Music recommendation method based on similarities

Abstract

The invention discloses a music similarity detection method based on mixed characteristics and a Gaussian mixed model. According to the basic thought, the method comprises the steps of using a gamma-tone cepstrum coefficient for conducting similarity detection, and using weighting similarities of various characteristics as a final detection result; providing a modulation spectrum characteristic based on a frame shaft, using the characteristic for representing a music long-time characteristic, and using the combination of the long-time characteristic and a short-time characteristic as the input of modeling in the next step; using the Gaussian mixed model for conducting modeling on the music characteristics, firstly, utilizing a dynamic K mean value method for conducting initialization on the model, then, using an expectation-maximization algorithm for conducting model training, obtaining accurate model parameters, and finally using a log-likelihood ratio algorithm for obtaining the similarities between the pieces of music. According to the music similarity detection method, the obtaining of the music characteristics is more sufficient and thorough, the accuracy degree of music recommendation is improved, the characteristic vector dimensionality can be reduced, the information memory content of a music database is reduced, and the accuracy degree of the music recommendation is improved.

Description

technical field [0001] The invention relates to an Internet music retrieval technology, in particular to a similarity-based music recommendation method. Background technique [0002] With the rapid development of network technology, a large amount of music is shared on the Internet, and the online music library is updated very quickly; the traditional music retrieval method is often carried out by entering the name of the song or singer, but when the music data is so huge, In the information age, it is difficult to find music whose songs and singers are unknown; music recommendation enables people to get the music they want more quickly. [0003] In recent years, many music websites have begun to provide music recommendation functions, but the recommendation effect is still not satisfactory to users. Many recommendations are only a list of popular songs, without personalized recommendations based on user interests; in view of the current situation where music recommendation ...

AI Technical Summary

Problems solved by technology

The disadvantages of this method are as follows: (1) use Mel-frequency cepstral coefficients to characterize the timbre features of music signals. There are certain limitations in the application; (2) The "bulldozer" distance is too dependent on the accuracy of the model and cannot reflect the time-varying nature of the signal, thus affecting the detection results of music similarity

Images