Singing evaluation method, apparatus, storage medium, and computing device
By using a multi-scale network model to process and fuse features of singing audio, the problem of singing evaluation without reference information is solved, and effective singing evaluation is achieved in various scenarios.
CN116524956BActive Publication Date: 2026-05-01HANGZHOU NETEASE CLOUD MUSIC TECH CO LTD
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202310504296.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-04
- Publication Date
- 2026-05-01
- Estimated Expiration
- 2043-05-04
AI Technical Summary
Technical Problem
Existing singing evaluation technologies require reference information and cannot be applied in scenarios without such information, thus limiting their application scenarios.
Method used
A multi-scale network model is used to extract features from the spectral features of singing audio, downsample at multiple time scales, perform convolution calculations and feature fusion, and combine with a multi-classifier to determine the evaluation result of the singing audio without the need for reference information.
Benefits of technology
It enables effective evaluation of singing audio without reference information, can obtain useful information at different time scales, and provides accurate singing evaluation results.
✦ Generated by Eureka AI based on patent content.
Abstract
Embodiments of the present disclosure provide a singing evaluation method and device, a storage medium and a computing device. The method comprises: performing feature extraction on obtained singing audio to obtain a sound spectrum feature of the singing audio; inputting the sound spectrum feature into a multi-scale network model for calculation; wherein the multi-scale network model comprises a model based on a convolutional neural network, and a convolutional layer in the multi-scale network model comprises convolutional kernels corresponding to multiple time scales; and determining a final evaluation result based on a probability distribution of different evaluation results output by the multi-scale network model.
Need to check novelty before this filing date? Find Prior Art