Sound effect recommendation method

Abstract

The invention discloses a sound effect recommendation method. The method comprises the following steps: acquiring a plurality of song listening stream data; acquiring a plurality of click stream data corresponding to the plurality of song listening stream data; combining the song listening stream data and the corresponding click stream data on the time axis; obtaining multiple pieces of sound effect evaluation sample data of the target sound effect, wherein each piece of sound effect evaluation sample data comprises a playing completion degree proportion and a target sound effect use proportion; setting a completion degree threshold value for the playing completion degree proportion; setting a sound effect use threshold value for the target sound effect use proportion; and judging whether the multiple pieces of sound effect evaluation sample data meet a preset standard or not, and if so, recommending the target sound effect. The invention further discloses a sound effect recommendation device, electronic equipment and a storage medium.

Description

technical field [0001] The present application relates to the technical field of multimedia content processing, and in particular to a sound effect recommendation method and device. The present application also relates to an electronic device and a storage medium. Background technique [0002] In music APPs, users' comments on sound effects are of great significance for the recommendation of sound effects or related push services and the improvement of sound effects. [0003] Existing sound effect evaluation or sound effect feedback requires users to actively participate and fill in pre-designed questionnaires. For example, users actively enter the sound effect feedback portal of music apps or actively participate in sound effect feedback plans, and then make choices on the preset questions on the evaluation page. For example, a review page for a sound effect would look like Figure 7 shown. [0004] Since the existing methods for sound effect evaluation or feedback mainly...

AI Technical Summary

Problems solved by technology

[0004] Since the existing methods for sound effect evaluation or feedback mainly rely on the active participation of users, there are problems of low user participation enthusiasm, resulting in fewer participating users and less data collection, which in turn leads to insufficient coverage of the songs being evaluated and insufficient user coverage.

Moreover, the effects of different sound effects applied to different songs and the degree of user acceptance are inconsistent. However, due to the way of questionnaires, only a small amount of data can be obtained, resulting in only a general evaluation of sound effects, and it is difficult to obtain specific evaluations of specific sound effects applied to specific songs.

In addition, due to the active operation of the user, it will interrupt the user's listening to music or other work and entertainment activities, causing trouble and poor experience to the user

Images