Song discrimination method, computer device, and storage medium
By acquiring the characteristic parameters and timbre features of the human voice signal in the song, and using the song identification model and timbre comparison, a comprehensive judgment is made on whether the song is a synthesized song. This solves the inaccuracy problem of traditional identification methods and achieves higher identification accuracy.
CN117765976BActive Publication Date: 2026-06-05TENCENT MUSIC ENTERTAINMENT TECH (SHENZHEN) CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TENCENT MUSIC ENTERTAINMENT TECH (SHENZHEN) CO LTD
- Filing Date
- 2023-12-11
- Publication Date
- 2026-06-05
Smart Images
Figure CN117765976B_ABST
Abstract
The application relates to a song identification method, a computer device and a storage medium, relates to the technical field of audio processing, and can improve the accuracy of an identification result of a synthesized song voice. The method comprises the following steps: acquiring a feature parameter corresponding to a human voice signal in a to-be-identified song; inputting the feature parameter corresponding to the human voice signal into a trained song identification model, outputting first identification information of the to-be-identified song by the song identification model according to the feature parameter corresponding to the human voice signal; comparing a timbre corresponding to the human voice signal in the to-be-identified song with each reference timbre, and determining second identification information of the to-be-identified song according to a comparison result; the reference timbre comprises a timbre corresponding to a synthesized song voice; and determining an identification result of the to-be-identified song according to the first identification information and the second identification information.
Need to check novelty before this filing date? Find Prior Art