A method and system for pushing audio based on scenarios for a vehicle

Abstract

The application discloses a method and system for pushing audio based on a scene of a vehicle, the method comprising: calculating a correlation score between a user scene feature and an audio content feature; calculating a correlation score between the user scene feature and navigation data features; calculating a correlation score between the user scene feature and calendar data features; calculating a correlation score between the user scene feature and weather data features; taking the corresponding user scene feature with the highest correlation score as the final user scene feature, and pushing audio to the user according to the audio content feature corresponding to the user scene feature with the highest correlation score, or calculating a comprehensive correlation score, finding out the navigation data features, the calendar data features and the weather data features corresponding thereto, and respectively pushing audio corresponding to the navigation data features, the calendar data features and the weather data features.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of automobile scene-based audio pushing, and more particularly to a method and system for automobile scene-based audio pushing. BACKGROUND

[0002] Currently, immersive driving experiences created by automobile intelligence and emotionalization are highly valued by various automobile manufacturers, including AI voice assistants, human-computer voice communication, active greetings, intelligent scene recommendation systems, and other new technologies, which provide users with rich emotional experiences. In driving scenarios, audio has the highest usage rate, and vehicle-mounted audio playback resources, audio App categories, audio models, and brand sound systems are becoming increasingly mature, but methods and systems for automatically pushing audio based on driving scenarios have not yet been established. SUMMARY

[0003] To solve the above technical problems, the present application provides a method for automobile scene-based audio pushing, comprising:

[0004] Obtaining a user scene feature dataset, an audio content feature dataset, and calculating the correlation score between the user scene features and the audio content features;

[0005] Obtaining a navigation data feature dataset and calculating the correlation score between the user scene features and the navigation data features based on the user scene feature dataset;

[0006] Obtaining a calendar data feature dataset and calculating the correlation score between the user scene features and the calendar data features based on the user scene feature dataset;

[0007] Obtaining a weather data feature dataset and calculating the correlation score between the user scene features and the weather data features based on the user scene feature dataset;

[0008] Ranking the correlation scores between the user scene features and the navigation data features, the correlation scores between the user scene features and the calendar data features, and the correlation scores between the user scene features and the weather data features, selecting the user scene features with the highest correlation scores as the final user scene features, and pushing audio corresponding to the user scene features with the highest correlation scores based on the audio content features with the highest correlation scores between the user scene features and the audio content features, or calculating a comprehensive correlation score, finding the navigation data features, calendar data features, and weather data features corresponding to the highest correlation score, and pushing audio corresponding to the navigation data features, calendar data features, and weather data features, respectively.

[0009] Further, the calculation of the correlation score between the user scene features and the audio content features is specifically as follows:

[0010]

[0011] Among them, f(S i , A i ) is the user scenario feature S i and audio content feature A i The correlation score between them, theta is the user scenario feature S i With audio content feature A i The angle between them, S i is the i-th user scenario feature, A i is the i-th audio content feature, ||S i || is the norm of the i-th user scenario feature, ||A i || is the norm of the i-th audio content feature.

[0012] Furthermore, the calculation of the correlation score between the user scenario feature and the navigation data feature is specifically as follows:

[0013]

[0014] Among them, m(S i , N i ) is the user scenario feature S i With navigation data feature N i The correlation score between s is the average value of user scenario features, N i is the i-th navigation data feature, Y N is the average value of the navigation data features.

[0015] Furthermore, the calculation of the correlation score between the user scenario feature and the calendar data feature is specifically as follows:

[0016]

[0017] Among them, n(S i , C i ) is the user scenario feature S i With calendar data feature C i The correlation score between i is the i-th calendar data feature, Y C is the mean value of the calendar data characteristics, and N′ is the number of samples.

[0018] Furthermore, the calculation of the correlation score between the user scenario feature and the weather data feature is specifically as follows:

[0019]

[0020] Among them, o(S i , Wi ) is a correlation score between the user scenario feature S i and the weather data feature W i , W i is the weather data feature, p(S i , W i ) is a joint probability distribution between the user scenario feature S i and the weather data feature W i , p(S i ) is an edge probability distribution of the user scenario feature S i , and p(W i ) is an edge probability distribution of the weather data feature W i .

[0021] Further, the comprehensive correlation score is calculated as follows:

[0022]

[0023] wherein r is the comprehensive correlation score, q1, q2 and q3 are weights.

[0024] The application further provides a system for pushing audio based on scenarios for a vehicle, comprising:

[0025] a module for calculating a correlation score between a user scenario feature and an audio content feature, configured to obtain a user scenario feature dataset and an audio content feature dataset, and calculate a correlation score between the user scenario feature and the audio content feature;

[0026] a module for calculating a correlation score between a user scenario feature and a navigation data feature, configured to obtain a navigation data feature dataset, and calculate a correlation score between the user scenario feature and the navigation data feature according to the user scenario feature dataset;

[0027] a module for calculating a correlation score between a user scenario feature and a calendar data feature, configured to obtain a calendar data feature dataset, and calculate a correlation score between the user scenario feature and the calendar data feature according to the user scenario feature dataset;

[0028] a module for calculating a correlation score between a user scenario feature and a weather data feature, configured to obtain a weather data feature dataset, and calculate a correlation score between the user scenario feature and the weather data feature according to the user scenario feature dataset;

[0029] The push module sorts the correlation scores between the user scene features and the navigation data features, the correlation scores between the user scene features and the calendar data features, and the correlation scores between the user scene features and the weather data features, takes the user scene feature corresponding to the highest correlation score as the final user scene feature, and pushes audio corresponding to the audio content feature with the highest correlation score between the user scene feature corresponding thereto and the audio content feature to the user, or calculates a comprehensive correlation score, finds out the navigation data feature, the calendar data feature, and the weather data feature corresponding thereto, and respectively pushes audio corresponding to the navigation data feature, the calendar data feature, and the weather data feature.

[0030] Further, the calculation of the correlation score between the user scene feature and the audio content feature is specifically:

[0031]

[0032] wherein f(S i , A i ) is the correlation score between the user scene feature S i and the audio content feature A i , theta is the included angle between the user scene feature S i and the audio content feature A i , S i i is the i-th user scene feature, A i i is the i-th audio content feature, ||S i || is the norm of the i-th user scene feature, and ||A i || is the norm of the i-th audio content feature.

[0033] Further, the calculation of the correlation score between the user scene feature and the navigation data feature is specifically:

[0034]

[0035] wherein m(S i , N i ) is the correlation score between the user scene feature S i and the navigation data feature N i , Y s is the average value of the user scene feature, N i i is the i-th navigation data feature, and Y N is the average value of the navigation data feature.

[0036] Further, the calculation of the correlation score between the user scene feature and the calendar data feature is specifically:

[0037]

[0038] wherein, n(S i , C i ) is the correlation score between the user scene feature S i and the calendar data feature C i , C i is the i-th calendar data feature, Y C is the average value of the calendar data feature, and N' is the sample quantity.

[0039] Further, the calculation of the correlation score between the user scene feature and the weather data feature is specifically:

[0040]

[0041] wherein, o(S i , W i ) is the correlation score between the user scene feature S i and the weather data feature W i , W i is the weather data feature, p(S i , W i ) is the joint probability distribution between the user scene feature S i and the weather data feature W i , p(S i ) is the marginal probability distribution of the user scene feature S i , and p(W i ) is the marginal probability distribution of the weather data feature W i .

[0042] Overall, compared with the prior art, the above technical scheme conceived by the present application has the following beneficial effects:

[0043] The present application can complete intelligent scene recommendation of audio based on the user driving scene and according to the user listening behavior. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is the flow chart of the method of the embodiment 1 of the present application;

[0045] Figure 2 is the structural diagram of the system of the embodiment 2 of the present application;

[0046] Figure 3 is the hardware system structural diagram of the embodiment 5 of the present application;

[0047] Figure 4 is the method flow chart of the embodiment 6 of the present application. DETAILED DESCRIPTION

[0048] For better understanding of the above technical solutions, the above technical solutions will be described in detail below in combination with the drawings of the specification and specific embodiments.

[0049] The method provided by the application can be implemented in a terminal environment, which can include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the embodiments below.

[0050] The processor can include one or more processing cores. The processor connects various parts in the entire terminal through various interfaces and lines, executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0051] The storage medium can include random access memory (RAM) and read-only memory (ROM). The storage medium can be used to store instructions, programs, codes, code sets or instructions.

[0052] The display screen is used to display the user interface of each application.

[0053] All subscripts in the formula of the application are only used to distinguish parameters and have no actual meaning.

[0054] In addition, those skilled in the art can understand that the structure of the terminal described above does not constitute a limitation on the terminal, and the terminal can include more or fewer components, or combine certain components, or different component arrangements. For example, the terminal also includes radio frequency circuit, input unit, sensor, audio circuit, power supply and other components, which will not be described here.

[0055] Embodiment 1

[0056] As shown in Figure 1 The embodiment of the application provides a method for pushing audio based on scene for a car, which includes:

[0057] Step 101, obtaining a user scene feature data set, an audio content feature data set, and calculating the correlation score between the user scene feature and the audio content feature;

[0058] Specifically, the calculation of the correlation score between the user scene feature and the audio content feature is specifically:

[0059]

[0060] Wherein, f(S i , A i) is the user scenario feature S i and audio content feature A i The correlation score between them, theta is the user scenario feature S i and audio content feature A i The angle between them, Si is the i-th user scene feature, A i is the i-th audio content feature, ||S i || is the norm of the i-th user scenario feature, ||A i || is the norm of the i-th audio content feature.

[0061] Step 102: Obtain a navigation data feature dataset, and calculate a correlation score between user scenario features and navigation data features based on the user scenario feature dataset;

[0062] Specifically, the calculation of the correlation score between the user scenario feature and the navigation data feature is as follows:

[0063]

[0064] Among them, m(S i , N i ) is the user scenario feature S i With navigation data feature N i The correlation score between s is the average value of user scenario features, N i is the i-th navigation data feature, Y N is the average value of the navigation data features.

[0065] Step 103: Acquire a calendar data feature dataset, and calculate a correlation score between the user scenario feature and the calendar data feature based on the user scenario feature dataset;

[0066] Specifically, the calculation of the correlation score between the user scenario feature and the calendar data feature is as follows:

[0067]

[0068] Among them, n(S i , C i ) is the user scenario feature S i With calendar data feature C i The correlation score between i is the i-th calendar data feature, Y C is the mean value of the calendar data characteristics, and N′ is the number of samples.

[0069] Step 104, obtaining a weather data feature dataset, and calculating a correlation score between the user scene feature and the weather data feature according to the user scene feature dataset;

[0070] Specifically, the calculation of the correlation score between the user scene feature and the weather data feature is specifically:

[0071]

[0072] Wherein, o(S i , W i ) is the correlation score between the user scene feature S i and the weather data feature W i , W i is the weather data feature, p(S i , W i ) is the joint probability distribution between the user scene feature S i and the weather data feature W i , p(S i ) is the marginal probability distribution of the user scene feature S i , and p(W i ) is the marginal probability distribution of the weather data feature W i .

[0073] Step 105, ranking the correlation scores between the user scene feature and the navigation data feature, the correlation scores between the user scene feature and the calendar data feature, and the correlation scores between the user scene feature and the weather data feature, taking the user scene feature corresponding to the highest correlation score as the final user scene feature, and pushing the audio corresponding to the audio content feature with the highest correlation score between the user scene feature and the audio content feature to the user, or calculating a comprehensive correlation score, finding the navigation data feature, the calendar data feature and the weather data feature corresponding thereto, and pushing the audio corresponding to the navigation data feature, the calendar data feature and the weather data feature respectively.

[0074] Specifically, the calculation of the comprehensive correlation score is specifically:

[0075]

[0076] Wherein, r is the comprehensive correlation score, and q1, q2 and q3 are weights.

[0077] Embodiment 2

[0078] As shown in Figure 2 , the embodiment of the application also provides a system for pushing audio based on scene for a car, comprising:

[0079] The computing user scene feature and audio content feature correlation score module is configured to obtain a user scene feature dataset and an audio content feature dataset, and calculate a correlation score between the user scene feature and the audio content feature.

[0080] Specifically, the computing user scene feature and audio content feature correlation score is specifically:

[0081]

[0082] wherein, f (S i , A i ) is a correlation score between the user scene feature S i and the audio content feature A i , theta is an included angle between the user scene feature S i and the audio content feature A i , S i is an i-th user scene feature, A i is an i-th audio content feature, ||S i || is a norm of the i-th user scene feature, and ||A i || is a norm of the i-th audio content feature.

[0083] The computing user scene feature and navigation data feature correlation score module is configured to obtain a navigation data feature dataset, and calculate a correlation score between the user scene feature and the navigation data feature according to the user scene feature dataset.

[0084] Specifically, the computing user scene feature and navigation data feature correlation score is specifically:

[0085]

[0086] wherein, m (S i , N i ) is a correlation score between the user scene feature S i and the navigation data feature N i , Y s is an average value of the user scene feature, N i is an i-th navigation data feature, and Y N is an average value of the navigation data feature.

[0087] The computing user scene feature and calendar data feature correlation score module is configured to obtain a calendar data feature dataset, and calculate a correlation score between the user scene feature and the calendar data feature according to the user scene feature dataset.

[0088] Specifically, the computing user scene feature and calendar data feature correlation score is specifically:

[0089]

[0090] wherein, n(S i , C i ) is the correlation score between the user scenario feature S i and the calendar data feature C i , C i is the ith calendar data feature, Y C is the average value of the calendar data features, and N' is the sample quantity.

[0091] The computing user scenario feature and weather data feature correlation score module is configured to obtain a weather data feature dataset and calculate a correlation score between the user scenario feature and the weather data feature according to the user scenario feature dataset.

[0092] Specifically, the correlation score between the user scenario feature and the weather data feature is specifically:

[0093]

[0094] wherein, o(S i , W i ) is the correlation score between the user scenario feature S i and the weather data feature W i , W i is the weather data feature, p(S i , W i ) is the joint probability distribution between the user scenario feature S i and the weather data feature W i , p(S i ) is the marginal probability distribution of the user scenario feature S i , and p(W i ) is the marginal probability distribution of the weather data feature W i .

[0095] The pushing module is configured to sort the correlation scores between the user scenario feature and the navigation data feature, the correlation scores between the user scenario feature and the calendar data feature, and the correlation scores between the user scenario feature and the weather data feature, take the user scenario feature corresponding to the highest correlation score as the final user scenario feature, and push the audio corresponding to the audio content feature with the highest correlation score between the user scenario feature and the audio content feature to the user, or calculate a comprehensive correlation score, find out the navigation data feature, the calendar data feature, and the weather data feature corresponding to the highest correlation score, and push the audio corresponding to the navigation data feature, the calendar data feature, and the weather data feature, respectively.

[0096] Specifically, the calculation of the comprehensive correlation score is specifically as follows:

[0097]

[0098] Wherein, r is the comprehensive correlation score, q1, q2 and q3 are weights.

[0099] Embodiment 3

[0100] The embodiment of the application further provides a storage medium which stores a plurality of instructions for implementing the method for pushing audio based on a scene for a vehicle.

[0101] Optionally, in the embodiment, the storage medium can be located in any one of computer terminals in a computer terminal group in a computer network, or in any one of mobile terminals in a mobile terminal group.

[0102] Optionally, in the embodiment, the storage medium is configured to store program code for performing the following steps: step 101, obtaining a user scene feature data set, an audio content feature data set, and calculating a correlation score between the user scene feature and the audio content feature;

[0103] Specifically, the calculation of the correlation score between the user scene feature and the audio content feature is specifically as follows:

[0104]

[0105] Wherein, f(S i , A i ) is the correlation score between the user scene feature S i and the audio content feature A i , theta is the included angle between the user scene feature S i and the audio content feature A i , S i is the i-th user scene feature, A i is the i-th audio content feature, ||S i || is the norm of the i-th user scene feature, and ||A i || is the norm of the i-th audio content feature.

[0106] Step 102, obtaining a navigation data feature data set, and calculating a correlation score between the user scene feature and the navigation data feature according to the user scene feature data set;

[0107] Specifically, the calculation of the correlation score between the user scene feature and the navigation data feature is specifically as follows:

[0108]

[0109] wherein m(S i , N i ) is the correlation score between the user scenario feature S i and the navigation data feature N i , Y s is the average of the user scenario features, N i is the i-th navigation data feature, and Y N is the average of the navigation data features.

[0110] Step 103, obtaining a calendar data feature dataset, and calculating a correlation score between the user scenario feature and the calendar data feature according to the user scenario feature dataset;

[0111] Specifically, the calculation of the correlation score between the user scenario feature and the calendar data feature is specifically:

[0112]

[0113] wherein n(S i , C i ) is the correlation score between the user scenario feature S i and the calendar data feature C i , C i is the i-th calendar data feature, Y C is the average of the calendar data features, and N' is the sample quantity.

[0114] Step 104, obtaining a weather data feature dataset, and calculating a correlation score between the user scenario feature and the weather data feature according to the user scenario feature dataset;

[0115] Specifically, the calculation of the correlation score between the user scenario feature and the weather data feature is specifically:

[0116]

[0117] wherein o(S i , W i ) is the correlation score between the user scenario feature S i and the weather data feature W i , W i is the weather data feature, p(S i , W i ) is the joint probability distribution between the user scenario feature S i and the weather data feature W i , p(S i ) is the marginal probability distribution of the user scenario feature S i , and p(W i ) is the marginal probability distribution of the weather data feature Wi an edge probability distribution.

[0118] Step 105, ranking the correlation degree score between the user scene feature and the navigation data feature, the correlation degree score between the user scene feature and the calendar data feature, and the correlation degree score between the user scene feature and the weather data feature, taking the user scene feature corresponding to the highest correlation degree score as the final user scene feature, and pushing the audio corresponding to the audio content feature with the highest correlation degree score between the audio content feature and the user scene feature to the user, or calculating the comprehensive correlation degree score, finding the navigation data feature, the calendar data feature, and the weather data feature corresponding to the highest comprehensive correlation degree score, and pushing the audio corresponding to the navigation data feature, the calendar data feature, and the weather data feature to the user.

[0119] Specifically, the comprehensive correlation degree score is specifically calculated as follows:

[0120]

[0121] Wherein, r is the comprehensive correlation degree score, q1, q2, and q3 are weights.

[0122] Embodiment 4

[0123] The embodiment of the application also provides an electronic device, which comprises a processor and a storage medium connected with the processor, and the storage medium stores a plurality of instructions which can be loaded and executed by the processor so that the processor can execute the method for pushing audio based on a scene of a vehicle.

[0124] Specifically, the electronic device of the embodiment can be a computer terminal, which can comprise one or more processors and a storage medium.

[0125] The storage medium can be used to store software programs and modules, such as the method for pushing audio based on a scene of a vehicle in the embodiment of the application, corresponding program instructions / modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, the above-mentioned method for pushing audio based on a scene of a vehicle is implemented. The storage medium can comprise a high-speed random storage medium and can also comprise a non-volatile storage medium, such as one or more magnetic storage systems, flash memories, or other non-volatile solid-state storage media. In some examples, the storage medium can further comprise storage media remotely arranged relative to the processor, and these remote storage media can be connected to the terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0126] The processor can store the information and application stored in the storage medium through the transmission system to perform the following steps: step 101, obtaining a user scene feature dataset, an audio content feature dataset, and calculating a correlation score between the user scene feature and the audio content feature;

[0127] Specifically, the calculation of the correlation score between the user scene feature and the audio content feature is specifically:

[0128]

[0129] Wherein, f(S i , A i ) is the correlation score between the user scene feature S i and the audio content feature A i , theta is the angle between the user scene feature S i and the audio content feature A i , S i is the i-th user scene feature, A i is the i-th audio content feature, ||S i || is the norm of the i-th user scene feature, and ||A i || is the norm of the i-th audio content feature.

[0130] Step 102, obtaining a navigation data feature dataset, and calculating a correlation score between the user scene feature and the navigation data feature according to the user scene feature dataset;

[0131] Specifically, the calculation of the correlation score between the user scene feature and the navigation data feature is specifically:

[0132]

[0133] Wherein, m(S i , N i ) is the correlation score between the user scene feature S i and the navigation data feature N i , Y s is the average value of the user scene feature, N i is the i-th navigation data feature, and Y N is the average value of the navigation data feature.

[0134] Step 103, obtaining a calendar data feature dataset, and calculating a correlation score between the user scene feature and the calendar data feature according to the user scene feature dataset;

[0135] Specifically, the calculation of the correlation score between the user scene feature and the calendar data feature is specifically:

[0136]

[0137] wherein, n(S i , C i ) is the correlation score between the user scenario feature S i and the calendar data feature C i , C i is the ith calendar data feature, Y C is the average value of the calendar data features, and N' is the sample quantity.

[0138] In step 104, a weather data feature dataset is obtained, and a correlation score between the user scenario feature and the weather data feature is calculated according to the user scenario feature dataset.

[0139] Specifically, the calculation of the correlation score between the user scenario feature and the weather data feature is specifically as follows:

[0140]

[0141] wherein, o(S i , W i ) is the correlation score between the user scenario feature S i and the weather data feature W i , W i is the weather data feature, p(S i , W i ) is the joint probability distribution between the user scenario feature S i and the weather data feature W i , p(S i ) is the marginal probability distribution of the user scenario feature S i , and p(W i ) is the marginal probability distribution of the weather data feature W i .

[0142] In step 105, the correlation scores between the user scenario feature and the navigation data feature, the correlation scores between the user scenario feature and the calendar data feature, and the correlation scores between the user scenario feature and the weather data feature are sorted, the user scenario feature corresponding to the highest correlation score is taken as the final user scenario feature, and the audio content feature corresponding to the highest correlation score between the user scenario feature and the audio content feature is pushed to the user as the audio, or a comprehensive correlation score is calculated, the navigation data feature, the calendar data feature, and the weather data feature corresponding to the highest correlation score are found out, and audios corresponding to the navigation data feature, the calendar data feature, and the weather data feature are respectively pushed.

[0143] Specifically, the calculation of the comprehensive correlation score is specifically as follows:

[0144]

[0145] Wherein, r is the comprehensive correlation score, q1, q2 and q3 are weights.

[0146] Example 5

[0147] As Figure 3 shown, the functional modules of the audio control mechanism of the present application scheme include: a vehicle-mounted entertainment audio system, a cloud data management platform, an audio management module, a system-level chip, a digital signal processing module, an audio micro control unit.

[0148] The functions of the vehicle-mounted entertainment audio system: interconnection between vehicle-mounted data and the cloud data management platform, audio application of online multimedia such as music, radio, and audio books, data application of real-time scenes such as navigation, calendar, and weather, and account application of collecting and entering user information such as personal center and log.

[0149] The functions of the cloud data management platform: the account data analysis module collects and analyzes data such as user gender, age, log, frequently used applications, and driving behavior, and manages them, the listening behavior analysis module collects and analyzes data such as audio content listened by the user, frequently listened audio types, and usage time / length, and manages them, and the driving scene acquisition module collects and analyzes data such as navigation, calendar, and weather, and manages them, and the driving scene query module matches the parameters.

[0150] The functions of the audio management module: sending analog audio signals to the digital signal processing module, and sending digital audio signals to the audio micro control unit.

[0151] The functions of the system-level chip: the system-level chip needs an analog-to-digital converter to convert analog signals into digital signals.

[0152] The functions of the audio micro control unit: classifying digital audio signals, retaining the current playing audio digital signal, removing other audio signals, and sending the results to the independent power amplifier through the A2B adapter line.

[0153] Example 6

[0154] As Figure 4 shown, the method of using the hardware system of example 5 includes:

[0155] Step one, the vehicle-mounted entertainment audio system needs to be connected to the network, allowing users to have their own vehicle networking account and deploy corresponding data resources to the cloud data management platform, and upload, store, and download data.

[0156] Step two, the car needs to have a car-mounted entertainment video system, which can carry audio applications such as music, radio, online multimedia audio books, real-time scene data applications such as navigation, calendar, weather, account applications for collecting and entering user information such as personal center and log;

[0157] Step three, the car-mounted entertainment video system needs to store the personal information set by the user to the cloud data management platform and bind it with its own account;

[0158] Step four, any car-mounted entertainment video system needs to obtain the cloud data storage information of personal information from the cloud data management platform after a certain user who has set personal information logs in;

[0159] Step five, the cloud data management platform needs functions including parameter model construction module and driving scene acquisition module;

[0160] Step six, the parameter model construction module needs to integrate the data of the account data analysis module and the listening behavior analysis module;

[0161] Step seven, the account data analysis module needs to collect and analyze and manage data such as user gender, age, log, frequently used applications, and driving behavior, and the listening behavior analysis module needs to collect and analyze and manage data such as audio content analysis module, frequently listened audio type analysis module, and use time / length analysis module;

[0162] Step eight, the audio content analysis module needs to analyze data such as user listening audio text, lyrics, and comments, and classify audio with high similarity into categories, and give different dimensions of labels set in advance, including first-level and second-level labels, first-level labels including types such as emotion, feeling, and dream, and second-level labels including types such as happy, upset, and melancholy under the emotion label, types such as family, friendship, and love under the feeling label, and types such as travel, study abroad, and getting rich under the dream label;

[0163] Step nine, the frequently listened audio type analysis module needs to collect and analyze and manage data such as music genre / singer, book type / author, news type / platform, and radio type / station, and regional location of audio;

[0164] Step ten, the driving scene acquisition module needs to collect and analyze and manage data such as navigation, calendar, and weather data analysis module;

[0165] Step eleven, the navigation data analysis module needs to obtain real-time positioning data, route data, and distance data of the user;

[0166] Step twelve, the calendar data analysis module needs to obtain real-time holiday data, birthday data, and schedule data of the user;

[0167] Step thirteen, the weather data analysis module needs to obtain the user's real-time weather, temperature, wind direction and other data;

[0168] Step fourteen, the above data is needed to match the driving scene query module data with the parameter model construction module data;

[0169] Step fifteen, the listening behavior analysis module needs to continuously mimic the magic cube learning of the user, based on the parameter model and the driving scene, to obtain the matching parameters of the audio playback;

[0170] Step sixteen, group the parameters according to different forms, set the iteration test period and times, and continuously verify the same scheme by generating an iteration test model;

[0171] Step seventeen, the system decides to select the push scheme according to the verification result, and automatically pushes the matching audio content to the user according to the result;

[0172] Step eighteen, the vehicle entertainment audio system needs an audio management module to send the audio analog signal to the digital signal processing module, and send the audio digital signal to the audio management micro control unit;

[0173] Step nineteen, the automotive system chip needs an analog-to-digital converter to convert analog signals to digital signals;

[0174] Step twenty, the car needs a digital signal processing module to convert analog signals to digital signals;

[0175] Step twenty-one, the car needs an audio management micro control unit to classify audio data, retain the current playing audio data, remove other audio data, and send the result to the independent power amplifier through the A2B adapter line;

[0176] Step twenty-two, the car needs an independent power amplifier, the audio management module sends the audio push result to the independent power amplifier, controls the loudspeaker, and drives the loudspeaker after simulating audio amplification;

[0177] Step twenty-three, the car needs a combination loudspeaker to carry out audio playback functions, which can realize the effect of automatically playing audio with driving scenes;

[0178] Step twenty-four, the vehicle entertainment audio system needs OTA function, which can continuously iterate and optimize the audio automatic push model.

[0179] The above embodiment numbers of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0180] In the above-mentioned embodiments of the present application, the description of each embodiment focuses on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0181] In several embodiments provided by the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the embodiments described above are merely schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, and can be electrical or other forms.

[0182] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0183] In addition, each functional unit in each embodiment of the present application can be integrated in a processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or software functional unit.

[0184] When the integrated unit is realized in the form of software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the essential part or all or part of the technical solutions which make contributions to the prior art can be embodied in the form of software product, and the computer software product is stored in a storage medium, including a plurality of instructions for making a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and various program code storage media.

[0185] Obviously, the above embodiments are merely example for clearly illustrating but not limitation to the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments need not and can not be enumerated. The obvious changes or variations derived from the above description are still within the protection scope of the present application.

Claims

1. A method for pushing audio based on a scene in a car, characterized in that: include: Obtain a user scenario feature dataset and an audio content feature dataset, and calculate the correlation score between the user scenario features and the audio content features; Obtaining a navigation data feature dataset, and calculating a correlation score between user scenario features and navigation data features based on the user scenario feature dataset; Obtaining a calendar data feature dataset, and calculating a correlation score between user scenario features and calendar data features based on the user scenario feature dataset; Obtaining a weather data feature dataset, and calculating a correlation score between user scenario features and weather data features based on the user scenario feature dataset; The correlation scores between the user scenario features and the navigation data features, the correlation scores between the user scenario features and the calendar data features, and the correlation scores between the user scenario features and the weather data features are sorted, and the corresponding user scenario feature with the highest correlation score is used as the final user scenario feature. Based on the audio content feature with the highest correlation score between the corresponding user scenario feature and the audio content feature, audio is pushed to the user, or a comprehensive correlation score is calculated to find the corresponding navigation data features, calendar data features, and weather data features, and the audio corresponding to the navigation data features, calendar data features, and weather data features are pushed respectively.

2. The method for pushing audio based on a scene in a car according to claim 1, characterized in that: The calculation of the correlation score between the user scenario feature and the audio content feature is specifically as follows: Among them, f(S i , A i ) is the user scenario feature S i With audio content feature A i Theta is the correlation score between the user scene feature Si and the audio content feature A i The angle between them, S i is the i-th user scenario feature, A i is the i-th audio content feature, ||S i || is the norm of the i-th user scenario feature, ||A i || is the norm of the i-th audio content feature.

3. The method for pushing audio based on a scene in a car according to claim 2, characterized in that: The calculation of the correlation score between the user scenario feature and the navigation data feature is specifically as follows: Among them, m(S i , N i ) is the user scenario feature S i With navigation data feature N i The correlation score between s is the average value of user scenario features, N i is the i-th navigation data feature, Y N is the average value of the navigation data features.

4. The method for pushing audio based on a scene in a car according to claim 3, characterized in that: The calculation of the correlation score between the user scenario feature and the calendar data feature is specifically as follows: Among them, n(S i , C i ) is the user scenario feature S i With calendar data feature C i The correlation score between i is the i-th calendar data feature, Y C is the mean value of the calendar data characteristics, and N′ is the number of samples.

5. The method for pushing audio based on a scene in a car according to claim 4, characterized in that: The calculation of the correlation score between the user scenario feature and the weather data feature is specifically as follows: Among them, o(S i , W i ) is the user scenario feature S i With weather data features W i The correlation score between i is the weather data feature, p(S i , W i ) is the user scenario feature S i With weather data features W i The joint probability distribution between i ) is the user scenario feature S i The marginal probability distribution, p(W i ) is the weather data feature W i The marginal probability distribution of .

6. The method for pushing audio based on a scene in a car according to claim 5, characterized in that: The calculation of the comprehensive relevance score is as follows: Among them, r is the comprehensive relevance score, and q1, q2 and q3 are weights.

7. A system for pushing audio based on scenes in a car, characterized in that: include: A module for calculating the correlation score between user scenario features and audio content features is used to obtain a user scenario feature dataset and an audio content feature dataset, and calculate the correlation score between the user scenario features and the audio content features; A module for calculating the correlation score between user scenario features and navigation data features is used to obtain a navigation data feature dataset and calculate the correlation score between the user scenario features and the navigation data features based on the user scenario feature dataset; A module for calculating the correlation score between user scenario features and calendar data features is used to obtain a calendar data feature dataset and calculate the correlation score between the user scenario features and the calendar data features based on the user scenario feature dataset; A module for calculating the correlation score between user scenario features and weather data features is used to obtain a weather data feature dataset and calculate the correlation score between the user scenario features and the weather data features based on the user scenario feature dataset; The push module is used to sort the correlation scores between user scenario features and navigation data features, the correlation scores between user scenario features and calendar data features, and the correlation scores between user scenario features and weather data features, and use the corresponding user scenario feature with the highest correlation score as the final user scenario feature. Based on the audio content feature with the highest correlation score between the corresponding user scenario feature and the audio content feature, the audio is pushed to the user, or the comprehensive correlation score is calculated to find the corresponding navigation data features, calendar data features, and weather data features, and the audio corresponding to the navigation data features, calendar data features, and weather data features are pushed respectively.

8. The system for pushing audio based on a scene in a car as claimed in claim 7, characterized in that: The calculation of the correlation score between the user scenario feature and the audio content feature is specifically as follows: Among them, f(S i , A i ) is the user scenario feature S i With audio content feature A i The correlation score between them, theta is the user scenario feature S i With audio content feature A i The angle between them, S i is the i-th user scenario feature, A i is the i-th audio content feature, ||S i || is the norm of the i-th user scenario feature, ||A i || is the norm of the i-th audio content feature.

9. The system for pushing audio based on a scene in a car as claimed in claim 8, characterized in that: The calculation of the correlation score between the user scenario feature and the navigation data feature is specifically as follows: Among them, m(S i , N i ) is the user scenario feature S i With navigation data feature N i The correlation score between s is the average value of user scenario features, N i is the i-th navigation data feature, Y N is the average value of the navigation data features.

10. The system for pushing audio based on a scene in a car as claimed in claim 9, characterized in that: The calculation of the correlation score between the user scenario feature and the calendar data feature is specifically as follows: Among them, n(S i , C i ) is the user scenario feature S i With calendar data feature C i The correlation score between i is the i-th calendar data feature, Y C is the mean value of the calendar data characteristics, and N′ is the number of samples.

11. The system for pushing audio based on a scene in a car as claimed in claim 10, characterized in that: The calculation of the correlation score between the user scenario feature and the weather data feature is specifically as follows: Among them, o(S i , W i ) is the user scenario feature S i With weather data features W i The correlation score between i is the weather data feature, p(S i , W i ) is the user scenario feature S i With weather data features W i The joint probability distribution between i ) is the user scenario feature S i The marginal probability distribution, p(W i ) is the weather data feature W i The marginal probability distribution of .

Images