Method and device for generating chord for melody based on perception curve

Through a perceptual curve-based method, combined with neural networks and mathematical models, chord sequences matching melody are generated, which solves the problem that it is difficult to capture the complexity of human auditory perception when melody generation chords in the prior art, and achieves high-quality and diverse music generation.

CN119724133APending Publication Date: 2025-03-28SHANGHAI UNIV
3 Cites 0 Cited by

Patent Information

Application Number
CN202411846282.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-16
Publication Date
2025-03-28

AI Technical Summary

Technical Problem

When generating chords for melody, it is difficult for the prior art to effectively capture the complexity and diversity of human auditory perception, resulting in the mediocrity and popularization of the generated music, and strictly limit the content of the chord database and the loss of complex chords in human works.

Method used

Using a method based on perception curves, a neural network model is combined with a mathematical model to generate the perceptual curve values ​​and time series that match the melody pitch and time series, and the candidate chord sequence is obtained through reverse generation, and the optimal chord sequence is finally selected.

Benefits of technology

It improves the quality and diversity of generated music, ensures that the chord sequence is more in line with human auditory preferences, has a higher sense of harmony and nature, and at the same time expands the types of chords that can be generated, reducing the requirements for music theory knowledge.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of audio processing, in particular to the technical field of computer-aided music generation. The invention provides a method and a device for generating chords for melody based on a perception curve. The method comprises the following steps: acquiring a training data set of deep learning; training a neural network model by using the obtained data set; generating a corresponding sensing curve sequence by using the completely trained neural network model; further regulating and controlling the predicted perceptual curve sequence; on the basis of the regulated sensing curve sequence, a candidate chord sequence is obtained through mathematical model calculation; and selecting the highest ranked chord sequence as a final chord sequence according to the candidate chord sequences. According to the method, the quality of a finished product generated by AI is improved, the creation efficiency is kept, the types of chords capable of being generated are expanded, and creation tools with stronger sense of participation are provided, so that a creator can further standardize a generation result according to actual listening feeling, and meanwhile, complicated music knowledge is not involved.
Need to check novelty before this filing date? Find Prior Art