Music push method, device, electronic device and readable storage medium

Abstract

An embodiment of the present invention provides a music push method, device, electronic device, and readable storage medium. The method includes: obtaining target information, wherein the target information includes at least one of a first music attribute that matches the in-vehicle scene information of a target vehicle, a second music attribute that matches the out-of-vehicle scene information of the target vehicle, a third music attribute that matches the user attribute information associated with the target vehicle, a fourth music attribute that matches music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, and a fifth music attribute that matches the collected voice information of people in the target vehicle; obtaining target music that matches the target information; and pushing the target music to an on-board terminal of the target vehicle. Therefore, in an embodiment of the present invention, the problem that the music pushed by the on-board terminal cannot meet the actual needs of users is solved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a music push method, device, electronic device and readable storage medium. Background Art

[0002] As traditional cars are developing towards intelligence, the car cabin has become the most important part of human-computer interaction in the car. More and more car cabins are equipped with online music applications. When people in the car open the music application, a lot of recommended music will be displayed on the homepage of the application, and people in the car can choose these recommended music.

[0003] Currently, in-car online music apps typically push a few songs randomly from the current popular music to users. However, even though the pushed music is currently popular, it may not meet the user's actual needs, and the method of only pushing popular music is relatively simple. Summary of the Invention

[0004] Embodiments of the present invention provide a music push method, device, electronic device, and readable storage medium, which are used to solve the problem that the music pushed by an in-vehicle terminal cannot meet the actual needs of users.

[0005] In order to solve the above problems, an embodiment of the present invention discloses a music push method, which includes:

[0006] Obtaining target information, wherein the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute, the first music attribute being a music attribute that matches in-vehicle scene information of a target vehicle, the second music attribute being a music attribute that matches in-vehicle scene information of a target vehicle, the third music attribute being a music attribute that matches user attribute information associated with the target vehicle, the fourth music attribute being an attribute that matches preset music, the preset music including music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, and the fifth music attribute being a music attribute that matches collected voice information of a person in the target vehicle;

[0007] Acquire target music matching the target information;

[0008] Push the target music to the vehicle-mounted terminal of the target vehicle.

[0009] The embodiment of the present invention further discloses a music push device, comprising:

[0010] a target information acquisition module, configured to acquire target information, wherein the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute, wherein the first music attribute is a music attribute that matches in-vehicle scene information of a target vehicle, the second music attribute is a music attribute that matches in-vehicle scene information of a target vehicle, the third music attribute is a music attribute that matches user attribute information associated with the target vehicle, the fourth music attribute is an attribute that matches preset music, the preset music including music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, and the fifth music attribute is a music attribute that matches collected voice information of a person in the target vehicle;

[0011] A target music module, configured to obtain target music that matches the target information;

[0012] The push module is used to push the target music to the vehicle-mounted terminal of the target vehicle.

[0013] Optionally, the target information acquisition module includes:

[0014] a feature information acquisition submodule, configured to acquire feature information of objects in the target vehicle, wherein the feature information includes at least one of the age of the person, the expression of the person, the number of people, and the type of item;

[0015] An in-vehicle scene determination submodule, configured to determine a target in-vehicle scene corresponding to the feature information of the object in the target vehicle based on a pre-established first correspondence between the feature information of the object and the in-vehicle scene;

[0016] The first music attribute determination submodule is configured to determine a music attribute that matches the target in-car scene according to a pre-established second correspondence between in-car scenes and music attributes, and determine the music attribute as the first music attribute.

[0017] Optionally, at least one camera is provided in the target vehicle;

[0018] The feature information acquisition submodule includes:

[0019] a first image acquisition unit, configured to acquire a first image of an object in the target vehicle captured by the at least one camera;

[0020] An extraction unit is used to extract feature information of an object in the target vehicle from the first image.

[0021] Optionally, the target information acquisition module includes:

[0022] A quantity acquisition submodule is used to obtain the number of vehicles and people within a preset range where the target vehicle is located;

[0023] a first determining submodule configured to determine, when the number of vehicles is greater than a second preset threshold, or the number of people is greater than a third preset threshold, that the exterior scene of the target vehicle belongs to a first preset scene, and to determine the predetermined music attribute corresponding to the first preset scene as the second music attribute;

[0024] The second determination submodule is used to determine that the outdoor scene of the target vehicle belongs to the second preset scene when the number of vehicles is less than or equal to the second preset threshold and the number of people is less than or equal to the third preset threshold, and to determine the music attribute predetermined corresponding to the second preset scene as the second music attribute.

[0025] Optionally, a panoramic camera is provided on the outside of the target vehicle;

[0026] The quantity acquisition submodule includes:

[0027] A second image acquisition unit, configured to acquire a second image captured by the panoramic camera;

[0028] a vehicle number determination unit, configured to obtain the number of vehicles in the second image to determine the number of vehicles within a preset range where the target vehicle is located;

[0029] The personnel number determining unit is configured to obtain the number of personnel in the second image to determine the number of personnel within a preset range where the target vehicle is located.

[0030] Optionally, the target information acquisition module includes:

[0031] A user determination submodule, configured to obtain at least one user who has logged into the vehicle-mounted terminal of the target vehicle and determine the user as the user associated with the target vehicle;

[0032] An attribute information acquisition submodule, configured to acquire attribute information of a user associated with the target vehicle;

[0033] a music attribute determination submodule, configured to determine the music attribute corresponding to the attribute information of the user associated with the target vehicle based on a predetermined third correspondence between the user attribute information and the music attribute;

[0034] The third determining submodule is configured to determine the music attribute corresponding to the attribute information of the user associated with the target vehicle in the third corresponding relationship as the third music attribute.

[0035] Optionally, the target information acquisition module further includes:

[0036] A play record acquisition submodule is used to acquire the music play history record of the user associated with the target vehicle within a second preset time period;

[0037] The fourth determining submodule is configured to obtain a first preset attribute of music in the music history playing record whose number of plays exceeds a fourth preset threshold, and determine the first preset attribute as the third music attribute.

[0038] Optionally, the target information acquisition module includes:

[0039] The fourth music attribute acquisition submodule is configured to acquire the second preset music attribute of the preset music and determine it as the fourth music attribute.

[0040] Optionally, the target information acquisition module includes:

[0041] An extraction submodule, configured to extract keywords from the voice information;

[0042] a fifth determining submodule, configured to determine a music attribute that matches the keyword from a plurality of predetermined music attributes;

[0043] The sixth determining submodule is configured to determine the music attribute that matches the keyword as the fifth music attribute.

[0044] Optionally, the target music acquisition module includes:

[0045] A feature vector acquisition submodule is configured to acquire a feature vector for each piece of music in a pre-established music library, wherein the i-th component of the feature vector for the j-th piece of music indicates whether the j-th piece of music has the i-th music attribute in the target information, where j is an integer from 1 to M, i is an integer from 1 to N, M represents the number of pieces of music in the music library, and N represents the number of music attributes included in the target information;

[0046] a calculation submodule, configured to calculate the similarity between each two pieces of music in the music library according to the feature vector and the predetermined weights of the music in the music library;

[0047] A sorting submodule, configured to sort the music in the music library according to the similarity and the weight to obtain a target sorting;

[0048] The seventh determination submodule is used to select a preset number of music based on the target ranking and determine them as target music that matches the target information.

[0049] Optionally, the calculation submodule is specifically used to:

[0050] Substitute the eigenvector of the first music and the eigenvector of the second music into the following formula , (i=1, 2, …, N)

[0051] Obtaining a similarity sim between the first music and the second music;

[0052] The first music and the second music are two pieces of music in the music library. represents the i-th component of the eigenvector of the first music, represents the i-th component of the feature vector of the second music, k1 represents the predetermined weight of the first music, and k2 represents the predetermined weight of the second music.

[0053] Optionally, the sorting submodule includes:

[0054] A first sorting unit is configured to sort the music in the music library in descending order of similarity to obtain a first sorting;

[0055] A second sorting unit is configured to sort the music in the first sorting according to the preset weights of the music, from large to small, based on the first sorting, to obtain a second sorting;

[0056] a third sorting unit, configured to determine a position where each piece of music in the music library first appears in the second sorting, and sort the music in the music library in a descending order of the positions to obtain the target sorting;

[0057] Optionally, the music pushing device further includes:

[0058] A feedback information acquisition module, used to obtain user feedback information on the target music;

[0059] An adjustment module is used to adjust the weight of the music in the music library according to the feedback information.

[0060] Optionally, the push module includes:

[0061] A total duration acquisition submodule is used to acquire the total duration of music playing by a target user within a third preset time period, wherein the target user is the user who last logged into the vehicle-mounted terminal of the target vehicle;

[0062] A first push submodule is configured to push the target music to the in-vehicle terminal using a first preset push mode when the total duration is greater than a preset duration;

[0063] The second push submodule is configured to push the target music to the vehicle-mounted terminal using a second preset push mode when the total duration is less than or equal to the preset duration.

[0064] Optionally, the first push submodule includes:

[0065] The first pushing unit is configured to push the third preset music attribute of the target music to the in-vehicle terminal.

[0066] Optionally, the second push submodule includes:

[0067] a second pushing unit, configured to push first information of the target music to the in-vehicle terminal, wherein the first information includes part of attribute information of a third preset music attribute;

[0068] The third pushing unit is configured to push second information to the vehicle terminal when the vehicle terminal logs in to the target user's account, wherein the second information includes attribute information of the third preset music attribute other than the first information.

[0069] The embodiments of the present invention include the following advantages:

[0070] In an embodiment of the present invention, target information can be obtained, wherein the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute, the first music attribute being a music attribute that matches the in-vehicle scene information of the target vehicle, the second music attribute being a music attribute that matches the out-of-vehicle scene information of the target vehicle, the third music attribute being a music attribute that matches the user attribute information associated with the target vehicle, the fourth music attribute being an attribute that matches preset music, the preset music including music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, the fifth music attribute being a music attribute that matches the collected voice information of the person in the target vehicle; then, the target music that matches the target information is obtained; thereby pushing the target music to the on-board terminal of the target vehicle.

[0071] Among them, the first music attribute is consistent with the scene inside the car where the user is at that time, the second music attribute is consistent with the scene outside the car where the user's car is at that time, the third music attribute is consistent with the user's attribute information, the fourth music attribute represents the attribute of the current popular music, and the fifth music attribute is consistent with the content mentioned in the conversation between the people in the car.

[0072] Therefore, the above-mentioned target music that matches the target information includes: music that matches the scene inside the car, music that matches the scene outside the car, music that matches the user's attribute information, current popular music, and music that matches the content mentioned in the conversation of people in the car. Therefore, in an embodiment of the present invention, music push can be performed from multiple angles, the push method is more flexible, and the pushed music matches the scenes at various angles, so that the pushed music is more in line with the actual needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0074] Figure 1 is a schematic diagram of a music push method provided by an embodiment of the present invention;

[0075] Figure 2 is a schematic diagram of recognizing an in-vehicle scene according to an embodiment of the present invention;

[0076] Figure 3 is a schematic diagram of identifying a scene outside a vehicle provided by an embodiment of the present invention;

[0077] Figure 4 is a schematic diagram of a music push system provided by an embodiment of the present invention;

[0078] Figure 5 Schematic diagram of 68 facial key points output by the facial expression matching model provided by an embodiment of the present invention;

[0079] Figure 6 This is a structural block diagram of a music pushing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0080] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0081] like Figure 1 FIG. 1 is a flow chart of a music push method according to an embodiment of the present invention. The method may include the following steps:

[0082] Step 101: Obtain target information.

[0083] In which, the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute. The first music attribute is a music attribute that matches the in-vehicle scene information of the target vehicle, the second music attribute is a music attribute that matches the out-of-vehicle scene information of the target vehicle, the third music attribute is a music attribute that matches the user attribute information associated with the target vehicle, the fourth music attribute is an attribute that matches the preset music, and the preset music includes music whose playback volume on the network platform exceeds a first preset threshold within a first preset time period. The fifth music attribute is a music attribute that matches the collected voice information of the people in the target vehicle.

[0084] The above preset music is currently popular music.

[0085] Therefore, the first music attribute is consistent with the scene inside the car where the user was at the time, the second music attribute is consistent with the scene outside the car where the user's car was at the time, the third music attribute is consistent with the user's attribute information, the fourth music attribute represents the attribute of the current popular music, and the fifth music attribute is consistent with the content mentioned in the conversation between the people in the car.

[0086] In addition, each of the first to fifth music attributes may include at least one music attribute.

[0087] Step 102: Acquire target music that matches the target information.

[0088] The target music that matches the target information includes at least one of: music that matches the in-car scene, music that matches the out-of-car scene, music that matches the user's attribute information, currently popular music, and music that matches the content of the conversations of people in the car. Therefore, in the embodiments of the present invention, music can be pushed from multiple perspectives, with a more flexible push method. The pushed music matches the scenes from various perspectives, thereby making the pushed music more in line with the user's actual needs.

[0089] Step 103: Push the target music to the vehicle-mounted terminal of the target vehicle.

[0090] In an embodiment of the present invention, a preset time interval can be set so that the target music is pushed to the on-board terminal of the target vehicle at every preset time interval, or a push instruction can be generated when a preset operation of the user is received (such as pressing a physical button or a physical combination button set on the on-board terminal), thereby triggering the execution of step 103.

[0091] After receiving the pushed target music, the in-vehicle terminal may display a list of the target music, so that the user can select at least one of the music to play.

[0092] In addition, it should be noted that, in an embodiment of the present invention, the target information includes at least one of the above-mentioned first music attribute, second music attribute, third music attribute, fourth music attribute, and fifth music attribute. Therefore, the above-mentioned target information may include one or more of the first to fifth music attributes.

[0093] Each of the first to fifth music attributes matches a scene from a different angle. Therefore, when the target information includes multiple music attributes, embodiments of the present invention can push music based on scenes from multiple different angles. These different angles represent different angles of the target vehicle's environment. Therefore, the pushed music is tailored to the environment of the occupants of the target vehicle. Therefore, when the target information includes multiple music attributes from the first to fifth music attributes, the accuracy of music push can be further improved.

[0094] Specifically, when the above-mentioned target information includes the first to fifth music attributes, an embodiment of the present invention pushes music by combining five angles: the interior scene of the target vehicle, the exterior scene, the attributes of the user associated with the target vehicle, the currently popular music, and the conversation scene of the people in the target vehicle. It should be noted that when combining these five angles for music push, it is not a simple combination of these five angles, that is, considering these five angles independently, matching the music that matches the scene of each angle respectively, and then pushing these music to the user. In this way, the single music pushed to the user may only match the scene of one of the angles, that is, from the perspective of the single music being pushed, the push accuracy is still poor.

[0095] The embodiments of the present invention respectively determine the music attributes that match the scenes at each of the above-mentioned angles to obtain the above-mentioned first to fifth music attributes, and then combine these music attributes to match the music. In this way, the music pushed to the user is music that has a high degree of matching with such music attributes, that is, the music pushed to the user has a higher degree of matching with the scenes at the above-mentioned five angles, thereby making the push accuracy of any of the pushed music higher.

[0096] It can be seen that when the above-mentioned target information includes the first to fifth music attributes, the embodiment of the present invention can achieve higher-accuracy music push, thereby enhancing the music experience of passengers to a greater extent.

[0097] Optionally, the process of obtaining the first music attribute includes:

[0098] Acquiring characteristic information of objects in the target vehicle, the characteristic information including at least one of the age of the person, the expression of the person, the number of the person, and the type of the item;

[0099] Determining a target vehicle interior scene corresponding to the feature information of the object in the target vehicle according to a pre-established first correspondence between feature information of the object and the vehicle interior scene;

[0100] According to a pre-established second correspondence between the in-car scene and the music attribute, a music attribute matching the target in-car scene is determined and determined as the first music attribute.

[0101] For example, the objects included in the target vehicle are a couple, a child, and a scooter, and the characteristic information includes the ages of the couple and the child, the number of people in the car, the toys, and the happy expressions of the couple and the child at this time. In the first correspondence, it can be determined that the scene in the car at this time is a happy scene, and thus in the second correspondence, the music attributes that match the happy scene can be determined, which can be "cheerful" or "relaxed".

[0102] Optionally, at least one camera is provided in the target vehicle; and obtaining feature information of objects in the target vehicle includes: obtaining a first image of the objects in the target vehicle captured by the at least one camera; and extracting feature information of the objects in the target vehicle from the first image.

[0103] The at least one camera may include a driver monitoring system (DMS) camera and an occupancy monitoring system (OMS) camera.

[0104] In addition, if Figure 2 As shown, the DMS camera can capture the driver's facial image, while the OMS camera can capture facial images of people in the passenger seat and back seat, as well as images of objects inside the vehicle. These images can be used to determine the age, facial expressions, number of people inside, and type of objects inside the target vehicle. It's important to note that both the DMS and OMS cameras must obtain the user's privacy authorization before capturing facial images.

[0105] In addition, if Figure 4 As shown, integrated face recognition (MediaPipe) can be used to detect facial images, thereby outputting 68 key point information, and using this information to construct a feature vector of facial expression. This feature vector is then input into a pre-established facial expression matching model to output the expression of the facial image.

[0106] It should be noted that the training process of the facial expression matching model includes:

[0107] A large number of facial sample images are collected and the expressions of each sample image are manually annotated; 68 key points of each face in each facial sample image are detected, and a feature vector for each face is constructed based on these key points; the parameters of the initial random forest algorithm are used to process the feature vector of each face and output the expression recognition results of each facial sample image; the expression recognition results of the facial sample images are compared with the manually annotated results, and the initial parameters of the random forest algorithm are modified until the convergence conditions are reached (that is, until the accuracy of the expression recognition results of the output facial image samples meets the preset requirements).

[0108] It should also be noted that target detection technology (yolov5) can be used to identify the type of object appearing in the aforementioned first image; wrinkle texture based on Hough transform (Hough) and skin color of facial image can be used to analyze the age of the facial image.

[0109] The driver's facial image captured by the DMS camera can be matched with facial information in a pre-stored facial information database. If a match is successful, the target vehicle's onboard terminal can access the driver's login interface. After the driver successfully logs in, the driver's age can be retrieved from the pre-stored driver information.

[0110] Optionally, the process of obtaining the second music attribute includes:

[0111] Obtaining the number of vehicles and people within a preset range of the target vehicle;

[0112] When the number of vehicles is greater than a second preset threshold, or the number of people is greater than a third preset threshold, determining that the exterior scene of the target vehicle belongs to a first preset scene, and determining the predetermined music attribute corresponding to the first preset scene as the second music attribute;

[0113] When the number of vehicles is less than or equal to the second preset threshold and the number of people is less than or equal to the third preset threshold, it is determined that the external scene of the target vehicle belongs to the second preset scene, and the predetermined music attribute corresponding to the second preset scene is determined as the second music attribute.

[0114] Among them, if the number of vehicles is greater than the second preset threshold, or the number of people is greater than the third preset threshold, it means that there are a large number of vehicles or people within the preset range where the target vehicle is currently located, and the first preset scene can be a congested scene; if the number of vehicles is less than or equal to the second preset threshold, and the number of people is less than or equal to the third preset threshold, it means that there are a small number of vehicles or people within the preset range where the target vehicle is currently located, and the second preset scene can be a non-congested scene; the music attribute corresponding to the first preset scene can be "soothing", and the music attribute corresponding to the second preset scene can be "cheerful".

[0115] Optionally, a panoramic camera is provided on the outside of the target vehicle;

[0116] The obtaining of the number of vehicles and the number of people within the preset range where the target vehicle is located includes:

[0117] Acquire a second image captured by the panoramic camera;

[0118] Acquiring the number of vehicles in the second image to determine the number of vehicles within a preset range where the target vehicle is located;

[0119] The number of people in the second image is obtained to determine the number of people within a preset range where the target vehicle is located.

[0120] Among them, such as Figure 3 As shown, a 360-degree panoramic camera is provided on the outside of the target vehicle for capturing a second image outside the vehicle and obtaining the scene outside the vehicle, so that music suitable for the scene outside the vehicle can be recommended to the user based on the scene outside the vehicle.

[0121] In addition, yolov5 can be used to detect vehicles and people in the second image.

[0122] Optionally, the process of obtaining the third music attribute includes:

[0123] Obtain at least one user who has logged into the vehicle-mounted terminal of the target vehicle and determine the user as the user associated with the target vehicle;

[0124] Acquiring attribute information of a user associated with the target vehicle;

[0125] determining the music attribute corresponding to the attribute information of the user associated with the target vehicle according to a predetermined third correspondence between the user attribute information and the music attribute;

[0126] The music attribute corresponding to the attribute information of the user associated with the target vehicle in the third corresponding relationship is determined as the third music attribute.

[0127] In which case, when the target vehicle's onboard terminal has logged in multiple users, the users associated with the target vehicle include all users who have logged in to the onboard terminal in the past, or users who have logged in to the onboard terminal of the target vehicle in the recent period (for example, 1 week or 1 month).

[0128] In addition, the user's attribute information may be "age", "gender", and "interests and hobbies", and the music attributes corresponding to the user's attribute information may be "cheerful", "sad", and "quiet"; for example, user A is a 22-year-old girl whose interest and hobby is reading; user B is a 29-year-old boy whose interest and hobby is traveling, then the music attribute corresponding to user A is "quiet", and the music attribute corresponding to user B is "cheerful".

[0129] Optionally, the process of obtaining the third music attribute further includes:

[0130] Obtaining music playback history records of a user associated with the target vehicle within a second preset time period;

[0131] A first preset attribute of music in the music history play record whose play times exceeds a fourth preset threshold is obtained and determined as the third music attribute.

[0132] The first preset music attribute may include the author, release time, music type, melody, rhythm, lyrics, etc. of the music.

[0133] For example, the user associated with the target vehicle has a music playback history record of music A played 5 times and music B played 7 times within a week; among which the fourth preset threshold is 6 times, then music B is music played more than 6 times, and the first preset music attribute of music B is determined to be the third music attribute.

[0134] Optionally, the process of obtaining the fourth music attribute includes:

[0135] A second preset music attribute of the preset music is obtained and determined as the fourth music attribute.

[0136] Among them, the preset music is the popular music on the current online platform (such as the latest released music, music with a large number of plays, music with high popularity, etc.); the second preset music attributes may include the author of the music, release time, music type, melody, rhythm, lyrics of the music, etc.

[0137] Optionally, the process of obtaining the fifth music attribute includes:

[0138] extracting keywords from the voice information;

[0139] Determining a music attribute that matches the keyword from a plurality of predetermined music attributes;

[0140] The music attribute that matches the keyword is determined as the fifth music attribute.

[0141] Among them, the on-board terminal of the target vehicle can collect the conversations of the people in the car, extract keywords from the conversations, and then find the corresponding music attributes from the correspondence between the predetermined keywords and multiple music attributes.

[0142] For example, the pre-selected multiple music attributes include "cheerful", "sad", and "quiet"; when the people in the car are talking about going out for a trip, the car terminal collects the conversation and identifies the keywords "travel" and "happy", then the music attribute corresponding to "travel" and "happy" is "cheerful", that is, the fifth music attribute is determined to be "cheerful".

[0143] Optionally, acquiring target music matching the target information includes:

[0144] Obtain a feature vector for each piece of music in a pre-established music library, wherein the i-th component of the feature vector for the j-th piece of music indicates whether the j-th piece of music has the i-th music attribute in the target information, j is an integer from 1 to M, i is an integer from 1 to N, M represents the number of pieces of music in the music library, and N represents the number of music attributes included in the target information;

[0145] Calculating the similarity between each two pieces of music in the music library according to the feature vector and the predetermined weights of the music in the music library;

[0146] Sorting the music in the music library according to the similarity and the weight to obtain a target sorting;

[0147] Based on the target ranking, a preset number of music pieces are selected and determined as target music pieces that match the target information.

[0148] Each feature vector of a piece of music indicates whether it possesses each of the music attributes in the target information. Optionally, the i-th component of the feature vector of the j-th piece of music takes a value of 1, indicating that the j-th piece of music possesses the i-th attribute in the target information, and a value of 0 indicates that the j-th piece of music does not possess the i-th attribute in the target information. For example, if the target information includes five music attributes, and the feature vector of music A is {00011}, then application A does not possess attributes 1 to 3 in the target information, but possesses attributes 4 to 5.

[0149] In addition, the higher the weight of a piece of music is, the higher the user's evaluation of the music is.

[0150] As can be seen from the above, the embodiment of the present invention can select music with a high correlation with the music attributes included in the aforementioned target information and a higher evaluation by calculating the aforementioned similarity and sorting the music in the music library according to the similarity and weight.

[0151] Among them, when the above-mentioned target information includes the first to fifth music attributes, the target information includes music attributes that match the scenes of the five angles of the environment in which the target vehicle is located. In this case, the above-mentioned feature vector represents the degree of matching between the music and the scenes of these five angles. By calculating the similarity between each two music in the music library through the feature vector and the weight of the music, the music that has a high degree of matching with the scenes of these five angles and has a high user evaluation can be screened out. In this way, the accuracy of music push can be further improved, that is, the music pushed to the user can be more in line with the scenes of the five angles of the environment in which the target vehicle is located, and more likely to be welcomed by the people in the target vehicle.

[0152] Optionally, the process of calculating the similarity between the first music and the second music based on the feature vector and the predetermined weights of the music in the music library includes:

[0153] Substitute the eigenvector of the first music and the eigenvector of the second music into the following formula , (i=1, 2, …, N)

[0154] Obtaining a similarity sim between the first music and the second music;

[0155] The first music and the second music are two pieces of music in the music library. represents the i-th component of the eigenvector of the first music, The i-th component of the feature vector representing the second music, i is an integer from 1 to N, N represents the number of music attributes included in the target information, k1 represents a predetermined weight of the first music, and k2 represents a predetermined weight of the second music.

[0156] In addition, d is the distance measure between two music attributes, and sim is the similarity between two music. The shorter the distance between two music attributes, the higher the similarity of the music.

[0157] For example, if the target information includes 5 music attributes, each music can be identified by a feature vector with a dimension of 5. For example, the feature vector of music A is X=( ), the characteristic vector of music B is Y=( ).

[0158] Optionally, sorting the music in the music library according to the similarity and the weight to obtain a target sort includes:

[0159] Sorting the music in the music library in descending order of the similarity to obtain a first sort;

[0160] Based on the first sorting, the music associated with the same similarity is sorted in descending order according to the preset weights of the music to obtain a second sorting;

[0161] Determine the position where each piece of music in the music library first appears in the second sorting, and sort the music in the music library in order from front to back according to the position to obtain the target sorting.

[0162] Among them, the first sorting is to arrange the music in descending order according to the similarity between each two songs in the music library; the second sorting is to arrange the music with the same similarity from large to small according to the preset weight of the music on the basis of the first sorting; the target sorting is to arrange the music from front to back according to the position where it first appears in the second sorting.

[0163] For example, there are three pieces of music A, B, and C in the music library, among which the similarity between A and B is 0.8, the similarity between A and C is 0.6, and the similarity between B and C is 0.7; the preset weight of A is 4, the preset weight of B is 8, and the preset weight of C is 5; then the first sorting is AB, BC, AC, the second sorting is BA, BC, CA, and the target sorting is B, A, C.

[0164] Optionally, after pushing the target music to the vehicle-mounted terminal of the target vehicle, the method further includes:

[0165] Obtaining user feedback information on the target music;

[0166] The weights of the music in the music library are adjusted according to the feedback information.

[0167] The feedback information may include the user's click rate on the target music, the number of times the target music is played within a preset time, and whether the user is satisfied with the evaluation of the target music.

[0168] In addition, the weight of the music may be adjusted based on the above feedback information at regular time intervals.

[0169] In addition, if the click rate of the third music by the user is greater than a preset click threshold, or the number of times it is played within a preset time is greater than a preset number, or the user's evaluation keyword indicates "satisfaction", then when the current weight value of the third music is greater than or equal to the preset maximum value, or the sum of the current weight value and the first preset value is greater than or equal to the preset maximum value, the current weight value of the third music remains unchanged; if the sum of the current weight value of the third music and the first preset value is less than the preset maximum value, the current weight value of the third music is increased by the first preset value;

[0170] When the user's click rate on the third music is less than or equal to the preset click threshold, and the number of times it is played within the preset time is less than or equal to the preset number of times, and the user's evaluation keyword indicates "dissatisfaction", then when the current weight value of the third music is less than or equal to the preset minimum value, or the difference between the current weight value and the second preset value is less than or equal to the preset minimum value, the current weight value of the third music remains unchanged; when the difference between the current weight value of the third music and the second preset value is greater than the preset minimum value, the current weight value of the third music is reduced by the second preset value.

[0171] For example, the target music pushed by the on-board terminal of the target vehicle are A and B, where the preset weight of A is 2 and the preset weight of B is 5; the user clicks A 3 times, plays it completely 5 times, and evaluates A as satisfactory; the user clicks B 1 time, plays it completely 1 time, and evaluates B as dissatisfied. Based on the user's feedback information, the weight of A can be adjusted to 5 and the weight of B can be adjusted to 2.

[0172] Optionally, pushing the target music to the vehicle-mounted terminal of the target vehicle includes:

[0173] Obtaining a total duration of music playback by a target user within a third preset time period, wherein the target user is the user who last logged into the vehicle-mounted terminal of the target vehicle;

[0174] When the total duration is greater than the preset duration, the target music is pushed to the vehicle terminal using the first preset push mode;

[0175] When the total duration is less than or equal to the preset duration, the target music is pushed to the vehicle terminal using a second preset push mode.

[0176] Among them, if the total time duration of music playing within the third preset time period is greater than the preset time duration, it indicates that the target user is an active user, that is, he often uses the on-board terminal of the target vehicle to play music; if the total time duration of music playing within the third preset time period is less than or equal to the preset time duration, it indicates that the target user is an inactive user, that is, he does not often use the on-board terminal of the target vehicle to play music.

[0177] In addition, the first preset push mode is a push mode for active users; the second preset push mode is a push mode for inactive users.

[0178] Optionally, the adopting the first preset push mode to push the target music to the in-vehicle terminal includes:

[0179] Pushing the third preset music attribute of the target music to the in-vehicle terminal.

[0180] Among them, if the total time the target user plays music within the third preset time period is greater than the preset time, it means that he often uses the on-board terminal of the target vehicle to play music. In this case, all the third preset music attributes of the target music to be pushed can be directly pushed to the user.

[0181] Optionally, the adopting of the second preset push mode to push the target music to the in-vehicle terminal includes:

[0182] Pushing first information of the target music to the in-vehicle terminal, wherein the first information includes partial attribute information of a third preset music attribute;

[0183] In a case where the in-vehicle terminal logs in to the target user's account, second information is pushed to the in-vehicle terminal, wherein the second information includes attribute information of the third preset music attribute except the first information.

[0184] In addition, if the total duration of the target user playing music within the third preset time period is less than or equal to the preset duration, it means that the target user does not frequently use the on-board terminal of the target vehicle to play music. In this case, the third preset music attributes of a part of the target music to be pushed can be pushed to the user. After the target user logs in to the on-board terminal of the target vehicle, the third preset music attributes of another part of the target music to be pushed can be pushed to the user.

[0185] In summary, a specific implementation of a system using the music push method according to an embodiment of the present invention, that is, a music push system, can be described as follows:

[0186] like Figure 5 As shown, the music push system 500 includes: a user behavior attribute acquisition module 501, a multi-attribute sorting module 502 and a user feedback module 503. The specific implementation plans of these three modules will be introduced below.

[0187] 1. User Behavior Attribute Acquisition Module 501

[0188] The user behavior attribute acquisition module 501 is the core module of the intelligent cockpit music recommendation system, which includes a visual information acquisition submodule, a user usage information acquisition submodule and a music attribute information acquisition submodule.

[0189] 1.1 Visual Information Acquisition Submodule

[0190] The visual information acquisition submodule includes: an in-vehicle scene determination unit and an out-of-vehicle scene determination unit.

[0191] The in-vehicle scene determination unit obtains the in-vehicle images captured by the DMS camera and the OMS camera, and then uses the yolov5 algorithm to detect the type of objects in the in-vehicle images, and uses the face key point detection algorithm to detect the facial information of the faces in the in-vehicle images to obtain the expressions, ages, and number of people in the car, and then determines the in-vehicle scene based on the detected expressions, ages, number of people, and type of objects in the car, and then determines the music attributes corresponding to the obtained in-vehicle scene based on the predetermined correspondence between different in-vehicle scenes and music attributes.

[0192] The vehicle exterior scene determination unit obtains exterior images captured by the 360-degree panoramic camera and uses the Yolov5 algorithm to detect the number of people and vehicles in the exterior image. Based on the obtained number of people and vehicles, it determines the exterior scene. Furthermore, based on the predetermined correspondence between different exterior scenes and music attributes, it determines the music attributes corresponding to the obtained exterior scene. Specifically, it identifies the density of traffic and pedestrian traffic outside the vehicle to construct the user's exterior scene. This allows it to obtain music attributes that match the exterior scene, both in congested and uncongested conditions. For example, when traffic is congested, it can recommend some pure music to alleviate the user's negative emotions.

[0193] Among them, in the above-mentioned visual information acquisition sub-module, the driver's facial information can also be collected and compared with the stored facial information. When there is facial information that matches the collected driver's facial information, the login interface of the music push system is entered, thereby triggering the in-vehicle scene judgment unit to judge the in-vehicle scene, and triggering the out-of-vehicle scene judgment unit to judge the out-of-vehicle scene, otherwise the relevant information of the face is re-entered.

[0194] 1.2 User Usage Information Acquisition Submodule

[0195] In the user usage information acquisition submodule, user attribute information is obtained based on the user's music playback history or user information (age, gender, interests, etc.). The user here can be a user who has logged into the music push system in the recent period (for example, within the last month).

[0196] 1.3 Music attribute information acquisition submodule

[0197] In the music attribute information acquisition submodule, the current popular music is obtained and the attributes of the popular music itself (such as the author of the music, release time, music type, melody, and rhythm) are extracted.

[0198] From the above, it can be seen that the processing based on the visual information acquisition module 501 can fully demonstrate the service performance of the music push system. The change from passive to active interaction mode can improve the accuracy of the in-car music recommendation system, enhance the user's music experience, and even alleviate some negative emotions when using the car, so that users can have a more comfortable car experience scenario.

[0199] 2. Multi-attribute sorting module 502

[0200] The multi-attribute sorting module 502 is an important module of the music push system. Based on the multiple music attributes output by the aforementioned user behavior attribute acquisition module, this module performs the following processing on the music in the music library:

[0201] Determine a feature vector for each piece of music, where the i-th component of the feature vector for the j-th piece of music indicates whether the j-th piece of music has the i-th music attribute output by the user behavior attribute acquisition module, where j is an integer from 1 to M, and i is an integer from 1 to N, where M represents the number of pieces of music in the music library, and N represents the number of music attributes output by the user behavior attribute acquisition module;

[0202] Calculate the similarity between each pair of songs in the music library based on the predetermined weights of each song and the feature vectors of each song (the specific method for calculating the similarity here is described above and will not be repeated here);

[0203] Sorting the music in the music library according to the similarity and the weight to obtain a target ranking (the specific sorting method here is described above and will not be repeated here);

[0204] Based on the target ranking, a preset number of music pieces are selected and determined as target music pieces that match the target information;

[0205] Push the target music to the user (the specific push method here is described above and will not be repeated here).

[0206] 3. User Feedback Module 503

[0207] The user feedback module 503 is used to collect the click rate, the number of times the pushed music is played within a preset time period, and the evaluation information of the users, so as to adjust the weight of the music in the music library according to the feedback information.

[0208] As can be seen from the above, the user's experience with the music being pushed can influence the music content recommended by the music push system. By directly feeding the user's current feedback information to the music push system, the efficiency of human-computer interaction is improved, allowing users to get closer to the system, improving the system's functionality and providing real-time feedback on system usage.

[0209] As can be seen from the above, the music push system of this embodiment of the present invention takes the user as its starting point, proactively identifying music preferences or interests. It then recommends desired music and related information based on the user's usage habits, the user group status within the vehicle, and the in-vehicle scene. By integrating in-vehicle camera video surveillance intelligent operation and maintenance system (VIMS), DMS, and 360-degree panoramic cameras to "perceive" information inside and outside the vehicle, it constructs the user's usage scenario. The system then uses the user's usage information and the attributes of the music as attribute information for the music recommendation system, providing users with a more enjoyable music listening experience.

[0210] In addition, by integrating the visual information on the car and then constructing the user's usage scenario based on the visual information inside and outside the car, the problem of the music recommendation function in the existing technology being inaccurate and not flexible enough to meet the usage needs in real scenarios is solved, giving users a better music experience and creating a better music interaction experience.

[0211] Specifically, this embodiment of the present invention utilizes in-vehicle visual information to identify the user's desired scenario, determine the in-vehicle scene status, and, based on the user's facial information, gather information such as the user's age and mood to understand their current physiological and psychological state. By analyzing the pitch, rhythm, and melody of different music, the system selects the optimal background music for the user to most effectively alleviate their emotions and create a more comfortable musical experience. Furthermore, the system can recommend music of interest and related music information from a vast library of available music. A personalized music recommendation system is developed based on the user's behavioral trajectory, providing a more intelligent music interaction experience.

[0212] It should be noted that for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0213] Reference Figure 6, shows a structural block diagram of a music push device according to an embodiment of the present invention. The music push device 600 may include the following modules:

[0214] The target information acquisition module 601 is used to acquire target information, wherein the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute, wherein the first music attribute is a music attribute that matches the in-vehicle scene information of the target vehicle, the second music attribute is a music attribute that matches the in-vehicle scene information of the target vehicle, the third music attribute is a music attribute that matches the user attribute information associated with the target vehicle, the fourth music attribute is an attribute that matches preset music, wherein the preset music includes music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, and the fifth music attribute is a music attribute that matches the collected voice information of the person in the target vehicle;

[0215] A target music acquisition module 602 is configured to acquire target music that matches the target information;

[0216] The push module 603 is used to push the target music to the vehicle-mounted terminal of the target vehicle.

[0217] Optionally, the target information acquisition module 601 includes:

[0218] a feature information acquisition submodule, configured to acquire feature information of objects in the target vehicle, wherein the feature information includes at least one of the age of the person, the expression of the person, the number of people, and the type of item;

[0219] An in-vehicle scene determination submodule, configured to determine a target in-vehicle scene corresponding to the feature information of the object in the target vehicle based on a pre-established first correspondence between the feature information of the object and the in-vehicle scene;

[0220] The first music attribute determination submodule is configured to determine a music attribute that matches the target in-car scene according to a pre-established second correspondence between in-car scenes and music attributes, and determine the music attribute as the first music attribute.

[0221] Optionally, at least one camera is provided in the target vehicle;

[0222] The feature information acquisition submodule includes:

[0223] a first image acquisition unit, configured to acquire a first image of an object in the target vehicle captured by the at least one camera;

[0224] An extraction unit is used to extract feature information of an object in the target vehicle from the first image.

[0225] Optionally, the target information acquisition module 601 includes:

[0226] A quantity acquisition submodule is used to obtain the number of vehicles and people within a preset range where the target vehicle is located;

[0227] a first determining submodule configured to determine, when the number of vehicles is greater than a second preset threshold, or the number of people is greater than a third preset threshold, that the exterior scene of the target vehicle belongs to a first preset scene, and to determine the predetermined music attribute corresponding to the first preset scene as the second music attribute;

[0228] The second determination submodule is used to determine that the outdoor scene of the target vehicle belongs to the second preset scene when the number of vehicles is less than or equal to the second preset threshold and the number of people is less than or equal to the third preset threshold, and to determine the music attribute predetermined corresponding to the second preset scene as the second music attribute.

[0229] Optionally, a panoramic camera is provided on the outside of the target vehicle;

[0230] The quantity acquisition submodule includes:

[0231] A second image acquisition unit, configured to acquire a second image captured by the panoramic camera;

[0232] a vehicle number determination unit, configured to obtain the number of vehicles in the second image to determine the number of vehicles within a preset range where the target vehicle is located;

[0233] The personnel number determining unit is configured to obtain the number of personnel in the second image to determine the number of personnel within a preset range where the target vehicle is located.

[0234] Optionally, the target information acquisition module 601 includes:

[0235] A user determination submodule, configured to obtain at least one user who has logged into the vehicle-mounted terminal of the target vehicle and determine the user as the user associated with the target vehicle;

[0236] An attribute information acquisition submodule, configured to acquire attribute information of a user associated with the target vehicle;

[0237] a music attribute determination submodule, configured to determine the music attribute corresponding to the attribute information of the user associated with the target vehicle based on a predetermined third correspondence between the user attribute information and the music attribute;

[0238] The third determining submodule is configured to determine the music attribute corresponding to the attribute information of the user associated with the target vehicle in the third corresponding relationship as the third music attribute.

[0239] Optionally, the target information acquisition module 601 further includes:

[0240] A play record acquisition submodule is used to acquire the music play history record of the user associated with the target vehicle within a second preset time period;

[0241] The fourth determining submodule is configured to obtain a first preset attribute of music in the music history playing record whose number of plays exceeds a fourth preset threshold, and determine the first preset attribute as the third music attribute.

[0242] Optionally, the target information acquisition module 601 includes:

[0243] The fourth music attribute acquisition submodule is configured to acquire the second preset music attribute of the preset music and determine it as the fourth music attribute.

[0244] Optionally, the target information acquisition module 601 includes:

[0245] An extraction submodule, configured to extract keywords from the voice information;

[0246] a fifth determining submodule, configured to determine a music attribute that matches the keyword from a plurality of predetermined music attributes;

[0247] The sixth determining submodule is configured to determine the music attribute that matches the keyword as the fifth music attribute.

[0248] Optionally, the target music acquisition module 602 includes:

[0249] A feature vector acquisition submodule is configured to acquire a feature vector for each piece of music in a pre-established music library, wherein the i-th component of the feature vector for the j-th piece of music indicates whether the j-th piece of music has the i-th music attribute in the target information, where j is an integer from 1 to M, i is an integer from 1 to N, M represents the number of pieces of music in the music library, and N represents the number of music attributes included in the target information;

[0250] a calculation submodule, configured to calculate the similarity between each two pieces of music in the music library according to the feature vector and the predetermined weights of the music in the music library;

[0251] A sorting submodule, configured to sort the music in the music library according to the similarity and the weight to obtain a target sorting;

[0252] The seventh determination submodule is used to select a preset number of music based on the target ranking and determine them as target music that matches the target information.

[0253] Optionally, the calculation submodule is specifically used to:

[0254] Substitute the eigenvector of the first music and the eigenvector of the second music into the following formula , (i=1, 2, …, N)

[0255] Obtaining a similarity sim between the first music and the second music;

[0256] The first music and the second music are two pieces of music in the music library. represents the i-th component of the eigenvector of the first music, represents the i-th component of the feature vector of the second music, k1 represents the predetermined weight of the first music, and k2 represents the predetermined weight of the second music.

[0257] Optionally, the sorting submodule includes:

[0258] A first sorting unit is configured to sort the music in the music library in descending order of similarity to obtain a first sorting;

[0259] A second sorting unit is configured to sort the music in the first sorting according to the preset weights of the music, from large to small, based on the first sorting, to obtain a second sorting;

[0260] The third sorting unit is used to determine the position where each piece of music in the music library first appears in the second sorting, and sort the music in the music library in order from front to back according to the position to obtain the target sorting.

[0261] Optionally, the music push device 600 further includes:

[0262] A feedback information acquisition module, used to obtain user feedback information on the target music;

[0263] An adjustment module is used to adjust the weight of the music in the music library according to the feedback information.

[0264] Optionally, the push module 603 includes:

[0265] A total duration acquisition submodule is used to acquire the total duration of music playing by a target user within a third preset time period, wherein the target user is the user who last logged into the vehicle-mounted terminal of the target vehicle;

[0266] A first push submodule is configured to push the target music to the in-vehicle terminal using a first preset push mode when the total duration is greater than a preset duration;

[0267] The second push submodule is configured to push the target music to the vehicle-mounted terminal using a second preset push mode when the total duration is less than or equal to the preset duration.

[0268] Optionally, the first push submodule includes:

[0269] The first pushing unit is configured to push the third preset music attribute of the target music to the in-vehicle terminal.

[0270] Optionally, the second push submodule includes:

[0271] a second pushing unit, configured to push first information of the target music to the in-vehicle terminal, wherein the first information includes part of attribute information of a third preset music attribute;

[0272] The third pushing unit is configured to push second information to the vehicle terminal when the vehicle terminal logs in to the target user's account, wherein the second information includes attribute information of the third preset music attribute other than the first information.

[0273] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0274] An embodiment of the present invention further provides an electronic device, including:

[0275] One or more processors; and one or more machine-readable media having instructions stored thereon, which, when executed by the one or more processors, cause the electronic device to perform the method described in the embodiment of the present invention.

[0276] The electronic device may be a terminal device or a server, and the terminal device may be a mobile phone, a tablet computer, or a vehicle-mounted terminal.

[0277] The embodiments of the present invention further provide one or more machine-readable media having instructions stored thereon, which, when executed by one or more processors, enable the processors to perform the methods described in the embodiments of the present invention.

[0278] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0279] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0280] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0281] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0282] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0283] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0284] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0285] The above is a detailed introduction to a music push method, device electronic device and readable storage medium provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A music push method, characterized in that: The method comprises: Obtaining target information, wherein the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute, the first music attribute being a music attribute that matches in-vehicle scene information of a target vehicle, the second music attribute being a music attribute that matches in-vehicle scene information of a target vehicle, the third music attribute being a music attribute that matches user attribute information associated with the target vehicle, the fourth music attribute being an attribute that matches preset music, the preset music including music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, and the fifth music attribute being a music attribute that matches collected voice information of a person in the target vehicle; Acquire target music matching the target information; Pushing the target music to the vehicle-mounted terminal of the target vehicle; The acquiring target music matching the target information includes: Obtaining a feature vector for each piece of music in a pre-established music library, wherein the i-th component of the feature vector for the j-th piece of music indicates whether the j-th piece of music has the i-th music attribute in the target information, j is an integer from 1 to M, i is an integer from 1 to N, M represents the number of pieces of music in the music library, and N represents the number of music attributes included in the target information; Calculating the similarity between each two pieces of music in the music library according to the feature vector and the predetermined weights of the music in the music library; Sorting the music in the music library according to the similarity and the weight to obtain a target sorting; Based on the target ranking, a preset number of music pieces are selected and determined as target music pieces that match the target information.

2. The method according to claim 1, characterized in that The process of obtaining the first music attribute includes: Acquiring characteristic information of objects in the target vehicle, the characteristic information including at least one of the age of the person, the expression of the person, the number of the person, and the type of the item; Determining a target vehicle interior scene corresponding to the feature information of the object in the target vehicle according to a pre-established first correspondence between feature information of the object and the vehicle interior scene; According to a pre-established second correspondence between the in-car scene and the music attribute, a music attribute matching the target in-car scene is determined and determined as the first music attribute.

3. The method according to claim 2, characterized in that At least one camera is provided in the target vehicle; and obtaining feature information of an object in the target vehicle includes: acquiring a first image of an object within the target vehicle captured by the at least one camera; Feature information of an object within the target vehicle is extracted from the first image.

4. The method according to claim 1, wherein The process of obtaining the second music attribute includes: Obtaining the number of vehicles and people within a preset range of the target vehicle; When the number of vehicles is greater than a second preset threshold, or the number of people is greater than a third preset threshold, determining that the exterior scene of the target vehicle belongs to a first preset scene, and determining the predetermined music attribute corresponding to the first preset scene as the second music attribute; When the number of vehicles is less than or equal to the second preset threshold and the number of people is less than or equal to the third preset threshold, it is determined that the external scene of the target vehicle belongs to the second preset scene, and the predetermined music attribute corresponding to the second preset scene is determined as the second music attribute.

5. The method according to claim 4, characterized in that A panoramic camera is provided on the outside of the target vehicle; The obtaining of the number of vehicles and the number of people within the preset range of the target vehicle includes: Acquire a second image captured by the panoramic camera; Acquiring the number of vehicles in the second image to determine the number of vehicles within a preset range where the target vehicle is located; The number of people in the second image is obtained to determine the number of people within a preset range where the target vehicle is located.

6. The method according to claim 1, characterized in that The process of obtaining the third music attribute includes: Obtain at least one user who has logged into the vehicle-mounted terminal of the target vehicle and determine the user as the user associated with the target vehicle; Acquiring attribute information of a user associated with the target vehicle; determining the music attribute corresponding to the attribute information of the user associated with the target vehicle according to a predetermined third correspondence between the user attribute information and the music attribute; The music attribute corresponding to the attribute information of the user associated with the target vehicle in the third corresponding relationship is determined as the third music attribute.

7. The method according to claim 6, characterized in that The process of obtaining the third music attribute further includes: Obtaining music playback history records of a user associated with the target vehicle within a second preset time period; A first preset attribute of music in the music history play record whose play times exceeds a fourth preset threshold is obtained and determined as the third music attribute.

8. The method according to claim 1, characterized in that The process of obtaining the fourth music attribute includes: A second preset music attribute of the preset music is obtained and determined as the fourth music attribute.

9. The method according to claim 1, characterized in that The process of obtaining the fifth music attribute includes: extracting keywords from the voice information; Determining a music attribute that matches the keyword from a plurality of predetermined music attributes; The music attribute that matches the keyword is determined as the fifth music attribute.

10. The method according to claim 1, characterized in that The process of calculating the similarity between the first music and the second music according to the feature vector and the predetermined weights of the music in the music library includes: Substitute the eigenvector of the first music and the eigenvector of the second music into the following formula , (i=1, 2, …, N) Obtaining a similarity sim between the first music and the second music; Wherein, d is the distance measure between two music attributes, the first music and the second music are two music in the music library, represents the i-th component of the eigenvector of the first music, represents the i-th component of the feature vector of the second music, k1 represents the predetermined weight of the first music, and k2 represents the predetermined weight of the second music.

11. The method according to claim 10, characterized in that Sorting the music in the music library according to the similarity and the weight to obtain a target sorting includes: Sorting the music in the music library in descending order of the similarity to obtain a first sort; Based on the first sorting, the music associated with the same similarity is sorted in descending order according to the preset weights of the music to obtain a second sorting; Determine the position where each piece of music in the music library first appears in the second sorting, and sort the music in the music library in order from front to back according to the position to obtain the target sorting.

12. The method according to claim 11, characterized in that After pushing the target music to the vehicle-mounted terminal of the target vehicle, the method further includes: Obtaining user feedback information on the target music; The weights of the music in the music library are adjusted according to the feedback information.

13. The method according to claim 1, wherein The pushing of the target music to the vehicle-mounted terminal of the target vehicle includes: Obtaining a total duration of music playback by a target user within a third preset time period, wherein the target user is the user who last logged into the vehicle-mounted terminal of the target vehicle; When the total duration is greater than the preset duration, the target music is pushed to the vehicle terminal using the first preset push mode; When the total duration is less than or equal to the preset duration, the target music is pushed to the vehicle terminal using a second preset push mode.

14. The method according to claim 13, wherein: The adopting the first preset push mode to push the target music to the vehicle-mounted terminal includes: Pushing the third preset music attribute of the target music to the in-vehicle terminal.

15. The method according to claim 14, characterized in that The adopting the second preset push mode to push the target music to the vehicle-mounted terminal includes: Pushing first information of the target music to the in-vehicle terminal, wherein the first information includes partial attribute information of a third preset music attribute; In a case where the in-vehicle terminal logs in to the target user's account, second information is pushed to the in-vehicle terminal, wherein the second information includes attribute information of the third preset music attribute except the first information.

16. A music push device, characterized in that: The device comprises: a target information acquisition module, configured to acquire target information, wherein the target information includes at least one of a first music attribute, a second music attribute, a third music attribute, a fourth music attribute, and a fifth music attribute, wherein the first music attribute is a music attribute that matches in-vehicle scene information of a target vehicle, the second music attribute is a music attribute that matches in-vehicle scene information of a target vehicle, the third music attribute is a music attribute that matches user attribute information associated with the target vehicle, the fourth music attribute is an attribute that matches preset music, the preset music including music whose playback volume on a network platform exceeds a first preset threshold within a first preset time period, and the fifth music attribute is a music attribute that matches collected voice information of a person in the target vehicle; A target music acquisition module, configured to acquire target music that matches the target information; A push module, configured to push the target music to the vehicle-mounted terminal of the target vehicle; The target music acquisition module includes: A feature vector acquisition submodule is configured to acquire a feature vector for each piece of music in a pre-established music library, wherein the i-th component of the feature vector for the j-th piece of music indicates whether the j-th piece of music has the i-th music attribute in the target information, where j is an integer from 1 to M, i is an integer from 1 to N, M represents the number of pieces of music in the music library, and N represents the number of music attributes included in the target information; a calculation submodule, configured to calculate the similarity between each two pieces of music in the music library according to the feature vector and the predetermined weights of the music in the music library; A sorting submodule, configured to sort the music in the music library according to the similarity and the weight to obtain a target sorting; The seventh determination submodule is used to select a preset number of music based on the target ranking and determine them as target music that matches the target information.

17. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the music push method according to any one of claims 1 to 15.

18. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the music pushing method according to any one of claims 1 to 15.

Images