CN-122019828-A - Music pushing method and device and electronic equipment

CN122019828ACN 122019828 ACN122019828 ACN 122019828ACN-122019828-A

Abstract

The invention provides a music pushing method, a device and electronic equipment, wherein a plurality of candidate music are obtained, each candidate music has corresponding text content, the text content is generated based on relevant data of the corresponding candidate music through a text generation model, target music is determined from the plurality of candidate music based on preference parameters of a target user on the plurality of candidate music, the preference parameters are determined based on historical behavior data of the target user on the candidate music, and the target music and the target text content corresponding to the target music are pushed to the target user. According to the method, the target music is determined based on preference parameters of the target user on the candidate music, the target music and texts corresponding to the target music are sent to the target user, the texts are generated based on multi-dimensional labels of the music through an artificial intelligence technology, are attached to the target music, are smooth in logic, and enable the user to be attracted by the texts more easily, so that interests are generated for the target music, and user experience is improved.

Inventors

ZHU MINGZHI
CHEN JINHAI

Assignees

杭州网易云音乐科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (10)

1. A music push method, the method comprising: The method comprises the steps of obtaining a plurality of candidate music, wherein each candidate music has corresponding text content, generating the text content based on relevant data of the corresponding candidate music through a text generation model, wherein the relevant data comprises a multi-dimensional tag of the music, and the multi-dimensional tag comprises a plurality of metadata, lyric characteristics and user feedback data of the music; Determining target music from the plurality of candidate music based on preference parameters of the target user for the plurality of candidate music, the preference parameters being determined based on historical behavior data of the target user for the candidate music; And pushing the target music and the target text content corresponding to the target music to the target user.
2. The method of claim 1, wherein the text content is determined by: Generating a multi-dimensional label of the candidate music; And inputting the multidimensional labels corresponding to the candidate music into the text generation model, and outputting text contents corresponding to the candidate music through the text generation model.
3. The method of claim 2, wherein the text generation model includes text reference information for prompting text content features for indicating at least one of keywords, emotional expressions, text lengths, and text styles, and/or text structural features for indicating a location of a tag of a specified category in text content, the text reference information being determined based on historical text content; The step of inputting the multi-dimensional labels corresponding to the candidate music into the text generation model comprises the following steps: And inputting the multi-dimensional labels corresponding to the candidate music and the text reference information into the text generation model.
4. The method of claim 1, wherein the step of determining target music from the plurality of candidate music based on preference parameters of the target user for the plurality of candidate music comprises: Determining music to be removed from the plurality of candidate music based on historical music pushing data corresponding to the target user, wherein the time difference between the time when the music to be removed is pushed to the target user and the current time is smaller than or equal to a preset duration threshold; Removing the music to be removed from the plurality of candidate music; And determining target music from the candidate music after being removed based on preference parameters of the target user on the plurality of candidate music.
5. The method of claim 1, wherein after pushing the target music and the target text content corresponding to the target music to the target user, the method further comprises: acquiring behavior data of the target user in a specified duration aiming at the target music; determining behavior parameters of the target user for the target music based on the behavior data; and updating the text generation model based on the behavior parameters.
6. The method of claim 5, wherein the text generation model includes text reference information, the text reference information and a tag generated based on the related data of the candidate music are input to the text generation model to cause the text generation model to generate text content corresponding to the candidate music; Updating the text generation model based on the behavior parameters, comprising: And if the behavior parameters meet preset conditions, updating text reference information based on the target text content.
7. The method of claim 6, wherein the step of updating text reference information based on the target text content comprises: Analyzing and processing the target text content, and determining text content characteristics and text structure characteristics of the target text content; And updating the text reference information based on the text content characteristics and the text structure characteristics of the target text content.
8. A music pushing device, the device comprising: the system comprises a music acquisition module, a text generation module and a user feedback module, wherein the music acquisition module is used for acquiring a plurality of candidate music, each candidate music has corresponding text content, the text content is generated based on relevant data of the corresponding candidate music through the text generation model, and the relevant data comprises a multi-dimensional label of the music, wherein the multi-dimensional label comprises a plurality of metadata, lyrics characteristics and user feedback data of the music; The target music determining module is used for determining target music from the plurality of candidate music based on preference parameters of a target user on the plurality of candidate music, wherein the preference parameters are determined based on historical behavior data of the target user on the candidate music; And the music pushing module is used for pushing the target music and the target text content corresponding to the target music to the target user.
9. An electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor, the processor executing the machine executable instructions to implement the music pushing method of any of claims 1-7.
10. A machine-readable storage medium storing machine-executable instructions which, when invoked and executed by a processor, cause the processor to implement the music pushing method of any of claims 1-7.

Description

Music pushing method and device and electronic equipment Technical Field The invention relates to the field of artificial intelligence and technology, in particular to a music pushing method, a device and electronic equipment. Background In the related art, a music platform can generally generate music text content in two ways. The method comprises the steps of manually generating text contents for songs by operators and issuing the text contents to platform users in batches, and presetting a plurality of matching rules, wherein each rule corresponds to a set of text templates, and matching user and song information to replace placeholders of the text templates for issuing when issuing. The text content generated by the manual mode has low efficiency and serious content homogenization, so that the user experience is poor. The method for generating text content based on the text of the template is easy to cause the problems of logic confusion and inconsistent style when the matching degree of the template and the actual song information is not high, and can also cause poor user experience. Disclosure of Invention Accordingly, the present invention is directed to a music pushing method, apparatus and electronic device, so that a user is more likely to be attracted by a pushed text of pushed music, thereby generating interest in the music and improving user experience. In a first aspect, an embodiment of the present invention provides a music pushing method, where the method includes obtaining a plurality of candidate music, each candidate music has corresponding text content, the text content is generated based on relevant data of the corresponding candidate music through a text generation model, the relevant data includes a multi-dimensional tag of the music, the multi-dimensional tag includes metadata of the music, lyrics features, and multiple kinds of user feedback data, determining target music from the plurality of candidate music based on preference parameters of the target user for the plurality of candidate music, determining the preference parameters based on historical behavior data of the target user for the candidate music, and pushing the target music and the target text content corresponding to the target music to the target user. In a second aspect, an embodiment of the invention provides a music pushing device, which comprises a music obtaining module, a target music determining module and a music pushing module, wherein the music obtaining module is used for obtaining a plurality of candidate music, each candidate music is provided with corresponding text content, the text content is generated based on relevant data of the corresponding candidate music through a text generation model, the relevant data comprises a multi-dimensional label of the music, the multi-dimensional label comprises a plurality of metadata, lyrics characteristics and user feedback data of the music, the target music determining module is used for determining target music from the plurality of candidate music based on preference parameters of a target user on the plurality of candidate music, the preference parameters are determined based on historical behavior data of the target user on the candidate music, and the music pushing module is used for pushing the target music and the target text content corresponding to the target music to the target user. In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory, where the memory stores machine executable instructions executable by the processor, and the processor executes the machine executable instructions to implement the music pushing method described above. In a fourth aspect, embodiments of the present invention provide a machine-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, cause the processor to implement the music pushing method described above. The embodiment of the invention has the following beneficial effects: The music pushing method, the device and the electronic equipment acquire a plurality of candidate music, each candidate music has corresponding text content, the text content is generated through a text generation model based on relevant data of the corresponding candidate music, the relevant data comprise multi-dimensional labels of the music, the multi-dimensional labels comprise metadata of the music, lyrics features and multiple types of user feedback data, target music is determined from the plurality of candidate music based on preference parameters of a target user for the plurality of candidate music, the preference parameters are determined based on historical behavior data of the target user for the candidate music, and the target music and the target text content corresponding to the target music are pushed to the target user. According to the method, the target music is determined based on preference parameters