, the drone approaches from a distance and moves away after passing by the sound collection component. The client of the sound collection component collects the audio sampling signals of the target mobile sound source at all sampling moments within the preset sampling range and sends them to the server. The server obtains the audio sampling signals and the audio sampling frequencies of each audio sampling signal; performs difference analysis processing on the audio sampling frequencies, finds the audio transition frequency from all the audio sampling frequencies, and performs speed conversion processing on all the audio sampling frequencies based on the audio transition frequency to obtain the speed at the first sampling moment of each audio sampling signal; performs feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequency of the target mobile sound source, and performs speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequency to obtain the speed at the second sampling moment of each audio sampling signal; determines the target sampling moment speed of each audio sampling signal according to the speed at the first sampling moment and the speed at the second sampling moment; performs restoration processing on each audio sampling signal according to the target sampling moment speed and the audio sampling frequency to obtain an audio restoration signal corresponding to each audio sampling signal.

[0014] In one embodiment, as Figure 1 shown, a mobile sound source signal processing method is provided, including the following steps S10-S50: S10. Obtain the audio sampling signals of the target mobile sound source at all sampling moments within the preset sampling range, and the audio sampling frequencies of each of the audio sampling signals.

[0015] Understandably, the target moving sound source refers to the moving sound emission source of the sound signal that needs to be restored. The preset sampling range refers to the sound signal acquisition area preset corresponding to the sound acquisition component. Since the sound acquisition component is in a stationary state, the sound signal can only be acquired when the target moving sound source is within the preset sampling range. The sampling moment refers to the time point when the sound acquisition component acquires the sound signal in a timed manner, and each sampling moment corresponds to an audio sampling signal. The audio sampling signal refers to the sound signal emitted by the target moving sound source acquired by the sound acquisition component. Each audio sampling signal corresponds to an audio sampling frequency, and the audio sampling frequency refers to the sound wave vibration frequency of the audio sampling signal, in hertz (Hz), and the frequency determines the pitch of the sound signal.

[0016] S20. Perform difference analysis processing on the audio sampling frequencies, find the audio transition frequency from all the audio sampling frequencies, and perform speed conversion processing on all the audio sampling frequencies based on the audio transition frequency to obtain the first sampling moment speed of each audio sampling signal.

[0017] Understandably, the target moving sound source is in a moving state, the sound acquisition component is in a stationary state, and the movement process of the target moving sound source is to approach from a distance and continue to move away after passing the sound acquisition component. Due to the Doppler effect, when there is relative motion between the wave source (target moving sound source) and the observer (sound acquisition component), the frequency received by the observer (sound acquisition component) will change compared with the actual frequency of the wave source. Specifically, when the target moving sound source moves closer to the sound acquisition component, the audio sampling frequency received by the sound acquisition component will be greater than the actual frequency of the target moving sound source; when the target moving sound source moves away from the sound acquisition component, the audio sampling frequency received by the sound acquisition component will be less than the actual frequency of the target moving sound source. Therefore, when the target moving sound source passes the sound acquisition component, the audio sampling frequency will change from large to small, that is, the audio sampling frequency changes from a value higher than the actual frequency to a value close to the actual frequency and then to a value lower than the actual frequency. Therefore, by performing difference analysis on two adjacent audio sampling frequencies, the audio transition frequency can be found from all the audio sampling frequencies. The audio transition frequency refers to the actual frequency of the target moving sound source determined based on the difference analysis, that is, the audio sampling frequency close to or equal to the actual frequency of the target moving sound source found from all the audio sampling frequencies.

[0018] Furthermore, during the movement of the target moving sound source, each sampling moment corresponds not only to the audio sampling frequency and the actual frequency of an audio sampling signal, but also to the speed at the sampling moment. Since the audio transition frequency is close to the actual frequency of the target moving sound source and is minimally affected by the Doppler effect (equivalent to the relative speed between the target moving sound source and the sound acquisition component being zero at the sampling moment corresponding to the audio transition frequency), it can be used as the inherent frequency (actual frequency) of the target moving sound source to perform conversions with other audio sampling frequencies except the audio transition frequency, and finally obtain the speed of the target moving sound source at each sampling moment. That is, according to the Doppler effect formula, the conversion relationship between speed and frequency can be obtained, and speed conversion processing is performed on all audio sampling frequencies based on the audio transition frequency to obtain the first sampling moment speed of each audio sampling signal. The first sampling moment speed refers to the sound source moving speed at each sampling moment obtained by performing speed conversion on the basis of determining the inherent frequency using the difference analysis method.

[0019] S30. Perform feature analysis processing on the audio sampling signal through a preset sound recognition model to obtain the sound source characteristic frequency of the target moving sound source, and perform speed conversion processing on all audio sampling frequencies based on the sound source characteristic frequency to obtain the second sampling moment speed of each audio sampling signal.

[0020] Understandably, for the sound emitted by the target moving sound source, the characteristics of the sound itself determine the sound type, and different sound types have different inherent frequencies. The inherent frequency of the sound source refers to the constant frequency maintained by the sound source when vibrating and emitting sound without being interfered by external forces. This inherent frequency is the actual frequency of the target moving sound source when not affected by the Doppler effect. At this time, by performing feature analysis processing on the audio sampling signal through a preset sound recognition model, the sound type can be analyzed, and then the sound source characteristic frequency of the target moving sound source can be obtained. The sound source characteristic frequency refers to the actual frequency of the target moving sound source determined based on feature analysis. The preset sound recognition model is a neural network model that has been pre-trained to predict the corresponding sound type based on the characteristics of the sound signal and determine the inherent frequency.

[0021] Furthermore, use the sound source characteristic frequency as the inherent frequency (actual frequency) of the target moving sound source to perform conversions with all audio sampling frequencies, and finally obtain the speed of the target moving sound source at each sampling moment. That is, according to the Doppler effect formula, the conversion relationship between speed and frequency can be obtained, and speed conversion processing is performed on all audio sampling frequencies based on the sound source characteristic frequency to obtain the second sampling moment speed of each audio sampling signal. The second sampling moment speed refers to the sound source moving speed at each sampling moment obtained by performing speed conversion on the basis of determining the inherent frequency using the feature analysis method.

[0022] S40. Determine the target sampling time velocity of each of the audio sampling signals based on the velocity at the first sampling time and the velocity at the second sampling time.

[0023] Understandably, there is only one sound source moving velocity corresponding to the target moving sound source at each sampling time. Since the velocity at the first sampling time and the velocity at the second sampling time are the sound source moving velocities at each sampling time obtained by using two different analysis methods, the velocity at the first sampling time and the velocity at the second sampling time may be the same or different for each sampling time. Therefore, it is necessary to determine the target sampling time velocity of each sampling time based on the velocity at the first sampling time and the velocity at the second sampling time. For example, the average value, weighted average value or fitting solution method is used for the velocity at the first sampling time and the velocity at the second sampling time to determine the target sampling time velocity. The target sampling time velocity refers to the sound source moving velocity of the target moving sound source at each sampling time determined after analyzing and converting the audio sampling signal and the audio sampling frequency.

[0024] S50. Perform reduction processing on each of the audio sampling signals based on the target sampling time velocity and the audio sampling frequency to obtain audio reduction signals corresponding to the audio sampling signals.

[0025] Understandably, through the analysis of the previous steps, the audio sampling frequency and the target sampling time velocity of the target moving sound source at each sampling time can be obtained. Each sampling time corresponds to an audio sampling signal. Based on the Doppler effect, the functional relationship between velocity, sampling frequency and actual frequency can be obtained. Therefore, the actual frequency can be obtained according to the target sampling time velocity and the audio sampling frequency of each audio sampling signal, and frequency compensation processing is performed on the audio sampling signal to finally obtain the audio reduction signal corresponding to each audio sampling signal. The audio reduction signal refers to the sound signal after eliminating the influence of the Doppler effect for each audio sampling signal.

[0026] In this embodiment, by obtaining the audio sampling signals of the target moving sound source at all sampling moments within the preset sampling range and the audio sampling frequencies of each audio sampling signal, performing difference analysis processing on the audio sampling frequencies, finding the audio transition frequencies from all the audio sampling frequencies, and performing speed conversion processing on all the audio sampling frequencies based on the audio transition frequencies to obtain the speeds at the first sampling moments of each audio sampling signal; performing feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequencies of the target moving sound source, and performing speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequencies to obtain the speeds at the second sampling moments of each audio sampling signal; determining the target sampling moment speeds of each audio sampling signal according to the speeds at the first sampling moments and the speeds at the second sampling moments; and performing restoration processing on each audio sampling signal according to the target sampling moment speeds and the audio sampling frequencies to obtain the audio restoration signals corresponding to each audio sampling signal. Based on the accurate analysis of the sampling frequencies and combined with the in-depth analysis of the sound source characteristics by the preset sound recognition model, this embodiment can accurately measure the speed of the moving sound source and ensure the credibility of the speed measurement. At the same time, based on the more accurate moving speed of the sound source, this embodiment realizes audio restoration, can effectively remove the interference of frequency offset, improves the accuracy and effect of sound source audio restoration, and greatly improves the audio quality of the moving sound source.

[0027] In one embodiment, in step S10, that is, the obtaining of the audio sampling signals of the target moving sound source at all sampling moments within the preset sampling range includes: S101. When it is detected that the target moving sound source enters the preset detection range, controlling the sound acquisition component to start timing to collect sound signals, and ending the collection when it is detected that the target moving sound source leaves the preset detection range, so as to obtain the initial sampling signals at multiple sampling moments; S102. Performing preprocessing on each of the initial sampling signals to obtain the audio sampling signals at each sampling moment.

[0028] Understandably, the server of the sound acquisition component continuously detects whether a moving sound source enters the preset detection range. When it detects that the target moving sound source enters the preset detection range and moves towards the sound acquisition component, it controls the client of the sound acquisition component to start timing the acquisition of sound signals, and ends the acquisition when it detects that the target moving sound source moves away in the opposite direction after passing the sound acquisition component and leaves the preset detection range, obtaining the initial sampling signals at multiple sampling moments. The timing sampling method is to sample at fixed time intervals, such as sampling once every 0.1 seconds. The initial sampling signal refers to the original sound signal received by the sound acquisition component. The original sound signal cannot be directly analyzed, and each initial sampling signal at each sampling moment needs to go through preprocessing to obtain the audio sampling signals at each sampling moment. The preprocessing process includes operations such as noise reduction, filtering, and normalization, which can remove the interference factors of the original sound signal and make the preprocessed signal more suitable for subsequent analysis and processing.

[0029] In one embodiment, as Figure 2 shown, S1 represents a target moving sound source moving at a certain speed, S2 represents the sound acquisition component, and the target moving sound source moves from left to right, approaches and passes the sound acquisition component, and then moves away in the opposite direction. Each sampling moment corresponds to the audio sampling frequency of an audio sampling signal. T0 represents the first sampling moment when it starts to detect that the target moving sound source enters the preset detection range, and F0 represents the audio sampling frequency at the T0 sampling moment. T1 represents the next sampling moment after the T0 moment when the target moving sound source enters the preset detection range, that is, the second sampling moment of the timing acquisition. F1 represents the audio sampling frequency at the T1 sampling moment. T n+m represents the last sampling moment when it detects that the target moving sound source leaves the preset detection range, and F n+m represents the audio sampling frequency at the T n+m sampling moment.

[0030] In this embodiment, when it detects that the target moving sound source is within the preset detection range, it automatically triggers the timing acquisition of sound signals, and can obtain the initial sampling signals at multiple sampling moments, ensuring the comprehensiveness and integrity of the sampling. In addition, through the preprocessing of the initial sampling signals, the quality of the audio sampling signals can be improved, ensuring the usability of subsequent analysis and processing.

[0031] In one embodiment, in step S20, that is, the difference analysis process of the audio sampling frequency to find the audio transition frequency from all the audio sampling frequencies includes: S201. Perform a difference operation on the audio sampling frequencies of the audio sampling signals at each sampling moment and the audio sampling signals at the next sampling moment to obtain the sampling frequency change values of each audio sampling signal; S202. Find out the audio turning sampling signals when the sampling frequency change value has positive and negative changes from all the audio sampling signals, and determine the audio sampling frequency of the audio turning sampling signals as the audio turning frequency.

[0032] Understandably, each sampling moment corresponds to an audio sampling signal and its audio sampling frequency. By performing a difference operation on the audio sampling frequencies of two adjacent audio sampling signals, the sampling frequency change values of each audio sampling signal can be obtained. The sampling frequency change value refers to the difference between the audio sampling frequencies of the audio sampling signals at two adjacent sampling moments. For example, by calculating the difference between the audio sampling frequency at the first sampling moment and the audio sampling frequency at the second sampling moment, the sampling frequency change value of the second sampling moment compared to the first sampling moment can be obtained. By judging all the sampling frequency change values, find out the audio turning sampling signals when the sampling frequency change value has positive and negative changes from all the audio sampling signals, and determine the audio sampling frequency of the audio turning sampling signals as the audio turning frequency. The audio turning sampling signal refers to the audio sampling signal whose audio sampling frequency at the current sampling moment has a positive or negative turn compared to the audio sampling frequency at the previous sampling moment, that is, the audio sampling signal when the target moving sound source passes through the sound acquisition component.

[0033] In one embodiment, as Figure 2 shown, when the target moving sound source moves at a constant speed, affected by the Doppler effect, both F1 and F0 are values greater than the inherent frequency of the target moving sound source, and the value of "F1 - F0" should theoretically be zero. For example, when an ambulance with a siren passes by an observer at a constant speed, the sound frequency heard by the stationary observer will change from a constant high frequency to a constant lower frequency, although the siren generates a constant source frequency. The value of "F1 - F0" represents the sampling frequency change value of the second sampling moment compared to the first sampling moment. T n represents the sampling moment when it is detected that the target moving sound source passes through the sound acquisition component, and F n represents T n the audio sampling frequency at the sampling moment. When the target moving sound source directly passes through the sound acquisition component instantaneously, the direction of motion of the target moving sound source is perpendicular to the connection direction between the sound acquisition component, and at this time the velocity component in the connection direction is zero, and the Doppler effect will not cause a frequency change, that is, the audio sampling frequency F at the T n sampling moment n ​is equal to the natural frequency of the target moving sound source, i.e., the audio transition frequency. When the target moving sound source approaches the sound acquisition component, the component velocity in the connection direction is towards the sound acquisition component, and the position of the target moving sound source is closer to the sound acquisition component at the next sampling moment than at the previous sampling moment, resulting in the compression of the wavefront and causing the frequency to increase. When the target moving sound source moves away from the sound acquisition component, the component velocity in the connection direction is away from the sound acquisition component, and the position of the target moving sound source is farther from the sound acquisition component at the next sampling moment than at the previous sampling moment, resulting in the stretching of the wavefront and causing the frequency to decrease. Therefore, from sampling moment T n-1 to T n and then to T n+1 the corresponding audio sampling frequencies F n-1 and F n and between F n and F n+1 the sign of the change value of the sampling frequency will change between positive and negative, while the sign of the change value of the sampling frequency between other sampling moments will not change.

[0034] In another embodiment, after finding the audio sampling signals when the sign of the change value of the sampling frequency changes from all the audio sampling signals, the audio sampling signals at two adjacent sampling moments are further obtained as the audio transition sampling signals. At this time, there are three audio transition sampling signals, such as Figure 2 the audio sampling signals at T n-1 T n and T n+1 as shown. Further, the average value between the audio sampling frequencies of the three audio transition sampling signals is determined as the audio transition frequency.

[0035] In this embodiment, through the method of difference analysis, the audio transition sampling signals can be effectively found, and the audio sampling frequency of the audio transition sampling signals is determined as the audio transition frequency, improving the accuracy of subsequent speed conversion.

[0036] In one embodiment, in step S201, that is, before performing the difference operation on the audio sampling frequencies of the audio sampling signal at each sampling moment and the audio sampling signal at the next sampling moment, it includes: S2011. Determine whether there is a signal loss in the audio sampling signals at each sampling moment; S2012. If there is a signal loss in the audio sampling signal, the audio sampling frequency of the missing audio sampling signal is corrected by an interpolation algorithm to obtain the audio sampling frequency of the corrected audio sampling signal.

[0037] Understandably, before performing a difference operation on the audio sampling frequencies of the audio sampling signal at each sampling moment and the audio sampling signal at the next sampling moment, it is necessary to check whether there is missing or abnormal data in the audio sampling signal at each sampling moment. Specifically, methods such as waveform and spectrogram analysis, and instantaneous frequency analysis can be used for comprehensive judgment to find the audio sampling signals with missing or abnormal data. When there is signal loss or abnormal signal data in the audio sampling signal, the audio sampling frequency of the missing or abnormal audio sampling signal is corrected by an interpolation algorithm to obtain the audio sampling frequency of the corrected audio sampling signal. When using the interpolation algorithm for correction, the audio sampling frequency of the missing audio sampling signal is complemented by linear interpolation or spline interpolation, and the audio sampling frequency of the abnormal audio sampling signal is modified to ensure the sequence continuity of the audio sampling frequencies of all audio sampling signals.

[0038] In this embodiment, after determining whether the signal is missing, the audio sampling frequency of the missing signal is corrected by an interpolation algorithm, and a more complete and accurate audio sampling signal can be obtained, avoiding deviations caused by data loss in subsequent analysis.

[0039] In one embodiment, in step S30, that is, the feature analysis process of the audio sampling signal by the preset sound recognition model to obtain the sound source characteristic frequency of the target moving sound source includes: S301. Input the audio sampling signal into the preset sound recognition model, perform feature extraction processing on the audio sampling signal, and obtain the frequency domain feature information and time domain feature information of the audio sampling signal; S302. Perform feature analysis processing on the frequency domain feature information and time domain feature information to obtain the sound source type result of the audio sampling signal; S303. Obtain the type standard frequency matching the sound source type result, and determine the type standard frequency as the sound source characteristic frequency of the target moving sound source.

[0040] Understandably, first, the server can extract the frequency-domain feature information and time-domain feature information of the audio sampling signal through a preset sound recognition model. The frequency-domain feature information describes the distribution and variation of the audio signal in terms of frequency, while the time-domain feature information describes the variation and characteristics of the audio signal in terms of time. For example, the frequency-domain features are obtained through short-time Fourier transform, and the waveform amplitude is extracted as the time-domain feature. Then, the classifier in the preset sound recognition model performs feature analysis and processing on the frequency-domain feature information and time-domain feature information to obtain the sound source type result of the audio sampling signal. The sound source type result refers to the matching result obtained based on the matching degree between the feature information of the audio sampling signal and the feature information of known sound source types, such as human voice, car sound, etc. Finally, the type standard frequency that matches the sound source type result is searched for in the preset type-frequency association data table. The type standard frequency is a predefined inherent frequency used to characterize different sound source types. The preset type-frequency association data table is a pre-established mapping data table for recording the mapping between different sound source types and inherent frequencies. For example, the inherent frequency corresponding to the car engine sound is 200 Hz.

[0041] In this embodiment, through the preset sound recognition model, feature extraction, analysis, and matching are performed on the audio sampling signal, and finally the sound source characteristic frequency of the target moving sound source is determined, ensuring the efficiency and accuracy of the sound source characteristic frequency analysis.

[0042] In one embodiment, in step S20, that is, the speed conversion process is performed on all audio sampling frequencies based on the audio transition frequency to obtain the speed at the first sampling moment of each audio sampling signal, including: S203. Use the Doppler effect formula to perform conversion processing on the audio transition frequency and audio sampling frequency of each audio sampling signal to obtain the speed at the first sampling moment of each audio sampling signal; In step S30, that is, the speed conversion process is performed on all audio sampling frequencies based on the sound source characteristic frequency to obtain the speed at the second sampling moment of each audio sampling signal, including: S304. Use the Doppler effect formula to perform conversion processing on the sound source characteristic frequency and audio sampling frequency of each audio sampling signal to obtain the speed at the second sampling moment of each audio sampling signal.

[0043] Understandably, whether it is performing speed conversion processing using the audio transition frequency and audio sampling frequency, or using the sound source characteristic frequency and audio sampling frequency, it needs to be realized through the conversion relationship of the Doppler effect formula, and finally the speed at the first sampling moment and the speed at the second sampling moment of each audio sampling signal are obtained. The Doppler effect formula is , where represents the audio sampling frequency; represents the natural frequency of the target moving sound source; represents the speed of sound; represents the moving speed of the sound collection component. When the sound collection component is in a stationary state ; represents the moving speed of the target moving sound source. When the target moving sound source moves closer to the sound collection component take a positive value. When the target moving sound source moves closer to the sound collection component take a negative value.

[0044] In this embodiment, the conversion relationship of the Doppler effect formula is used to convert frequency to speed, and the speed at the first sampling moment and the speed at the second sampling moment of each audio sampling signal are obtained based on different frequencies, ensuring the data accuracy of subsequent speed analysis.

[0045] In one embodiment, in step S40, that is, determining the target sampling moment speed according to the speed at the first sampling moment and the speed at the second sampling moment includes: S401. Perform weighted calculation processing on the speed at the first sampling moment and the speed at the second sampling moment of each audio sampling signal to obtain the target sampling moment speed of each audio sampling signal.

[0046] It can be understood that the speed at the first sampling moment and the speed at the second sampling moment are determined based on different natural frequency analysis methods. By comprehensively analyzing the speed at the first sampling moment and the speed at the second sampling moment using a preset weight, a reasonable sampling speed can be obtained. Specifically, perform weighted calculation processing on the speed at the first sampling moment and the speed at the second sampling moment of each audio sampling signal according to the preset weight to obtain the target sampling moment speed of each audio sampling signal. The preset weight is a coefficient that is preset to assign different importance to different variables in a specific calculation process, and is used to adjust or balance the influence of different factors on the final result. The preset weight can be a fixed weight value determined according to the importance or reliability of the speed at the first sampling moment and the speed at the second sampling moment, or a dynamic weight value calculated and adjusted based on actual condition changes.

[0047] In this embodiment, the target sampling moment speed of each audio sampling signal can be obtained through weighted calculation processing, which can better represent the true speed of the audio signal at the sampling moment based on the speed at the first sampling moment and the speed at the second sampling moment, improving the rationality and effectiveness of the target sampling moment speed.

[0048] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0049] In one embodiment, a mobile sound source signal processing device is provided, which corresponds one-to-one with the mobile sound source signal processing method in the above embodiment. As Figure 3 shown, the mobile sound source signal processing device includes a sampling data acquisition module 10, an audio transition frequency conversion module 20, an audio characteristic frequency conversion module 30, a target speed determination module 40, and an audio signal restoration module 50. The detailed description of each functional module is as follows: The sampling data acquisition module 10 is configured to acquire audio sampling signals of a target mobile sound source at all sampling moments within a preset sampling range, and the audio sampling frequencies of each of the audio sampling signals; The audio transition frequency conversion module 20 is configured to perform difference analysis processing on the audio sampling frequencies, find the audio transition frequency from all the audio sampling frequencies, and perform speed conversion processing on all the audio sampling frequencies based on the audio transition frequency to obtain the first sampling moment speed of each of the audio sampling signals; The audio characteristic frequency conversion module 30 is configured to perform feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequency of the target mobile sound source, and perform speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequency to obtain the second sampling moment speed of each of the audio sampling signals; The target speed determination module 40 is configured to determine the target sampling moment speed of each of the audio sampling signals according to the first sampling moment speed and the second sampling moment speed; The audio signal restoration module 50 is configured to perform restoration processing on each of the audio sampling signals according to the target sampling moment speed and the audio sampling frequency to obtain an audio restoration signal corresponding to each of the audio sampling signals.

[0050] In one embodiment, the sampling data acquisition module 10 includes: An initial signal sampling unit, configured to control a sound acquisition component to start timing to collect sound signals when it is detected that the target mobile sound source enters the preset detection range, and end the collection when it is detected that the target mobile sound source leaves the preset detection range, so as to obtain initial sampling signals at multiple sampling moments; An initial signal processing unit, configured to perform preprocessing on each of the initial sampling signals to obtain audio sampling signals at each sampling moment.

[0051] In one embodiment, the audio transition frequency conversion module 20 includes: A frequency difference calculation unit, configured to perform difference operation on the audio sampling frequencies of the audio sampling signals at each sampling moment and the audio sampling signals at the next sampling moment to obtain the sampling frequency change values of each audio sampling signal; An audio transition frequency determination unit is configured to find out audio transition sampling signals when the sampling frequency change value undergoes positive and negative changes from all audio sampling signals, and determine the audio sampling frequency of the audio transition sampling signals as the audio transition frequency.

[0052] In one embodiment, the audio transition frequency conversion module 20 further includes: A signal loss judgment unit is configured to judge whether there is signal loss in the audio sampling signals at each sampling moment; A sampling frequency correction unit is configured to, if there is signal loss in the audio sampling signals, correct the audio sampling frequency of the missing audio sampling signals through an interpolation algorithm to obtain the audio sampling frequency of the corrected audio sampling signals.

[0053] In one embodiment, the audio feature frequency conversion module 30 includes: A feature extraction processing unit is configured to input the audio sampling signals into a preset voice recognition model, perform feature extraction processing on the audio sampling signals, and obtain the frequency domain feature information and time domain feature information of the audio sampling signals; A feature analysis processing unit is configured to perform feature analysis processing on the frequency domain feature information and time domain feature information to obtain the sound source type result of the audio sampling signals; A sound source feature frequency determination unit is configured to obtain a type standard frequency matching the sound source type result, and determine the type standard frequency as the sound source feature frequency of the target moving sound source.

[0054] In one embodiment, the audio transition frequency conversion module 20 further includes: A first speed determination unit is configured to perform conversion processing on the audio transition frequency and audio sampling frequency of each audio sampling signal by using the Doppler effect formula to obtain the first sampling moment speed of each audio sampling signal; The audio feature frequency conversion module 30 further includes: A second speed determination unit is configured to perform conversion processing on the sound source feature frequency and audio sampling frequency of each audio sampling signal by using the Doppler effect formula to obtain the second sampling moment speed of each audio sampling signal.

[0055] In one embodiment, the target speed determination module 40 includes: A speed weighting processing unit is configured to perform weighted calculation processing on the first sampling moment speed and the second sampling moment speed of each audio sampling signal to obtain the target sampling moment speed of each audio sampling signal.

[0056] For the specific limitations of the mobile sound source signal processing device, reference can be made to the limitations of the mobile sound source signal processing method in the foregoing text, which will not be elaborated here. Each module in the above mobile sound source signal processing device can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0057] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 4 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer-readable instructions, and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The database of the computer device is used to store the data involved in the mobile sound source signal processing method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a mobile sound source signal processing method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0058] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored on the memory and executable on the processor. When the processor executes the computer-readable instructions, the following steps are implemented: Obtain the audio sampling signals of the target mobile sound source at all sampling moments within a preset sampling range, and the audio sampling frequencies of each of the audio sampling signals; Perform difference analysis processing on the audio sampling frequencies, find the audio transition frequency from all the audio sampling frequencies, and perform speed conversion processing on all the audio sampling frequencies based on the audio transition frequency to obtain the first sampling moment speed of each of the audio sampling signals; Perform feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequency of the target mobile sound source, and perform speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequency to obtain the second sampling moment speed of each of the audio sampling signals; Determine the target sampling moment speed of each of the audio sampling signals according to the first sampling moment speed and the second sampling moment speed; Restore each of the audio sampling signals according to the speed at the target sampling moment and the audio sampling frequency to obtain audio restoration signals corresponding to the respective audio sampling signals.

[0059] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. Computer-readable instructions are stored on the readable storage media. When the computer-readable instructions are executed by one or more processors, the following steps are implemented: Obtain audio sampling signals of a target moving sound source at all sampling moments within a preset sampling range, and the audio sampling frequencies of the respective audio sampling signals; Perform difference analysis processing on the audio sampling frequencies, find an audio transition frequency from all the audio sampling frequencies, and perform speed conversion processing on all the audio sampling frequencies based on the audio transition frequency to obtain the speed at the first sampling moment of each audio sampling signal; Perform feature analysis processing on the audio sampling signals through a preset sound recognition model to obtain the sound source characteristic frequency of the target moving sound source, and perform speed conversion processing on all the audio sampling frequencies based on the sound source characteristic frequency to obtain the speed at the second sampling moment of each audio sampling signal; Determine the speed at the target sampling moment of each audio sampling signal according to the speed at the first sampling moment and the speed at the second sampling moment; Restore each of the audio sampling signals according to the speed at the target sampling moment and the audio sampling frequency to obtain audio restoration signals corresponding to the respective audio sampling signals.

[0060] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through computer-readable instructions. The computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided by the present invention can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0061] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0062] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for processing a mobile sound source signal, characterized in that: include: Acquire audio sampling signals of the target moving sound source at all sampling moments within a preset sampling range, and the audio sampling frequency of each of the audio sampling signals; Performing difference analysis processing on the audio sampling frequency, finding the audio turning frequency from all the audio sampling frequencies, and performing speed conversion processing on all the audio sampling frequencies based on the audio turning frequency to obtain the first sampling time speed of each of the audio sampling signals; Performing feature analysis processing on the audio sampling signal through a preset sound recognition model to obtain a sound source characteristic frequency of the target moving sound source, and performing speed conversion processing on all audio sampling frequencies based on the sound source characteristic frequency to obtain a second sampling time speed of each of the audio sampling signals; Determine a target sampling time speed of each of the audio sampling signals according to the first sampling time speed and the second sampling time speed; Each of the audio sampling signals is restored according to the target sampling time speed and the audio sampling frequency to obtain an audio restoration signal corresponding to each of the audio sampling signals.

2. The method for processing a mobile sound source signal according to claim 1, wherein: The step of obtaining the audio sampling signals of the target moving sound source at all sampling moments within a preset sampling range includes: When the target moving sound source is detected to enter the preset detection range, the sound collection component is controlled to start collecting sound signals at a fixed time, and the collection ends when the target moving sound source is detected to leave the preset detection range, thereby obtaining initial sampling signals at multiple sampling moments; Each of the initial sampling signals is preprocessed to obtain an audio sampling signal at each sampling moment.

3. The method for processing a mobile sound source signal according to claim 1, wherein: The performing difference analysis on the audio sampling frequencies to find the audio turning frequency from all the audio sampling frequencies includes: Performing a difference operation on the audio sampling frequency of the audio sampling signal at each sampling moment and the audio sampling signal at the next sampling moment to obtain a sampling frequency change value of each audio sampling signal; An audio transition sampling signal when the sampling frequency change value changes positively or negatively is found from all audio sampling signals, and the audio sampling frequency of the audio transition sampling signal is determined as the audio transition frequency.

4. The method for processing a mobile sound source signal according to claim 3, wherein: Before performing a difference operation on the audio sampling frequency of the audio sampling signal at each sampling moment and the audio sampling signal at the next sampling moment, the method includes: Determine whether there is a signal loss in the audio sampling signal at each sampling moment; If there is a missing signal in the audio sampling signal, the audio sampling frequency of the missing audio sampling signal is corrected by an interpolation algorithm to obtain the audio sampling frequency of the corrected audio sampling signal.

5. The method for processing a mobile sound source signal according to claim 1, wherein: The performing feature analysis processing on the audio sampling signal by using a preset sound recognition model to obtain the sound source characteristic frequency of the target moving sound source includes: Inputting the audio sampling signal into a preset sound recognition model, performing feature extraction processing on the audio sampling signal, and obtaining frequency domain feature information and time domain feature information of the audio sampling signal; Performing feature analysis on the frequency domain feature information and the time domain feature information to obtain a sound source type result of the audio sampling signal; A type standard frequency matching the sound source type result is obtained, and the type standard frequency is determined as the sound source characteristic frequency of the target moving sound source.

6. The method for processing a mobile sound source signal according to claim 1, wherein: The performing speed conversion processing on all audio sampling frequencies based on the audio transition frequency to obtain the first sampling time speed of each audio sampling signal includes: The audio corner frequency and the audio sampling frequency of each of the audio sampling signals are converted by using a Doppler effect formula to obtain a first sampling time velocity of each of the audio sampling signals; The performing speed conversion processing on all audio sampling frequencies based on the sound source characteristic frequency to obtain the second sampling time speed of each audio sampling signal includes: The Doppler effect formula is used to convert the sound source characteristic frequency and the audio sampling frequency of each of the audio sampling signals to obtain the second sampling time velocity of each of the audio sampling signals.

7. The method for processing a mobile sound source signal according to claim 1, wherein: The step of determining a target sampling time speed according to the first sampling time speed and the second sampling time speed comprises: A weighted calculation process is performed on the first sampling time rate and the second sampling time rate of each of the audio sampling signals to obtain a target sampling time rate of each of the audio sampling signals.

8. A mobile sound source signal processing device, characterized in that: include: A sampling data acquisition module, used to acquire audio sampling signals of the target mobile sound source at all sampling moments within a preset sampling range, and the audio sampling frequency of each of the audio sampling signals; An audio corner frequency conversion module, used to perform difference analysis processing on the audio sampling frequency, find the audio corner frequency from all the audio sampling frequencies, and perform speed conversion processing on all the audio sampling frequencies based on the audio corner frequency to obtain the first sampling time speed of each of the audio sampling signals; An audio characteristic frequency conversion module, used to perform characteristic analysis processing on the audio sampling signal through a preset sound recognition model to obtain the sound source characteristic frequency of the target moving sound source, and perform speed conversion processing on all audio sampling frequencies based on the sound source characteristic frequency to obtain the second sampling time speed of each of the audio sampling signals; a target speed determining module, configured to determine a target sampling time speed of each of the audio sampling signals according to the first sampling time speed and the second sampling time speed; The audio signal restoration module is used to restore each of the audio sampling signals according to the target sampling time speed and the audio sampling frequency to obtain an audio restoration signal corresponding to each of the audio sampling signals.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the mobile sound source signal processing method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the mobile sound source signal processing method according to any one of claims 1 to 7.

Images