A music automatic labeling method based on label depth analysis

Abstract

The invention discloses a music automatic labeling method based on label depth analysis. The method comprises the following steps: S1, collecting music data and cleaning the data by combining a musiclabel system; S2, sampling the music data, converting the music data into a Mel-frequency spectrogram, and slicing the Mel-frequency spectrogram; S3, constructing an audio multi-level feature extraction network based on the one-dimensional convolutional network, and performing parameter pre-training through supervised learning; S4, performing music label vector representation learning based on thetwo-dimensional convolutional network, and obtaining music label characteristics; S5, realizing feature aggregation of the audio multi-level features and the music tag features; and S6, performing final music label prediction based on the aggregation characteristics. According to the method, the difficulty that a traditional music labeling mode cannot be applied to a large-scale music data set isovercome, the music is automatically labeled according to the audio content, the workload of manually maintaining a music label library is reduced, and the method has very good usability.

Description

technical field [0001] The invention relates to the field of music information research, in particular to an automatic music tagging method based on tag depth analysis. Background technique [0002] In recent years, digital music has become increasingly popular, and the number of music that users can access on the Internet has exceeded 30 million. In addition, as users often post a large number of original songs, cover songs and other multimedia resources on social media, the types of music are increasingly diversified. As an effective means of organizing massive music data, rich music annotation information is of great value for online music products. Music annotation information also provides high potential economic value for music retrieval and music discovery services. Based on this, music automatic labeling, as an effective method for enriching music information, has received more and more attention and research. [0003] Deep neural networks have been successfully a...

AI Technical Summary

Problems solved by technology

These problems are an important reason for the label noise in the dataset, and the label noise in the dataset will lead to poor learning effect of the deep neural network.

Images