Audio fingerprint coding method and device, computer equipment and storage medium

An audio fingerprint and coding method technology, applied in computer parts, calculations, calculation models, etc., can solve the problems of poor coding accuracy of audio fingerprints and the inability of audio fingerprints to accurately reflect different audio characteristics, and achieve the effect of improving coding accuracy.
CN111428078APending Publication Date: 2020-07-17TENCENT TECH (SHENZHEN) CO LTD
0 Cites 4 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
TENCENT TECH (SHENZHEN) CO LTD
Publication Date
2020-07-17

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to an audio fingerprint encoding method and device, computer equipment and a storage medium. The method comprises the following steps: before feature points are extracted from aninitial spectrogram of an audio signal, firstly, an initial spectrogram is processed through a weight matrix, a processed spectrogram is obtained, feature point pair extraction and coding are carriedout on the processed spectrogram; the weight matrix is obtained by training the audio sample and the actual user behavior characteristics of the user for the audio sample; when the training samples are enough, the user behavior characteristics can reflect the condition that different audios are concerned by the user; correspondingly, after the initial spectrogram is processed through the weight matrix obtained through training, the obtained processed spectrogram can highlight the possible attention part of the user in the audio, the obtained audio fingerprint can more accurately reflect the possible attention characteristic of the user, and therefore the encoding accuracy of the audio fingerprint is improved.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The embodiments of the present application relate to the technical field of artificial intelligence, and in particular to an audio fingerprint encoding method, device, computer equipment, and storage medium. Background technique

[0002] Nowadays, with the continuous development of network audio services, more and more audio applications provide users with audio matching services. Audio matching refers to matching the audio fingerprints of two audios, and the audio fingerprints can be obtained by encoding according to the spectrogram of the audio.

[0003] In related technologies, when audio is encoded to obtain an audio fingerprint, the spectrogram of the audio can be obtained first, and feature points with large spectral values ​​are extracted from the spectrogram, and then selected from the extracted feature points in the preset frequency range Then match and encode the selected feature points to obtain the audio fingerprint of the audio.

[0004] ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More