Music generation method and training method of music generation model

Abstract

The invention discloses a music generation method, which comprises the following steps: obtaining a chord comprising a plurality of chords, each chord having a specified duration and comprising a plurality of notes; respectively converting the plurality of chords of the chord into a plurality of first MIDI messages, wherein each first MIDI message comprises a chord feature representing the corresponding chord and a note feature representing the plurality of notes of the corresponding chord; determining a chord arrangement sequence of the plurality of chords performed by the chord; inputting the multiple first MIDI messages and the chord arrangement sequence into a music generation model based on a natural language neural network, and sequentially generating multiple second MIDI messages corresponding to the multiple first MIDI messages and mapping the chord arrangement sequence, wherein each second MIDI message comprises note features representing multiple generated notes, the plurality of generated notes generally have the specified duration; and splicing the plurality of second MIDI messages into a generated music MIDI message according to the generation sequence.

Description

technical field [0001] The present application relates to the technical field of multimedia content processing, and in particular to a method for generating music. The present application also relates to a training method for a music generation model. Background technique [0002] Music creation involves a lot of professional knowledge. Currently, MIDI (Musical Instrument Digital Interface) technology has become a widely used technology in the field of music creation. With the popularization of various music applications, users' pursuit of music personalization is getting higher and higher, and more and more users want to create music by themselves. However, for non-professional ordinary users, it is often impossible to realize music creation due to lack of professional music knowledge and unfamiliarity with MIDI technology-related protocols, standards, or complex software and hardware operations. [0003] Moreover, even for professional users, it is very likely that due ...

AI Technical Summary

Problems solved by technology

However, for non-professional ordinary users, it is often impossible to realize music creation due to lack of professional music knowledge and unfamiliarity with MIDI technology-related protocols, standards, or complex software and hardware operations.

[0003] Moreover, even for professional users, it is very likely that due to time constraints, complex software and hardware operations in the creation process, and lack of ideas for modifying unsatisfactory works, etc., it is difficult to present creative ideas into desired music works

Images