CN113641902A - Music information pushing method and device, computer equipment and storage medium thereof - Google Patents

Music information pushing method and device, computer equipment and storage medium thereof Download PDF

Info

Publication number
CN113641902A
CN113641902A CN202110917597.6A CN202110917597A CN113641902A CN 113641902 A CN113641902 A CN 113641902A CN 202110917597 A CN202110917597 A CN 202110917597A CN 113641902 A CN113641902 A CN 113641902A
Authority
CN
China
Prior art keywords
music
information
user
login user
acquiring
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110917597.6A
Other languages
Chinese (zh)
Inventor
黎财可
杨文应
谢华为
王玉屏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aimyunion Technology Ltd
Original Assignee
Aimyunion Technology Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Aimyunion Technology Ltd filed Critical Aimyunion Technology Ltd
Priority to CN202110917597.6A priority Critical patent/CN113641902A/en
Publication of CN113641902A publication Critical patent/CN113641902A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/43Querying
    • G06F16/435Filtering based on additional data, e.g. user or group profiles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/45Clustering; Classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16Sound input; Sound output
    • G06F3/165Management of the audio stream, e.g. setting of volume, audio stream path

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application relates to a music information pushing method, a music information pushing device, a computer device and a storage medium thereof, wherein the method comprises the following steps: acquiring user information of a login user, and acquiring user image information of the login user according to the user information; reading currently detected real-time interaction information of the login user; selecting target music from a music library according to the user portrait information and the real-time interaction information; and pushing the target music to the login user. According to the technical scheme, the current real interest and demand of the user can be identified more accurately in more occasions, the pushing accuracy is improved, and the user experience is improved.

Description

Music information pushing method and device, computer equipment and storage medium thereof
Technical Field
The present application relates to the field of music push technologies, and in particular, to a music information push method and apparatus, a computer device, and a storage medium thereof.
Background
In music information pushing, a variety of recommendation systems are used, interest information, historical behavior information and the like of a user are collected, tagged user image information is utilized to train an artificial intelligence recommendation model and the like, audio and video content is matched according to the user image information, and then recommendation is carried out on the audio and video content to the user; for example, chinese patent application publication No.: CN108509534A, discloses "a personalized music recommendation system based on deep learning and its implementation method", chinese patent application publication No.: CN105930429A, a music recommendation method and apparatus, similar to these technologies, selects recommended music by using a trained recommendation model.
In practical application, however, the applicant finds that, when a model is trained, the model is often trained on the basis of historical data, so that in many occasions, it is difficult to accurately identify the current real interest and demand of a user, and the model trained by a large amount of sample data is beneficial to improving the overall accuracy of recommending information content, but the response to the interest change of the user is slow, and on the contrary, it is difficult to accurately identify the real-time real demand of the user, so that the pushing accuracy is affected.
Disclosure of Invention
The application provides a music information pushing method and device, a computer device and a storage medium thereof, which are used for solving one of the defects and improving the accuracy of music pushing.
A music information pushing method comprises the following steps:
acquiring user information of a login user, and acquiring user image information of the login user according to the user information;
reading currently detected real-time interaction information of the login user;
selecting target music from a music library according to the user portrait information and the real-time interaction information;
and pushing the target music to the login user.
In one embodiment, the real-time interactive information is somatosensory interactive information of a logged-in user, which is synchronously detected by the client device when music is played;
the client equipment comprises an image processing system, a system host and a sound system;
the image processing system is used for acquiring image data associated with a login user;
the system host is used for outputting music to the sound system for playing and receiving somatosensory image data fed back by the image processing system in real time;
and the sound system is used for receiving and playing the music output by the system host.
In one embodiment, selecting the target music from the music library based on the user profile information and the real-time interaction information comprises:
calling a pre-trained first recommendation model; the recommendation model is obtained by training samples based on user image information of the login user and music which is operated historically;
and screening a candidate music set from the music library according to the interaction information, and inputting the portrait information of the user and the tag information of the music in the selected music list into the first recommendation model to obtain the target music.
In one embodiment, the music information pushing method further includes:
acquiring historical record information of music operation of the login user;
extracting characteristic information of the music according to the historical record information;
and acquiring a user label of the login user according to the characteristic information, and updating the user image information according to the user label.
In one embodiment, the music information pushing method further includes:
acquiring music information operated by the login user;
extracting key information of music operated by the login user, and searching the same type of music and other operated users according to the key information;
acquiring corresponding user portrait information according to the user information of the other users;
and carrying out statistical analysis on the user portrait information of other users to obtain initial user portrait information of the login user.
In one embodiment, the music information pushing method further includes:
acquiring a current sound effect mode of the client equipment;
selecting target music from a music library according to the user portrait information and the real-time interaction information, and the method comprises the following steps:
determining the currently interested music type of the login user according to the sound effect mode, and selecting a music set to be recommended from the music library according to the music type;
and selecting target music from the music set according to the user portrait information and the interaction information.
In one embodiment, the music information pushing method further includes:
acquiring music operated by the login user within a recent time period;
predicting the interest change trend of the login user according to the operated music;
and updating the user portrait information of the login user according to the change trend and changing the output result of the first recommendation model.
In one embodiment, the music information pushing method further includes:
acquiring new music which is newly put in a warehouse;
labeling the new music;
inputting the label information into a pre-trained second recommendation model, and identifying a target user set; the second recommendation model is obtained by training samples based on label information of a plurality of pieces of music of the same type as the new music and user information corresponding to users with the plurality of pieces of music operated;
selecting a number of users from the set of target users and pushing the new music to the users.
In one embodiment, the music information pushing method further includes:
acquiring an environment image of the login user based on client equipment;
determining the current application scene of the login user according to the environment image;
before the candidate music set is screened out from the music library according to the interaction information, the method further comprises the following steps:
and screening out a music set which is classified correspondingly to the application scene from the music library according to the application scene.
A music information pushing apparatus comprising:
the image acquisition unit is used for acquiring user information of a login user and acquiring user image information of the login user according to the user information;
the interaction detection unit is used for reading the currently detected real-time interaction information of the login user;
the target selection module is used for selecting a target unit from a music library according to the user portrait information and the real-time interaction information;
and the music pushing unit is used for pushing the target music to the login user.
A computer device, the computer device comprising:
one or more processors;
a memory;
one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to: the music information pushing method according to any of the above embodiments is performed.
A computer-readable storage medium storing at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded by a processor and executes the music information pushing method of any of the above embodiments.
The technical scheme of this application has following beneficial effect:
when music information is pushed, target music is selected from a music library by combining currently detected real-time interaction information of a login user with user portrait information, and then pushing is carried out.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a hardware environment diagram for implementing a music information pushing method according to an embodiment;
fig. 2 is a flowchart of a music information pushing method of the present application;
FIG. 3 is a flow diagram of updating user representation information;
FIG. 4 is a flow diagram of establishing initial user representation information;
FIG. 5 is a flow chart of the filtering of push music based on sound effect mode;
FIG. 6 is a flow chart for pushing music based on user trends;
FIG. 7 is a push flow diagram of new music being newly binned;
FIG. 8 is a flow chart of music push based on user usage scenarios;
fig. 9 is a schematic structural diagram of a music information pushing device of the present application;
FIG. 10 is a schematic diagram of a client device of an embodiment;
FIG. 11 is an electrical schematic diagram of a client device of an embodiment;
FIG. 12 is an exploded view of the body-sensory interactive device hardware architecture of one embodiment.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary only for the purpose of explaining the present application and are not to be construed as limiting the present application.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, or operations, but do not preclude the presence or addition of one or more other features, integers, steps, operations, or groups thereof.
Referring to fig. 1, fig. 1 is a schematic diagram of a hardware environment for implementing a music information pushing method according to an embodiment, where a client device 100 is connected to a motion capture device 120 and connected to a backend server 110 through a network; a user can input interactive information to the client device 100, wherein the real-time interactive information is somatosensory interactive information of a logged-in user synchronously detected when the client device 100 plays music; for example, the motion sensing interaction information input to the client device 100 through the motion capture device 120 in the figure may be motion interaction information obtained by capturing an image, or interaction information input through other devices, such as outputting operation information through a wireless microphone, and the like, and for the client device, the client device 100 in the embodiment may be, or other devices that can interact with a user, such as other similar smart speakers.
The technical scheme of the application can be applied to client equipment, such as the client equipment 100, and can also be applied to the background server 110, so that music information is pushed to a login user of the client equipment 100; it should be noted that the music information referred to in this application may be audio, video, or other audio-visual entertainment information besides audio. In addition, the client device 100 may also input somatosensory interaction information in an image manner, and thus, for the client device 100, it may be a somatosensory interaction device, which may include an image processing system, a system host, and an audio system; the image processing system can be used for acquiring image data associated with a login user, the system host is used for outputting music to the sound system for playing and receiving somatosensory image data fed back by the image processing system in real time, and the sound system is used for receiving the music output by the system host and playing.
Referring to fig. 2, fig. 2 is a flowchart of a music information pushing method according to the present application, and the technical solution of the present application may be utilized on the client device shown in fig. 1, based on which a user may automatically obtain music information of interest.
As shown in the figure, a music information pushing method includes:
and S10, acquiring the user information of the login user, and acquiring the user image information of the login user according to the user information.
The user can log in through the client device 100, each user corresponds to one account, and each account corresponds to binding user information and related record information of accessing, downloading and selecting music; the user image information can be obtained by the labeling information processing of the user.
And S20, reading the currently detected real-time interaction information of the login user.
In this step, the real-time interaction information of the login user may be detected by the client device 100, for example, in fig. 1, detected in real time by the motion capture device 120, input in real time by another device, or captured by a camera of the client device 100.
And S30, selecting target music from the music library according to the user portrait information and the real-time interaction information.
In the step, the user portrait information and the real-time interaction information are utilized to select the target music from the music library, and the user portrait information is user labels marked on the basis of historical information, so that the labels are difficult to reflect the real-time requirements and interest points of the user, and when the user carries out somatosensory interaction, the real-time interaction information of the user can reflect the current requirements for music pushing more easily.
The user can act as interaction information with the client device through voice or somatosensory action, so that the user can act as the client device to push music as an influence factor; for example, a user may input interaction information in a voice control manner as a reference for pushing music, and similarly, may also input interaction information, such as body movement between the user and the client device, as a reference for pushing music.
In one embodiment, the method for selecting the target music from the music library in step S30 may include the following steps:
(1) calling a pre-trained first recommendation model; the recommendation model is obtained by training samples based on user image information of the login user and music which is operated historically;
(2) and screening a candidate music set from the music library according to the interaction information, and inputting the portrait information of the user and the tag information of the music in the selected music list into the first recommendation model to obtain the target music.
According to the scheme of the embodiment, firstly, the interaction information is utilized to quickly screen out the candidate music set, the music types which are not currently interested by the login user are screened out, and accurate classification is firstly carried out; and then, the pre-trained first recommendation model is used for recognition, and the target music can be obtained and pushed to the login user by inputting user image information of the login user, operation on the music and the like into the first recommendation model.
And S40, pushing the target music to the login user.
In the step, the screened target music is pushed to the login user, so that the current interest point of the user can be quickly and accurately identified, and the pushed music can be closer to the real requirement of the user.
According to the technical scheme, when music information is pushed, target music is selected from a music library by means of the currently detected real-time interaction information of the login user and the user portrait information, and then the target music is pushed.
In the scheme of the embodiment, when the target music is screened, the recommendation model can be trained by using the user portrait information, and the music which is interested by the user is identified through the user recommendation model, for example, when the recommendation model of the user is trained, the music which is operated once and comprises the log-in user history record can be used as a sample, and the operations comprise searching, downloading, listening trial or playing and the like; the information of the song name, the singer information, the song style, the record year and the like of the music operated by the user can be extracted, the user is labeled, and then the recommendation model is trained.
In one embodiment, in order to improve the accuracy of the trained recommendation model, before extracting samples and training, the method further comprises filtering the historical record information of the user to remove music samples which are disliked by the user. For example, considering that some music is disliked by the user and the trial listening is opened, the playing time of the music samples is generally not very long and is almost not repeated, so the music samples can be filtered before the training of the model, and the music samples which are disliked by the user and have short playing time are correspondingly deleted from the samples, and the music data are not used, so that the accuracy influence on the trained recommendation model is avoided.
In one embodiment, to improve recognition accuracy, reference is made to FIG. 3, which is a flow chart illustrating updating user portrait information; the music information pushing method can further comprise the following steps:
s201, obtaining historical record information of music operation of the login user; specifically, information such as music searched, downloaded, listened to on trial or played in the user history may be acquired.
S201, extracting characteristic information of the music according to the historical record information; wherein the characteristic information includes singer information, creation time or genre information, and the like.
S203, acquiring the user label of the login user according to the characteristic information, and updating the user image information according to the user label.
In the embodiment, music-related behaviors of the logged-in user are acquired through the history information, the characteristic information is extracted as an influence factor, and the user label is generated to update the user portrait information, so that behavior operations of the user can be continuously reflected to the user portrait information.
Generally, after a new user registers, favorite music is often pushed to the user for the convenience of using the user, user characteristics are difficult to judge through user information due to incompleteness of the user information during user registration, and meanwhile, the music pushed by adopting a platform unified model is easy to deviate from the real requirements of a login user, so that the scheme for more accurately acquiring the portrait information of the user is provided.
In one embodiment, for a user who initially logs in or a user who logs in less frequently, the following method may be used to quickly establish the initial user portrait information, as shown in fig. 4, where fig. 4 is a flowchart for establishing the initial user portrait information; the method specifically comprises the following steps:
s301, acquiring music information operated by the login user; specifically, in the early stage of the first login of the user or the relatively few login times, the operations of searching, downloading, playing and the like of the user are utilized, the operations related to the user are responded, and meanwhile the operations of the user are obtained, and the operations relatively reflect the actual requirements of the user.
S302, extracting key information of music operated by the login user, and searching the same type of music and other operated users according to the key information; specifically, similar music is searched through key information of the music, and similar users are found according to the similar music.
S303, acquiring corresponding user portrait information according to the user information of the other users; specifically, user portrait information of other users is used as an influence factor.
S304, carrying out statistical analysis on the user portrait information of other users to obtain the initial user portrait information of the login user.
According to the scheme of the embodiment, similar music is found through key information of music operated by the user, similar users are found through the key information, user portrait information of other users is used as an influence factor, initial user portrait information of the logged-in user is counted, and through the scheme, user portrait information can be quickly established for the user under the condition that the user information amount is insufficient, so that interested music can be pushed for the logged-in user accurately.
In addition, in order to improve the accuracy of pushing music, according to the technical scheme of the application, the pushed music can be screened based on the current sound effect mode of the client device, as shown in fig. 5, fig. 5 is a flowchart of screening the pushed music based on the sound effect mode.
Accordingly, the music information pushing method of the present application may further include the following:
s401, acquiring a current sound effect mode of the client equipment; specifically, a sound effect mode used by the client device currently used by the login user is obtained, and the current sound effect mode may be considered as a point of interest for the user to select the type of music.
S402, determining the currently interested music type of the login user according to the sound effect mode, and selecting a music set to be recommended from the music library according to the music type.
For example, taking dance music as an example, the sound effect mode of dance music includes a plurality of dance style classifications: dancing in china, rock, disco, square dance, hip hop, HI-POP, taiji, etc.; similarly, if the sound effect mode corresponding to the taiji direct broadcasting music is selected, the user is considered to be inclined to select the taiji direct broadcasting music; if the user selects the mix sound effect mode of karaoke, it can be considered that the user tends to select karaoke music.
In addition, music is pushed through the sound effect mode, the selection operation time of a user can be saved, when the user places the client device in a certain sound effect mode, the client device/the background server can automatically go to the background to search for related music, the pushing precision and the intelligent degree are greatly improved, and the user experience is improved.
S403, selecting target music from the music set according to the user portrait information and the interaction information.
Specifically, after the music set to be recommended is selected, the target music can be further selected according to the user portrait information and the interaction information; it should be noted that if two factors, namely user portrait information and interaction information, are not considered, the music collection to be recommended may be pushed to the login user, or may be filtered in combination with other conditions and then pushed to the login user. If two factors of the user portrait information and the interaction information are considered, the music set to be recommended is further filtered from the two dimensions, and then the target music is obtained.
For example, the user may request a song by voice through the wireless microphone, and send interactive information, such as: the user can push a plurality of songs of different versions of the forgetting water on the client equipment, and the different versions can be selected by combining the user portrait information, such as an original singing version, a DJ version or a singing version. ,
according to the technical scheme of the embodiment, music is pushed based on the sound effect mode of the client device, the pushing precision can be improved, the intelligent degree of music searching of a user can be improved, and the user experience is further improved.
According to the technical scheme of the embodiment, the first recommendation model is obtained by training based on user portrait information and music operated by history of the user portrait information as a sample, and is obtained by training based on historical record data, so that the response to interest change of the user is slow, the most accurate interest point and the real requirement of the user can not be predicted conveniently, and the pushing accuracy is improved in order to respond to the change of the user timely; referring to fig. 6, fig. 6 is a flowchart for pushing music based on a user's trend of change.
Accordingly, the music information pushing method of the present application may further include the steps of:
s501, acquiring music operated by the login user within a recent time period; specifically, when the recommendation model is trained by using the historical record data, the music condition of the user which is operated within a period of time in the near future is counted, such as searching, downloading or playing music, watching audio and video, accessing an audio and video page, and the like.
S502, predicting the interest change trend of the login user according to the operated music; specifically, the information such as the type and style of the music is used to predict whether the user has a point of interest transition.
S503, updating the user portrait information of the login user and changing the output result of the first recommendation model according to the change trend; specifically, when the user portrait information is updated, the user portrait information is not only dependent on the historical record information, but also further utilizes the change trend, and when the interest change of the user is predicted, the user portrait information is timely adjusted to a new interest direction, so that the response hysteresis of a recommendation model is avoided.
The proposal of the embodiment aims at the recommendation model trained based on the user portrait information, when the recommendation model is trained by using the user portrait information as a sample, the change trend of the user is further considered, and the change trend is used as an interference factor to act on the recommendation model training process, so that the trend direction transition of the recommendation model is increased, and the music liked by the user is more accurately pushed.
For example, a login user originally likes classical music, after new rock music is found, the true desirability of the rock music is higher, and the user portrait information is formed by adopting historical record information to train a recommendation model, so that the recognition result of the recommendation model can be converted from the classical music to the type of the rock music within a long time, the response to the user is very slow, and the accuracy of the recommendation result is not facilitated; by adopting the technical scheme, after the interest change trend of the login user is predicted to be rock music, the output result of the first recommendation model can be changed to be the rock music type, so that the interest point and the real requirement of the user can be determined more accurately.
In the foregoing embodiment, a technical solution for pushing interesting music for a login user is listed, and for a platform, as users and contents increase, after a piece of music is created, it needs to be added to a music library, and for new music in a new library, in order to push the new music to a suitable user more quickly and accurately, the present application further provides a new music pushing solution, and as shown in fig. 7, fig. 7 is a flow chart for pushing new music in a new library; specifically, the scheme comprises the following steps:
s601, acquiring new music stored in a warehouse; specifically, the newly stored new music may be original music provided by the platform, and in order to promote the new music better, the new music needs to be pushed to a suitable user group.
S602, labeling information on the new music; specifically, the labeling can be used for model training, so that the characteristic information of the music is formed, and the characteristic description is carried out on the music; such as the name of the music, the creator information, the genre, genre.
S603, inputting the label information into a pre-trained second recommendation model, and identifying a target user set; the second recommendation model is obtained by training samples based on label information of a plurality of pieces of music of the same type as the new music and user information corresponding to users with the plurality of pieces of music operated; as to which users the type of music was downloaded, on demand, etc., so that a second recommendation model can be built that identifies characteristic information for these user groups, and further identifies the tag input model for new music to target user combinations.
S604, selecting a plurality of users from the target user set, and pushing the new music to the users; specifically, the number of pushed users can be selected according to requirements.
Compared with a conventional full-network pushing mode of new music, the scheme of the embodiment processes the new music through a more scientific and technical means, and pushes the new music to a part of most suitable users, so that on one hand, the pushing effect is ensured, and on the other hand, the sense of reaction of the users who are not interested is avoided. In the implementation process, a second recommendation model is established for new music created by the creator to identify a target user group, and appropriate new music can be pushed for users who continuously select changes, so that new works can be better and more accurately pushed to the users, and user experience is improved.
In order to further improve the intelligent degree and the pushing accuracy of the pushed music, according to the technical scheme, the user using scene can be determined according to the somatosensory interaction information between the login user and the client device, and therefore the candidate music of the user can be screened out according to the application scene.
Accordingly, referring to fig. 8, fig. 8 is a flowchart of music push based on a user usage scenario; in one embodiment, the music information pushing method may further include:
s701, acquiring an environment image of the login user based on client equipment; specifically, interactive equipment can be felt to the client equipment, take camera function intelligence audio amplifier, mini KTV room etc. and these client equipment can be used for in the middle of the different scenes.
S702, determining the current application scene of the login user according to the environment image; specifically, the environment image acquired by the client device can determine a user usage scene, for example, a hotel scene, a KTV scene, a family scene, a square dance scene, a mini KTV room, and the like.
S703, screening out a music set which is classified correspondingly to the application scene from the music library according to the application scene; and then screening out a candidate music set from the music library according to the interaction information.
According to the scheme of the embodiment, different use scenes have different music selection requirements, and the use scenes are identified through the environment images, so that music suitable for the types of the use scenes can be pushed, the pushing accuracy is further improved, and the user experience is improved.
An embodiment of the music information pushing apparatus of the present application is described below, and referring to fig. 9, fig. 9 is a schematic structural diagram of the music information pushing apparatus of the present application, including:
an image acquisition unit 10 for acquiring user information of a login user and acquiring user image information of the login user according to the user information;
the interaction detection unit 20 is configured to read currently detected real-time interaction information of the login user;
a target selection module 30, configured to select a target unit from a music library according to the user portrait information and the real-time interaction information;
and the music pushing unit 40 is used for pushing the target music to the login user.
The music information pushing apparatus of this embodiment may execute a music information pushing method provided in the embodiments of the present disclosure, and the implementation principles thereof are similar, the actions performed by the modules in the music information pushing apparatus in the embodiments of the present disclosure correspond to the steps in the music information pushing method in the embodiments of the present disclosure, and for the detailed functional description of the modules in the music information pushing apparatus, reference may be specifically made to the description in the corresponding music information pushing method shown in the foregoing, and details are not repeated here.
An embodiment of a computer device of the present application is set forth below, the computer device comprising:
one or more processors;
a memory;
one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to: the music information pushing method according to any of the above embodiments is performed.
The following sets forth an embodiment of a computer-readable storage medium of the present application, which stores at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded by the processor and executes the music information pushing method of any of the above embodiments.
As in the foregoing embodiments, the client device of the present application may be a smart speaker, a somatosensory interactive device, or the like; in order to make application scenarios of the present application clearer, an embodiment of the motion sensing interactive device is given below by way of example.
Referring to fig. 10, fig. 10 is a schematic structural diagram of a client device according to an embodiment, and a technical solution of the music information pushing method according to the present application may be applied to the client device 100 shown in fig. 10, where the device includes: an image processing system 01, a system host 13, and an audio system 03; the image processing system 01 is connected to the system host 13, and the audio system 03 is connected to the system host 13.
Wherein the image processing system 01 is used for acquiring image data associated with a user; the system host 13 is configured to acquire audio data to be played, perform sound effect processing on the audio data, output the audio data to the sound system 03, and receive somatosensory image data fed back by the image processing system 01 in real time; the audio system 03 is configured to receive audio data output by the system host 13, perform sound effect processing on the audio data, and play the audio data.
As shown in fig. 10, the image processing system 01 includes an image pickup device 12 and an image recognition module 19, and the sound system 03 includes an audio processing circuit 14 and a speaker unit 15, wherein the image pickup device 12 is configured to pick up a real-time image of a user's somatosensory interaction and output image data in a specified format; the image recognition module 19 is configured to recognize the real-time image and extract the limb motion data for output. The system host 13 outputs audio data to the audio processing circuit 14, and the audio processing circuit 14 performs sound effect processing and then outputs the audio data to the speaker unit 15 for playing.
Illustratively, as shown in fig. 10, the client device 100 specifically includes: the system comprises a box body 11, a camera device 12, an image recognition module 19, a sound system 03 and a system host 13; the image processing system 01 comprises a camera device 12 and an image recognition module 19, wherein the camera device 12 acquires image data through the camera device 12, the camera device 12 is connected with a system host 13 through the image recognition module 19, and the system host 13 is connected with a loudspeaker unit 15 through an audio processing circuit 14; the camera 12 captures real-time motion images of the user in front of the box body 11 and sends the images to the image recognition module 19, and the image recognition module 19 performs recognition processing on the image data through a motion capture algorithm, captures body motion data and the like of the user, and then sends the relevant body motion data to the system host 13.
The image recognition module 19, the system host 13 and the audio processing circuit 14 are matched with each other, the system host 13 realizes related functions such as system control, analysis and processing, the image recognition module 19 is concentrated in image processing for capturing high-speed user limb actions, the audio processing circuit 14 is mainly responsible for processing audio data, the functions are matched with each other in a division manner, performance optimization combination is achieved, higher image data processing efficiency and sound effect processing effect are achieved, faster response speed is achieved, and user application experience is improved.
Illustratively, referring to FIG. 11, FIG. 11 is a schematic diagram of an electrical configuration of a client device of one embodiment; as shown in the figure, the audio processing circuit 14 includes a DSP chip 141, a digital-to-analog conversion chip 142 and a power amplifier module 143, which are connected in sequence, and a wireless microphone receiving module 144 connected to the DSP chip 141; the DSP chip 141 is connected to the system host 13, and the power amplifier module 143 is connected to the speaker unit 15. The operating principle of the audio processing circuit 14 is as follows: the DSP chip 141 performs sound effect processing on the audio data output by the system host 13; the digital-to-analog conversion chip 142 can convert digital signals into analog signals; the power amplifier module 143 amplifies the power of the analog audio data and outputs the amplified analog audio data to the speaker unit 15 for playing; the voice signal of the wireless microphone can be directly processed by the DSP chip 141 and then sent to the speaker unit 15 for playing, or can be input to the system host 13 as a control signal for processing.
Illustratively, as shown in fig. 11, in order to enrich the peripheral functions of the client device 100, a video signal interface 131a, a microphone peripheral interface 131b, and the like may also be provided on the system host 13; among them, the video signal interface 131a may connect an external display device, such as a television, a projector, or the like. The microphone peripheral interface 131b may connect an external microphone; the user can input the operation instruction through the keys of the microphone, and the system host 13 executes the function corresponding to the operation instruction, for example, the volume of the microphone is increased or decreased through the keys on the microphone; increasing or decreasing the volume of the music; cutting songs; the first yeast and the next yeast; original singing and accompanying singing are switched; mute and the like. Meanwhile, the system host 13 may also be connected to an infrared receiving module 132, a bluetooth module 133, a voice recognition module 134, and the like; the infrared receiving module 132 can receive a control signal of the infrared remote controller, the bluetooth module 133 can receive a control signal of the bluetooth remote controller (such as a wristwatch, a bracelet, a U-section microphone integrated with bluetooth transmission, etc.), and the voice recognition module 134 can recognize a voice signal input by the user through the pickup microphone 134A.
Illustratively, as shown in fig. 11, the system host 13 may further be provided with a motion capture peripheral interface 135 for connecting at least one external motion capture device 135A, wherein the motion capture device 135A may be a dance mat, a dance pedal, etc. for detecting and transmitting the step motion data of the user to the system host 13; the connection mode may be a wired connection or a wireless connection. For example, the USB connection mode can also adopt a wireless mode, such as Bluetooth, WIFI, 2.4G/5.8G and other connection modes. Preferably, the client device 100 may be connected to a plurality of motion capture devices 135A, and the camera 12 simultaneously captures real-time motion images of a plurality of users.
Exemplarily, referring to fig. 12, fig. 12 is an exploded view of a hardware structure of a motion sensing interactive device according to an embodiment; the front part of the box 11 is provided with a perforated panel 110, so as to avoid the vibration of the camera device 12 caused by the vibration of the speaker unit 15 during operation, the camera device 12 is fixed on the perforated panel 110 through a damping structure 111, and the perforated panel 110 and the box 11 can be flexibly connected through a third flexible damping piece 111 e.
As shown in fig. 12, the horn unit 15 of the present application preferably has a 2.1 two-channel design, specifically a "2.1 channel + inverter" design, and as shown, includes a two-voice coil center-to-center channel horn 154, a left channel tweeter 157L, and a right channel tweeter 157R, with an inverter 155 disposed on the back plate 112 of the housing 11. Preferably, the tweeter of the speaker unit 15 is mounted on the hole site 110 a; a plurality of circular holes 110a are arranged on the perforated panel 110, and the camera 12 and the speaker unit 15 (tweeter) can be fitted to any circular hole 110a on the perforated panel 110; a plurality of imaging devices 12 can be fitted to the circular hole 110 a; in addition, a plurality of tweeters are matched with the panel with holes 110 in combination with selection of the loudspeaker units 15, so that requirements of different functions and different costs can be met. Preferably, a speaker mesh 15A is provided between the speaker unit 15 and the perforated panel 110; pickup microphone 134A and voice recognition module 134 are disposed at the top of housing 11 to facilitate capturing of voice signals.
Illustratively, as shown in fig. 12, an infrared receiving module 132 is disposed at the front of the box 11 for receiving infrared signals; the power module 16 and the power amplifier module 143 are arranged on a PCB board card, and a shielding case 113 is arranged between the system host 13 and the power amplifier module 143 to avoid mutual interference; a peripheral interface 135, a microphone peripheral interface 131b, an external power supply interface, and the like may be provided at the rear of the backplate 112. In addition, an electric rotating base 18 may be provided under the housing 11, and the housing 11 is rotated by the rotating base 18 to control the shooting direction of the imaging device 12, thereby tracking the moving position of the user and shooting.
The foregoing is only a partial embodiment of the present application, and it should be noted that, for those skilled in the art, several modifications and decorations can be made without departing from the principle of the present application, and these modifications and decorations should also be regarded as the protection scope of the present application.

Claims (10)

1. A music information pushing method is characterized by comprising the following steps:
acquiring user information of a login user, and acquiring user image information of the login user according to the user information;
reading currently detected real-time interaction information of the login user;
selecting target music from a music library according to the user portrait information and the real-time interaction information;
and pushing the target music to the login user.
2. The music information pushing method according to claim 1, wherein the real-time interactive information is somatosensory interactive information of a logged-in user synchronously detected by the client device when music is played;
the client equipment comprises an image processing system, a system host and a sound system;
the image processing system is used for acquiring image data associated with a login user;
the system host is used for outputting music to the sound system for playing and receiving somatosensory image data fed back by the image processing system in real time;
and the sound system is used for receiving and playing the music output by the system host.
3. The method of claim 1, wherein selecting the target music from a music library according to the user profile information and the real-time interaction information comprises:
calling a pre-trained first recommendation model; the recommendation model is obtained by training samples based on user image information of the login user and music which is operated historically;
and screening a candidate music set from the music library according to the interaction information, and inputting the portrait information of the user and the tag information of the music in the selected music list into the first recommendation model to obtain the target music.
4. The music information pushing method according to claim 1, further comprising:
acquiring historical record information of music operation of the login user;
extracting characteristic information of the music according to the historical record information;
and acquiring a user label of the login user according to the characteristic information, and updating the user image information according to the user label.
5. The music information pushing method according to claim 1, further comprising:
acquiring music information operated by the login user;
extracting key information of music operated by the login user, and searching the same type of music and other operated users according to the key information;
acquiring corresponding user portrait information according to the user information of the other users;
and carrying out statistical analysis on the user portrait information of other users to obtain initial user portrait information of the login user.
6. The music information pushing method according to claim 2, further comprising:
acquiring a current sound effect mode of the client equipment;
selecting target music from a music library according to the user portrait information and the real-time interaction information, and the method comprises the following steps:
determining the currently interested music type of the login user according to the sound effect mode, and selecting a music set to be recommended from the music library according to the music type;
and selecting target music from the music set according to the user portrait information and the interaction information.
7. The music information pushing method according to claim 3, further comprising:
acquiring music operated by the login user within a recent time period;
predicting the interest change trend of the login user according to the operated music;
and updating the user portrait information of the login user according to the change trend and changing the output result of the first recommendation model.
8. The music information pushing method according to claim 1, further comprising:
acquiring new music which is newly put in a warehouse;
labeling the new music;
inputting the label information into a pre-trained second recommendation model, and identifying a target user set; the second recommendation model is obtained by training samples based on label information of a plurality of pieces of music of the same type as the new music and user information corresponding to users with the plurality of pieces of music operated;
selecting a number of users from the set of target users and pushing the new music to the users.
9. The music information pushing method according to claim 1, further comprising:
acquiring an environment image of the login user based on client equipment;
determining the current application scene of the login user according to the environment image;
before the candidate music set is screened out from the music library according to the interaction information, the method further comprises the following steps:
and screening out a music set which is classified correspondingly to the application scene from the music library according to the application scene.
10. A music information pushing apparatus, comprising:
the image acquisition unit is used for acquiring user information of a login user and acquiring user image information of the login user according to the user information;
the interaction detection unit is used for reading the currently detected real-time interaction information of the login user;
the target selection module is used for selecting a target unit from a music library according to the user portrait information and the real-time interaction information;
and the music pushing unit is used for pushing the target music to the login user.
CN202110917597.6A 2021-08-10 2021-08-10 Music information pushing method and device, computer equipment and storage medium thereof Pending CN113641902A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110917597.6A CN113641902A (en) 2021-08-10 2021-08-10 Music information pushing method and device, computer equipment and storage medium thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110917597.6A CN113641902A (en) 2021-08-10 2021-08-10 Music information pushing method and device, computer equipment and storage medium thereof

Publications (1)

Publication Number Publication Date
CN113641902A true CN113641902A (en) 2021-11-12

Family

ID=78420698

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110917597.6A Pending CN113641902A (en) 2021-08-10 2021-08-10 Music information pushing method and device, computer equipment and storage medium thereof

Country Status (1)

Country Link
CN (1) CN113641902A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115648847A (en) * 2022-10-26 2023-01-31 东莞市皓龙激光科技有限公司 Laser decoration method, system and storage medium
CN115648847B (en) * 2022-10-26 2024-05-24 东莞市皓龙激光科技有限公司 Laser decoration method, system and storage medium

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105787069A (en) * 2016-03-01 2016-07-20 中山大学深圳研究院 Personalized music recommendation method
CN105930429A (en) * 2016-04-19 2016-09-07 乐视控股(北京)有限公司 Music recommendation method and apparatus
CN106844660A (en) * 2017-01-23 2017-06-13 腾讯科技(深圳)有限公司 A kind of music recommends method and device
CN108920585A (en) * 2018-06-26 2018-11-30 深圳市赛亿科技开发有限公司 The method and device of music recommendation, computer readable storage medium
CN110083738A (en) * 2019-03-20 2019-08-02 广东工业大学 A kind of music recommended method and terminal device based on vocal print and context aware
CN110175245A (en) * 2019-06-05 2019-08-27 腾讯科技(深圳)有限公司 Multimedia recommendation method, device, equipment and storage medium
CN110275980A (en) * 2019-06-26 2019-09-24 徐州工业职业技术学院 One kind having an X-rayed music recommended method based on group
CN111439268A (en) * 2020-03-31 2020-07-24 重庆长安汽车股份有限公司 Method and device for actively providing personalized service, cloud server and automobile
CN112328849A (en) * 2020-11-02 2021-02-05 腾讯科技(深圳)有限公司 User portrait construction method, user portrait-based dialogue method and device
CN112541093A (en) * 2020-12-10 2021-03-23 新华网股份有限公司 Music recommendation method, system, storage medium and electronic device

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105787069A (en) * 2016-03-01 2016-07-20 中山大学深圳研究院 Personalized music recommendation method
CN105930429A (en) * 2016-04-19 2016-09-07 乐视控股(北京)有限公司 Music recommendation method and apparatus
CN106844660A (en) * 2017-01-23 2017-06-13 腾讯科技(深圳)有限公司 A kind of music recommends method and device
CN108920585A (en) * 2018-06-26 2018-11-30 深圳市赛亿科技开发有限公司 The method and device of music recommendation, computer readable storage medium
CN110083738A (en) * 2019-03-20 2019-08-02 广东工业大学 A kind of music recommended method and terminal device based on vocal print and context aware
CN110175245A (en) * 2019-06-05 2019-08-27 腾讯科技(深圳)有限公司 Multimedia recommendation method, device, equipment and storage medium
CN110275980A (en) * 2019-06-26 2019-09-24 徐州工业职业技术学院 One kind having an X-rayed music recommended method based on group
CN111439268A (en) * 2020-03-31 2020-07-24 重庆长安汽车股份有限公司 Method and device for actively providing personalized service, cloud server and automobile
CN112328849A (en) * 2020-11-02 2021-02-05 腾讯科技(深圳)有限公司 User portrait construction method, user portrait-based dialogue method and device
CN112541093A (en) * 2020-12-10 2021-03-23 新华网股份有限公司 Music recommendation method, system, storage medium and electronic device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115648847A (en) * 2022-10-26 2023-01-31 东莞市皓龙激光科技有限公司 Laser decoration method, system and storage medium
CN115648847B (en) * 2022-10-26 2024-05-24 东莞市皓龙激光科技有限公司 Laser decoration method, system and storage medium

Similar Documents

Publication Publication Date Title
CN108304441B (en) Network resource recommendation method and device, electronic equipment, server and storage medium
CN100489761C (en) Interactive voice reproduction
KR102290419B1 (en) Method and Appratus For Creating Photo Story based on Visual Context Analysis of Digital Contents
US8793727B2 (en) System and method for selecting audio/video content for presentation to a user in response to monitored user activity
CN103026673A (en) Multi-function remote control device
US20190220492A1 (en) Display apparatus and method of controlling the same
CN109657236B (en) Guidance information acquisition method, apparatus, electronic apparatus, and storage medium
CN106484856B (en) Audio frequency playing method and device
CN112511850B (en) Wheat connecting method, live broadcast display device, equipment and storage medium
KR20110110147A (en) Method and system for generating data for controlling a system for rendering at least one signal
CN107592339B (en) Music recommendation method and music recommendation system based on intelligent terminal
CN110047497B (en) Background audio signal filtering method and device and storage medium
CN100546267C (en) The system, device, method, recording medium and the computer program that are used for process information
JP2003259316A (en) Stream processing system and stream processing program
CN111416996A (en) Multimedia file detection method, multimedia file playing device, multimedia file equipment and storage medium
US10210545B2 (en) Method and system for grouping devices in a same space for cross-device marketing
CN109792502A (en) Information processing equipment, information processing method, program and information processing system
JP2009169795A (en) Information-processing device, method, and program
CN113641902A (en) Music information pushing method and device, computer equipment and storage medium thereof
CN104866477A (en) Information processing method and electronic equipment
CN105930522A (en) Intelligent music recommendation method, system and device
CN105657510A (en) Image display apparatus and information providing method thereof
US20160133243A1 (en) Musical performance system, musical performance method and musical performance program
KR101625900B1 (en) Smart Audio apparatus based on IoT
CN113127678A (en) Audio processing method, device, terminal and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination