An audio fingerprint generation method, device and storage medium

Abstract

The present application discloses an audio fingerprint generation method, device, and storage medium, wherein the method includes: downsampling the target audio; performing spectrum analysis on the downsampled target audio to obtain the spectrum data of the target audio, and converting the spectrum of the target audio The data is combined into an audio image of audio data; the deep learning network is called to analyze the audio features contained in the audio image to obtain the audio fingerprint of the target audio. This application converts the high-dimensional target audio into a low-dimensional audio image by sequentially performing downsampling, spectrum analysis, and data combination on the target audio, and then uses a deep learning network to perform feature analysis on the audio image to obtain audio fingerprints. It can be seen that the present application quickly generates audio fingerprints by performing dimensionality reduction and deep learning on the target audio, and the audio fingerprints generated by this method are more accurate. Overall, this application provides an efficient, accurate and fast audio fingerprint generation method.

Description

technical field [0001] The present application relates to the field of audio data processing, in particular to an audio fingerprint generation method, device and storage medium. Background technique [0002] With the development of network technology, various network resources, such as audio resources, are carried on the network. Massive audio resources are convenient for people to download, but they also face the problems of copyright protection and audio management. The above problems can be solved by using audio fingerprint technology. Audio fingerprint technology refers to the method of extracting the audio fingerprint of the audio from the audio. The audio fingerprint contains the characteristic information of the audio and can be used as the identification information of the audio. By comparing the audio fingerprints between the audios to determine the similarity between the audios, you can quickly search for similar audios, such as listening to songs and recognizing ...

AI Technical Summary

Problems solved by technology

[0003] In the process of realizing the present invention, the inventor found that using the existing technology to extract audio fingerprints, when the audio has a slight change in the tune, it cannot be recognized, for example, there may be out of tune in the cover audio, etc.

Overall, an efficient audio fingerprint generation method is still missing

Images