A midi multitrack sequence representation method and application
A sequence representation and sequence technology, applied in the field of music design, can solve problems such as unfavorable association information, ignoring music bar information, etc., and achieve a strong musical effect
Active Publication Date: 2022-08-09
ZHEJIANG UNIV +1
View PDF11 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] The purpose of the present invention is to provide a MIDI multi-track sequence representation method, to solve the problem that existing encoding methods ignore music bar information, which is not conducive to model learning related information between different bars
Method used
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
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0032] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and do not limit the protection scope of the present invention.
[0033] In order to solve the problem that the existing encoding method ignores the music bar information, it is not conducive to the model to learn the correlation information between different bars. This embodiment provides a MIDI multi-track sequence representation method, and the representation method is suitable for multi-track encoding, and has universality in the composition model based on deep learning.
[0034] Specifically, the MIDI multi-track sequence representation method provided by the embodiment comprises the following steps:
[0035] S101 , parse th...
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 PUM
Login to View More Abstract
The invention discloses a MIDI multi-track sequence representation method and application, comprising: parsing a MIDI file, splitting a MIDI message contained in the MIDI file into note performance information, note stop information, time transfer information and timbre transformation information; The performance information, note stop information, time transition information and timbre transformation information are converted into bar sequences, each bar sequence contains bar start marks and multiple intra-bar time step sequences, and each time step sequence contains time step labels arranged in sequence , track mark and note mark; wherein, a measure is divided into 32 time steps, the note mark is determined according to the note performance information, the note stop information, and the time transition information, and the track mark is determined according to the timbre transformation information. In order to solve the problem of ignoring the music measure information in the existing coding method, it is not conducive to the model to learn the correlation information between different measures.
Description
technical field [0001] The invention relates to the field of music design, in particular to a MIDI multi-track sequence representation method, and a method for generating music based on the MIDI multi-track sequence representation and utilizing a deep learning model. Background technique [0002] Music is an audio form that organizes the information of sound according to certain purpose and rules and is used to express emotions and ideas. With the rapid development of artificial intelligence and deep learning, many automatic composition techniques have been proposed. For automatic composition technology, music needs to first be transformed into a form that machines can understand. In order to serialize a musical score into a sequence that can be understood by a machine, many previous works have tried different encodings, which usually have their own applicable scenarios and advantages and disadvantages. [0003] The mainstream MIDI (Musical Instrument Digital Interface, mu...
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 Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G10H1/00G06K9/62G06N20/00
CPCG10H1/0025G06N20/00G06F18/214
Inventor 任意李晨啸张克俊
Owner ZHEJIANG UNIV