A machine learning based in-vehicle audio sound processing system

Abstract

The present application relates to the technical field of vehicle-mounted equipment analysis, and particularly relates to a vehicle-mounted audio frequency processing system based on machine learning, which comprises: a vehicle information acquisition module, which is used to acquire running information and audio equipment information of a vehicle; an information storage and classification module, which is used to classify the running information of the vehicle to obtain an acquisition information type, and store the acquired running information, audio equipment information and acquisition information type of the vehicle; a vehicle feature analysis module, which is used to analyze vehicle carrying features and vehicle driving features; a storage data analysis module, which is used to analyze stored audio features; an audio data analysis module, which is used to analyze audio quality parameters according to similar data, vehicle carrying features, vehicle driving features, type difference features and historical fluctuation features; and an audio frequency regulation module, which is used to control the volume of the audio equipment of the vehicle. The present application realizes accurate analysis and processing of vehicle-mounted audio frequency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle-mounted equipment analysis, and in particular to a vehicle-mounted audio frequency processing system based on machine learning. BACKGROUND

[0002] With the development of automobile intelligence, vehicle-mounted audio frequency processing technology is increasingly important. As a core technology, digital signal processors can convert analog signals into digital signals for efficient real-time processing, optimize sound quality, improve driving safety and comfort, and meet users' demand for high-quality audio experience.

[0003] Chinese Patent Publication No. CN117221786A discloses a loudness compensation method for a vehicle-mounted audio system and a vehicle-mounted audio system, comprising the following steps: Step A, obtaining a plurality of actual window areas of a vehicle, processing the plurality of actual window areas to obtain a plurality of weighted window areas; Step B, obtaining a plurality of loudness compensation amounts from the plurality of weighted window areas, the plurality of loudness compensation amounts being one-to-one corresponding to a plurality of loudspeakers, and performing loudness compensation on the plurality of loudspeakers based on the plurality of loudness compensation amounts. The present application realizes the analysis of the loudness compensation amount of the audio according to the window area of the vehicle, and does not realize the comprehensive analysis of the quality of the audio data of the audio in combination with the bearing data, driving data and in-vehicle audio data of the vehicle. The analysis efficiency of the vehicle-mounted audio data is low, and the vehicle-mounted audio processing is not accurate. SUMMARY

[0004] The present application aims to provide a vehicle-mounted audio frequency processing system based on machine learning to solve at least one of the problems in the prior art.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] A vehicle-mounted audio frequency processing system based on machine learning, comprising:

[0007] A vehicle information acquisition module for acquiring vehicle operating information and audio equipment information;

[0008] An information storage and classification module for classifying the vehicle operating information to obtain the type of collected information, and storing the collected vehicle operating information, audio equipment information and collected information type;

[0009] A vehicle feature analysis module for analyzing vehicle bearing characteristics and vehicle driving characteristics based on vehicle operating information and audio equipment information;

[0010] The storage data analysis module is configured to analyze audio parameters of each type of collected information according to stored vehicle operation information and audio equipment information, and to analyze type difference characteristics and historical fluctuation characteristics according to the stored vehicle operation information, audio equipment information and audio parameters;

[0011] The audio data analysis module is configured to extract similar data according to the type of collected information, operation information and audio equipment information, and to analyze audio quality parameters according to the similar data, vehicle carrying characteristics, vehicle driving characteristics, type difference characteristics and historical fluctuation characteristics.

[0012] The sound audio control module is configured to control the volume of the audio equipment of the vehicle according to the audio quality parameters.

[0013] Further, the information storage classification module calculates the average value of the opening degree of each window of the vehicle as the average opening degree of the window, and compares the average opening degree of the window P(t) with the opening degree threshold p, and judges the type of collected information according to the comparison result, the type of collected information including type 1, type 2 and type 3.

[0014] Further, the vehicle characteristic analysis module is provided with a carrying characteristic analysis unit, which is configured to analyze the vehicle carrying characteristics according to the number of passengers N1(t), the received signal strength D1(t) and the audio signal strength D2(t) when the type of collected information is type 2 and type 3, to obtain the vehicle carrying characteristics A(t).

[0015] The vehicle characteristic analysis module is also provided with a driving characteristic analysis unit, which is configured to analyze the vehicle driving characteristics according to the driving speed V(t), the received signal strength D1(t) and the audio signal strength D2(t), to obtain the vehicle driving characteristics B(t).

[0016] Further, the storage data analysis module is provided with a type difference analysis unit, which is configured to analyze the audio parameters of the stored vehicle operation information and audio equipment information of type 1, type 2 and type 3 respectively, and to analyze the type difference characteristics according to the audio parameters L(j, t), to obtain the type difference characteristics C, wherein j represents the type parameter.

[0017] Further, the storage data analysis module is also provided with a historical fluctuation analysis unit, which is configured to analyze the historical fluctuation characteristics R(j) of the stored vehicle operation information and audio equipment information of type 1, type 2 and type 3 respectively.

[0018] Further, the audio data analysis module is provided with a similar data extraction unit, which is configured to extract similar data of the same type of collected information as the type of collected information of the currently analyzed data and |D2(t maxThe stored vehicle operation information and audio equipment information of ) / D2(t|t∈U(1)∪U(2)∪U(3))-1|≤α are used as the first similar data, and the similar data extraction unit extracts |V(t) max ) / V(t|t∈U(1)∪U(2)∪U(3))-1|≤α and|D1(t) max The stored vehicle operation information and audio device information of ) / D1(t|t∈U(1)∪U(2)∪U(3))-1|≤α are used as the second similar data, where D1(t max D2(t) represents the received signal strength currently being analyzed. max V(t) represents the intensity of the audio signal currently being analyzed. max ) represents the current driving speed being analyzed, and α represents the similarity comparison threshold.

[0019] Furthermore, the audio data analysis module also includes an audio quality analysis unit, which is used to analyze the audio quality parameters based on the vehicle load characteristics A(t), vehicle driving characteristics B(t), first similar data and type difference characteristics C, so as to obtain the audio quality parameters F(t).

[0020] Furthermore, the audio data analysis module is also equipped with a historical feature analysis unit, which is used to compare the historical fluctuation parameter R(j) with the fluctuation comparison threshold r. When the historical fluctuation parameter does not meet the threshold, the analysis process of the audio quality parameter is adjusted, and the adjusted audio quality parameter is F1(t).

[0021] Furthermore, the audio data analysis module is also equipped with a similar data analysis unit, which is used to count the number of second similar data whose information type is the same as the information type of the data being analyzed, and compare the second similar number N2 with the number comparison threshold β. When the second similar number meets the threshold, the adjustment process of the audio quality parameters is optimized, and the optimized audio quality parameters are F2(t).

[0022] Furthermore, the audio control module compares the audio quality parameter F(t) with the quality comparison thresholds f1 and f2, and controls the speaker volume of the vehicle according to the comparison result, so that the collected audio device information satisfies D1(t) / [D2(t)+1]=F(t).

[0023] The beneficial effects of the present application are as follows: through the collection of vehicle operation information and audio equipment information by the vehicle information collection module, and the analysis of the collected data by other modules, intelligent control of the vehicle audio audio in various driving environments is realized, the vehicle audio audio is ensured to meet the driving needs of the user, thereby improving the analysis efficiency of the system on the vehicle audio audio data and improving the accuracy of the vehicle audio audio processing. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0025] Figure 1 The structure diagram of the vehicle audio audio processing system based on machine learning of the present embodiment.

[0026] Figure 2 The structure diagram of the vehicle feature analysis module of the present embodiment.

[0027] Figure 3 The structure diagram of the storage data analysis module of the present embodiment.

[0028] Figure 4 The structure diagram of the audio data analysis module of the present embodiment. DETAILED DESCRIPTION

[0029] In order to more clearly illustrate the present application, the present application will be further described below in combination with preferred embodiments and drawings. In the drawings, similar components are denoted by the same reference numerals. Those skilled in the art should understand that the specific description below is illustrative rather than limiting, and should not limit the protection scope of the present application.

[0030] It should be noted that although the terms first, second, third, etc. may be used in the embodiments of the present application, these descriptions should not be limited to these terms. These terms are only used to distinguish the description. For example, without departing from the scope of the embodiments of the present application, the first can also be called the second, and similarly, the second can also be called the first.

[0031] Please refer to Figure 1 As shown in the figure, the vehicle audio audio processing system based on machine learning of the present embodiment comprises:

[0032] The vehicle information collection module is used to collect the running information and audio equipment information of the vehicle, the running information includes the number of passengers, the opening degree of the window, and the driving speed, the opening degree of the window is the percentage of the unopened part of the window to the whole window size, 1 indicates that the window is opened, and 0 indicates that the window is fully opened, the audio equipment information includes audio information and received signal strength, the audio information includes the speaker volume and the audio signal strength, the received signal strength is the signal strength of the sound data received by the microphone, the collection mode of the opening degree of the window and the driving speed is obtained through the vehicle operation data management platform data import, and the number of passengers is obtained through the pressure sensor installed below the vehicle.

[0033] Specifically, the embodiment is applied to a vehicle equipment operation management platform cloud, and analysis is performed on the audio data of the vehicle audio during driving, so that the audio volume of the vehicle audio meets the driving needs of the user.

[0034] Please continue to refer to Figure 1 As shown in the figure, the machine learning-based vehicle audio processing system further includes:

[0035] The information storage and classification module is used to classify the running information of the vehicle to obtain the collection information type, and store the collected running information of the vehicle, audio equipment information and collection information type, and the information storage and classification module is connected with the vehicle information collection module.

[0036] Specifically, in the embodiment, the information storage and classification module calculates the average value of the opening degree of each window of the vehicle as the average opening degree of the window, sets the average opening degree of the window as P(t), and sets The average opening degree of the window is compared with the opening degree threshold value, and the collection information type is judged according to the comparison result, if P(t) < p, the information storage and classification module determines that the collection information type is type I; if p ≤ P(t) < 1, the information storage and classification module determines that the collection information type is type II; if P(t) = 1, the information storage and classification module determines that the collection information type is type III; wherein i represents the window number, t represents the collection time number, N i represents the number of vehicle windows, P1(i, t) represents the opening degree of the window, p represents the opening degree threshold value, and 0.4 ≤ p ≤ 0.7. The window number is defined to distinguish the numbers of different windows on the vehicle, and the collection time number is defined to distinguish the numbers of different collection times. It can be understood that the value of the opening degree threshold value is not specifically limited in the embodiment, and a person skilled in the art can freely set it, as long as it meets the judgment of the collection information type, and the best value of the opening degree threshold value is p = 0.6.

[0037] Specifically, the information storage and classification module analyzes the operation information of the vehicle in this embodiment to classify and store the collected data according to the opening degree of the vehicle window, increase the sample diversity of system analysis, ensure the accuracy of the data used by the system when analyzing historical data, thereby improving the analysis efficiency of the system for the audio data of the vehicle audio and improving the accuracy of the audio processing of the vehicle audio.

[0038] Please continue to refer to Figure 1 As shown in the figure, the vehicle audio processing system based on machine learning further comprises:

[0039] A vehicle feature analysis module is configured to analyze the vehicle carrying feature and the vehicle driving feature according to the operation information of the vehicle and the audio equipment information, and the vehicle feature analysis module is connected to the information storage and classification module.

[0040] Please refer to Figure 2 As shown in the figure, the vehicle feature analysis module comprises:

[0041] A carrying feature analysis unit is configured to analyze the vehicle carrying feature according to the number of passengers, the type of collected information and the audio equipment information, and the vehicle carrying feature represents the characteristic relationship between the vehicle carrying data and the audio data of the vehicle audio and the sound data received by the microphone.

[0042] Specifically, the carrying feature analysis unit analyzes the vehicle carrying feature according to the number of passengers and the audio equipment information when the type of collected information is two and three in this embodiment, sets the vehicle carrying feature as A(t), and sets A(t) = [D1(t) / (D2(t)+1)]xlnN1(t), where D1(t) represents the received signal strength, D2(t) represents the audio signal strength, and N1(t) represents the number of passengers. The carrying feature analysis unit analyzes the type of collected information, the number of passengers and the audio equipment information to analyze the vehicle carrying feature, realizes the comprehensive analysis of the vehicle carrying data and the audio data of the vehicle audio, distinguishes the order of magnitude of the characteristic data in the subsequent analysis of the carrying data, thereby improving the analysis efficiency of the system for the audio data of the vehicle audio and improving the accuracy of the audio processing of the vehicle audio.

[0043] Please continue to refer to Figure 2 As shown in the figure, the vehicle feature analysis module further comprises:

[0044] A driving feature analysis unit is configured to analyze the vehicle driving feature according to the driving speed and the audio equipment information, and the vehicle driving feature represents the characteristic relationship between the driving speed of the vehicle and the audio data of the vehicle audio and the sound data received by the microphone, and the driving feature analysis unit is connected to the carrying feature analysis unit.

[0045] Specifically, the driving feature analysis unit in this embodiment analyzes the vehicle driving features according to the driving speed, the received signal strength and the audio signal strength, sets the vehicle driving features as B(t), and sets B(t) = [D1(t) / (D2(t)+1)]xlgV(t), where V(t) represents the driving speed. Through the analysis of the driving speed and the audio device information by the driving feature analysis unit, the comprehensive analysis of the vehicle driving data and the audio data is realized, the magnitude of the feature data in the subsequent driving data analysis is distinguished, the analysis efficiency of the system for the vehicle audio data is improved, and the accuracy of the vehicle audio processing is improved.

[0046] Please continue to refer to Figure 1 As shown in the figure, the machine learning-based vehicle audio processing system further includes:

[0047] The stored data analysis module is used to analyze the type difference features and the historical fluctuation features according to the stored vehicle operation information, the audio device information and the collected information types, and is connected with the information storage and classification module.

[0048] Please refer to Figure 3 As shown in the figure, the stored data analysis module includes:

[0049] The type difference analysis unit is used to analyze the type difference features according to the stored vehicle operation information, the audio device information and the collected information types, and use the type difference features to represent the feature difference between the data of different collected information types in the historical data.

[0050] Specifically, the type difference analysis unit in this embodiment analyzes the audio parameters of the stored vehicle operation information and the audio device information of the collected information types of one type, two types and three types, respectively, sets the audio parameters as L(j,t), and sets analyzes the type difference features according to the audio parameters, sets the type difference features as C, and sets where j represents the type parameter, U(j) represents the set of time numbers corresponding to each collected information type, NU(j) represents the number of elements in the set of time numbers corresponding to each collected information type, j=1 represents that the collected information type is one type, j=2 represents that the collected information type is two types, and j=3 represents that the collected information type is three types. Through the analysis of the stored vehicle operation information and the audio device information of each collected information type by the type difference analysis unit, the audio parameters are analyzed, the feature relationship between the received sound data and the audio data in the data of each collected information type is represented by the audio parameters, the type difference features are analyzed, the comprehensive analysis of the audio data features of different collected information types is realized, the analysis efficiency of the system for the vehicle audio data is improved, and the accuracy of the vehicle audio processing is improved.

[0051] Please continue to see Figure 3 The storage data analysis module further includes:

[0052] A historical fluctuation analysis unit is configured to analyze historical fluctuation characteristics based on the stored vehicle operation information, audio equipment information, and collection information types, and the historical fluctuation analysis unit is connected to the type difference analysis unit.

[0053] Specifically, the historical fluctuation analysis unit analyzes historical fluctuation characteristics of the stored vehicle operation information, audio equipment information, and collection information types of categories one, two, and three, respectively, sets the historical fluctuation characteristics as R(j), and sets Through the analysis of the historical fluctuation analysis unit on the data of each collection information type, the historical fluctuation parameters are analyzed, the fluctuation characteristics of the data in each collection information type are represented by the historical fluctuation parameters, the analysis of the change amplitude of the data in each collection information type is realized, and thus the analysis efficiency of the system on the vehicle audio frequency data is improved, and the accuracy of the vehicle audio frequency processing is improved.

[0054] Please continue to see Figure 1 The vehicle audio frequency processing system based on machine learning further includes:

[0055] An audio data analysis module is configured to extract similar data based on the collection information type, the operation information, and the audio equipment information, and to analyze the audio quality parameters based on the similar data, the vehicle carrying characteristics, the vehicle driving characteristics, the type difference characteristics, and the historical fluctuation characteristics, and the audio data analysis module is connected to the vehicle characteristic analysis module and the storage data analysis module.

[0056] Please see Figure 4 The audio data analysis module includes:

[0057] A similar data extraction unit is configured to extract similar data based on the collection information type and the operation information.

[0058] Specifically, the similar data extraction unit extracts the stored vehicle operation information and audio equipment information of the same collection information type as the currently analyzed data and |D2(t max ) / D2(t|t∈U(1)∪U(2)∪U(3))-1|≤α as the first similar data, and the similar data extraction unit extracts |V(t max ) / V(t|t∈U(1)∪U(2)∪U(3))-1|≤α and |D1(t max) as second similar data, where D1(t max ) represents the current analyzed received signal strength, D2(t max ) represents the current analyzed audio signal strength, V(t max ) represents the current analyzed driving speed, and a represents a similarity comparison threshold, 0.1≤a≤0.2; the similarity data extraction unit analyzes the collected information type and the operation information to extract similar data, the first similar data represents historical data of the same type and similar audio data, and the second similar data represents historical data of similar vehicle driving data and similar microphone received sound data, which increases the diversity of system analysis and improves the analysis efficiency of the system for the vehicle-mounted audio data and the accuracy of the vehicle-mounted audio processing. It can be understood that the value of the similarity comparison threshold is not specifically limited in the embodiment, and can be freely set by those skilled in the art, as long as the extraction of similar data is met. The optimal value of the similarity comparison threshold is a=0.15.

[0059] Please continue to refer to Figure 4 As shown in FIG. 6, the audio data analysis module further includes:

[0060] An audio quality analysis unit is configured to analyze an audio quality parameter according to the vehicle carrying feature, the vehicle driving feature, the first similar data, and the type difference feature. The audio quality analysis unit is connected to the similarity data extraction unit.

[0061] Specifically, the audio quality analysis unit analyzes the audio quality parameter according to the vehicle carrying feature, the vehicle driving feature, the first similar data, and the type difference feature. The audio quality parameter is set as F(t), and the following equation is set: where U(k) represents a set of collection time numbers corresponding to the first similar data, and NU(k) represents the number of elements in the set of collection time numbers corresponding to the first similar data. The audio quality analysis unit analyzes the vehicle carrying feature, the vehicle driving feature, the first similar data, and the type difference feature to analyze the audio quality parameter, which represents the feature relationship among the vehicle-mounted data, the driving data, and the audio data, so as to realize comprehensive analysis of the carrying data, the driving data, and the audio data of the vehicle, thereby improving the analysis efficiency of the system for the vehicle-mounted audio data and the accuracy of the vehicle-mounted audio processing.

[0062] Please continue to refer to Figure 4 As shown in FIG. 6, the audio data analysis module further includes:

[0063] The historical characteristic analysis unit is used to adjust the analysis process of the audio quality parameter according to the historical fluctuation characteristic, so that the adjusted audio quality parameter is related to the fluctuation characteristic of the stored historical data, and the historical characteristic analysis unit is connected with the audio quality analysis unit.

[0064] Specifically, the historical characteristic analysis unit compares the historical fluctuation parameter with the fluctuation comparison threshold, and adjusts the analysis process of the audio quality parameter according to the comparison result. If R(j)≤r, the historical characteristic analysis unit determines that the historical fluctuation parameter meets the threshold, and does not adjust the analysis process of the audio quality parameter. If R(j)>r, the historical characteristic analysis unit determines that the historical fluctuation parameter does not meet the threshold, and adjusts the analysis process of the audio quality parameter. The adjusted audio quality parameter is F1(t), and F1(t) is set as F(t) / e R(j) / r-1 ; wherein r represents the fluctuation comparison threshold, and 0.05≤r≤0.15. The analysis of the historical fluctuation parameter by the historical characteristic analysis unit adjusts the analysis process of the audio quality parameter, realizes the correlation analysis between the historical data with the same type of collected information as the current analysis data and the current analysis data, thereby improving the analysis efficiency of the system for the audio data of the vehicle audio and improving the accuracy of the audio processing of the vehicle audio. It can be understood that the value of the fluctuation comparison threshold is not specifically limited in the embodiment, and a person skilled in the art can freely set it, as long as it can adjust the analysis process of the audio quality parameter. The best value of the fluctuation comparison threshold is r=0.1.

[0065] The similar data analysis unit is used to optimize the adjustment process of the audio quality parameter according to the second similar data, so that the optimized audio quality parameter is related to the data in the historical data which is similar to the current driving data and the received sound data. The similar data analysis unit is connected with the historical characteristic analysis unit.

[0066] Specifically, the similar data analysis unit in the embodiment counts the number of the second similar data with the same type of collected information as the currently analyzed data as a second similar number, compares the second similar number with the quantity comparison threshold, and optimizes the adjustment process of the audio quality parameter according to the comparison result. If N2 / NU2≤β, the similar data analysis unit determines that the second similar number does not meet the threshold, and does not optimize the adjustment process of the audio quality parameter. If N2 / NU2>β, the similar data analysis unit determines that the second similar number meets the threshold, and optimizes the adjustment process of the audio quality parameter. The optimized audio quality parameter is F2(t), and F2(t) is set as F1(t)×e 1-β; wherein, N2 represents a second similar quantity, NU2 represents a quantity of corresponding collection time numbers in the second similar data; the analysis of the second similar data by the similar data analysis unit is to analyze the second similar quantity, and to represent the quantity of data of the same collection data type as the current analysis data in the second similar data by the second similar data, so as to optimize the adjustment process of the audio quality parameter when the quantity of data of the same collection data type as the current analysis data in the second similar data is relatively large, thereby improving the analysis efficiency of the system on the audio data of the vehicle audio and improving the accuracy of the audio processing of the vehicle audio.

[0067] Please continue to refer to Figure 1 As shown in the figure, the vehicle audio audio processing system based on machine learning further comprises:

[0068] The audio audio regulation module is used to control the volume of the audio equipment of the vehicle according to the audio quality parameter, and the audio audio regulation module is connected with the audio data analysis module.

[0069] Specifically, the audio audio regulation module compares the audio quality parameter with the quality comparison threshold value, and controls the volume of the loudspeaker of the vehicle according to the comparison result. If F(t) < f1 or F(t) > f2, the audio audio regulation module controls the volume of the loudspeaker so that the collected audio equipment information satisfies D1(t) / [D2(t)+1]=F(t); if f1≤F(t)≤f2, the audio audio regulation module does not control the volume of the loudspeaker; wherein, f1 represents a first quality comparison threshold value, 0.7≤f1≤0.9, and f2 represents a second quality comparison threshold value, 1.6≤f2≤1.8. Through the analysis of the audio quality parameter by the audio audio regulation module, the volume of the audio equipment of the vehicle is controlled, so that the audio data of the vehicle audio meets the user's demand, and the volume of the audio is appropriately adjusted in different environments, thereby improving the analysis efficiency of the system on the audio data of the vehicle audio and improving the accuracy of the audio processing of the vehicle audio. It can be understood that the value of the quality comparison threshold value is not specifically limited in this embodiment, and those skilled in the art can freely set it, as long as the control of the volume of the audio equipment of the vehicle is met. The best value of the quality comparison threshold value is f1=0.8 and f2=1.7.

[0070] Obviously, the above embodiments of the present application are only examples for clearly illustrating the present application, and are not limitations on the embodiments of the present application. For those skilled in the art, other different forms of changes or variations can be made on the basis of the above description, and here it is impossible to enumerate all the embodiments, and any obvious changes or variations derived from the technical solutions of the present application still fall within the protection scope of the present application.

Claims

1. A machine learning based in-vehicle audio sound processing system, characterized by, The application relates to a vehicle audio quality control system, which comprises: a vehicle information collection module for collecting vehicle operation information and audio equipment information; an information storage and classification module for classifying the vehicle operation information to obtain collection information types and storing the collected vehicle operation information, audio equipment information and collection information types; the information storage and classification module calculates the average value of the opening degrees of the vehicle windows as the average opening degree of the windows and compares the average opening degree P(t) of the windows with the opening degree threshold value p, judges the collection information types according to the comparison result, if P(t) < p, the information storage and classification module judges that the collection information types are type one; if p <= P(t) < 1, the information storage and classification module judges that the collection information types are type two; if P(t) = 1, the information storage and classification module judges that the collection information types are type three; wherein i represents the window number and t represents the collection time number; a vehicle characteristic analysis module for analyzing vehicle carrying characteristics and vehicle driving characteristics according to the vehicle operation information and the audio equipment information; a stored data analysis module for analyzing audio parameters of each collection information type according to the stored vehicle operation information and audio equipment information and for analyzing type difference characteristics and historical fluctuation characteristics according to the stored vehicle operation information, audio equipment information and audio parameters; The stored data analysis module is provided with a type difference analysis unit, which is used to analyze the stored vehicle operation information and audio parameters of the audio equipment information of the collected information types of one, two and three respectively, set the audio parameters as L(j, t), set and analyze the type difference characteristics according to the audio parameters, set the type difference characteristics as C, and set wherein j represents the type parameter, U(j) represents a set of time numbers corresponding to each collected information type, NU(j) represents the number of elements in the set of time numbers corresponding to each collected information type, j=1 represents that the collected information type is one, j=2 represents that the collected information type is two, j=3 represents that the collected information type is three, D1(t) represents the received signal strength, and D2(t) represents the audio signal strength. the stored data analysis module is further provided with a historical fluctuation analysis unit for analyzing the historical fluctuation characteristics R(j) of the stored vehicle operation information and audio equipment information of the collection information types one, two and three respectively, an audio data analysis module for extracting similar data according to the collection information types, operation information and audio equipment information and for analyzing audio quality parameters according to the similar data, vehicle carrying characteristics, vehicle driving characteristics, type difference characteristics and historical fluctuation characteristics; The audio data analysis module is provided with a similar data extraction unit which extracts stored vehicle operation information and audio equipment information of which the collection information type is the same as the collection information type of the currently analyzed data and |D2(t max ) / D2(t|t∈U(1)∪U(2)∪U(3))-1|≤α as first similar data, and extracts stored vehicle operation information and audio equipment information of which |V(t max ) / V(t|t∈U(1)∪U(2)∪U(3))-1|≤α and |D1(t max ) / D1(t|t∈U(1)∪U(2)∪U(3))-1|≤α as second similar data, wherein D1(t max ) represents the currently analyzed received signal strength, D2(t max ) represents the currently analyzed audio signal strength, V(t max ) represents the currently analyzed driving speed, and α represents a similarity comparison threshold value. The audio data analysis module is also provided with an audio quality analysis unit, which is used to analyze the audio quality parameter according to the vehicle carrying feature A(t), the vehicle driving feature B(t), the first similar data and the type difference feature C, so as to obtain the audio quality parameter F(t), Wherein, U(k) represents a set of collection time numbers corresponding to the first similar data, NU(k) represents the number of elements in the set of collection time numbers corresponding to the first similar data, A(t) represents the vehicle carrying feature, and B(t) represents the vehicle driving feature. the audio data analysis module compares the historical fluctuation parameters with fluctuation comparison threshold values and adjusts the analysis process of the audio quality parameters when the historical fluctuation parameters do not conform to the threshold values; the audio data analysis module counts the number of second similar data of the same collection information type as the currently analyzed data as a second similar number and compares the second similar number with a number comparison threshold value, and optimizes the adjustment process of the audio quality parameters when the second similar number conforms to the threshold value; a sound audio control module for controlling the volume of the audio equipment of the vehicle according to the audio quality parameters.

2. The machine learning based in-vehicle sound system audio processing system of claim 1, wherein, the vehicle characteristic analysis module is provided with a carrying characteristic analysis unit for analyzing the vehicle carrying characteristics according to the number of passengers N1(t), the received signal strength D1(t) and the audio signal strength D2(t) when the collection information types are type two and type three to obtain the vehicle carrying characteristics A(t), A(t) = [D1(t) / (D2(t)+1)]xlnN1(t). The vehicle feature analysis module is also provided with a driving feature analysis unit, which is used to analyze the vehicle driving features according to the driving speed V(t), the received signal strength D1(t) and the audio signal strength D2(t) to obtain the vehicle driving feature B(t), B(t) = [D1(t) / (D2(t)+1)]×lgV(t).

3. The machine learning based in-vehicle sound system audio processing system of claim 2, wherein, The audio data analysis module is also provided with a historical characteristic analysis unit, which is used to compare the historical fluctuation parameter R(j) with the fluctuation ratio threshold value r, and adjust the analysis process of the audio quality parameter when the historical fluctuation parameter does not conform to the threshold value. The adjusted audio quality parameter is F1(t), F1(t)=F(t) / e R(j) / r-1 .

4. The machine learning based in-vehicle sound system audio processing system of claim 3, wherein, The audio data analysis module is also provided with a similar data analysis unit, which is used to count the number of second similar data whose collection information type is the same as the collection information type of the current analyzed data as a second similar number, and compare the second similar number N2 with the number ratio threshold β, when the second similar number meets the threshold, the adjustment process of the audio quality parameter is optimized, and the optimized audio quality parameter is F2(t), F2(t)=F1(t)×e 1-β .

5. The machine learning based in-vehicle sound system audio processing system of claim 4, wherein, The audio frequency control module compares the audio quality parameter F(t) with the quality ratio threshold values f1 and f2, and controls the speaker volume of the vehicle according to the comparison result, so that the collected audio equipment information satisfies D1(t) / [D2(t)+1] = F(t).

Images