Drum music digital music score creating method and related device capable of automatically identifying common word scores and generating three-score contrast

Abstract

The invention discloses a digitized drumbeat music score making method capable of automatically identifying a common word score and generating a three-score contrast and a related device. The method comprises the following steps: acquiring the common word score; inputting into the trained YOLOv5 network for identification to obtain an identified common word spectrum; carrying out digitization and standardization processing on the recognized commonly-known character spectrum to generate a commonly-known character spectrum; adding stave lines to the generated commonly-known notation to obtain a double-notation contrast music score; and adding numbered musical notation lines into the music score with double-music-score comparison to obtain a drum music score file with three-music-score comparison. The method comprises the following steps: identifying handwritten and printed ancient notation characters through a computer vision technology, converting the notation characters into standardized digital formats, and establishing an original commonly-known notation database; according to the method, by developing the digital spectrum making method of three-spectrum comparison, the common word spectrum has flexible editability and mobility, standardization, standardization and digitization of the common word spectrum are achieved, the making efficiency and spectrum surface quality of the common word spectrum are greatly improved, and a large number of manpower and material resources are saved.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic music notation recognition and notation transcription, and in particular relates to a digital drum music notation transcription method and related devices for automatically recognizing colloquial notation and generating three-notation comparison. Background Art

[0002] Xi'an drum music is a representative work of humanity's intangible cultural heritage, a treasure of traditional Chinese culture, and known as a "living fossil of ancient Chinese music." Developed in Xi'an (ancient Chang'an) and surrounding areas, it originated from Tang Dynasty banquet music, later integrated into imperial court music, and spread to the general public during the An-Shi Rebellion, following the exile of imperial musicians. Xi'an drum music still maintains a remarkably complete repertoire, musical notation, structure, instruments, and performance styles. The repertoire is rich and diverse, featuring diverse modes and styles, and complex and extensive musical structures. Performances range from seated to mobile, and include suites, free-style songs, songs, and recitations.

[0003] Musical notation refers to various written forms and methods used to record music. Simplified notation and five-line notation are the dominant methods of musical notation today and are widely used in the study and dissemination of modern music. In ancient China, traditional Chinese music was primarily recorded using written text. Early Chinese musical notation consisted of handwritten scrolls—written notation. Throughout Chinese history, various musical notation systems have emerged, including lülü notation, sezi notation, and gongche notation. Additionally, notation based on the fingering of wind and plucked instruments exists, such as tablature notation and phoneme notation.

[0004] Xi'an drum music uses the "Song Dynasty Su Zi Pu" notation. Experts have discovered that the ancient notation for Xi'an drum music uses the "five-character gongche notation" system, with the notation "five-character gongche notation" (five-character gongche notation, upper one, four, six, hooked together) to represent the notes. This notation may be the same as the "Yan Yue Ban Zi Pu" notation system. Its notation style is similar to the Dunhuang Tangren Qupu (a manuscript dating from 933), including some of the same notation characters and beat symbols. Its origins can be traced back to the Tang and Song dynasties. Xi'an drum music Su Zi Pu is a rare example of musical notation in China that uses historical notation to record musical forms. The transmission of these notations has played a positive role in preserving ancient musical culture. However, due to the unique notation system and regional characteristics of Xi'an drum music, Su Zi Pu still uses traditional handwritten Su Zi Pu. Research since 1952 has revealed that approximately 100 Xi'an drum music notations have survived, containing over 3,000 drum music pieces. Experts have compiled over 1,000 pieces, but only 200 to 300 pieces are truly playable after being rhymed, leaving a significant number of scores yet to be "revived." In recent years, old artists who could rhyme music in various music societies have passed away one after another, and a large number of music scores have no one to rhyme, resulting in the loss and inheritance crisis of popular music scores!

[0005] With the development of computer technology, the digital preservation and notation recognition of traditional music have gradually gained attention. In the research on the automatic recognition of Xi'an drum music folk notation, a variety of background technologies jointly form the foundation for achieving this goal. First, the development of image processing, computer vision, pattern recognition, deep learning, and other technologies has made automatic recognition of Xi'an drum music folk notation possible, while natural language processing and database technology provide powerful tools for subsequent music analysis and processing. The combined application of these technologies provides strong technical support for the digital preservation and inheritance of traditional music culture. However, due to the lack of data on Xi'an drum music folk notation and the difficulty in collecting it, as well as the highly variable form of the notation characters, the automatic recognition of folk notation is relatively difficult.

[0006] Most existing music notation software supports the input and editing of both simplified and five-line musical notation. Mainstream notation software includes Sibelius, MuseScore, and Lilypond. Sibelius is known for its user-friendly interface and powerful editing features, making it suitable for a wide range of users, but it comes with a higher price tag. MuseScore is free and open-source software with comprehensive functionality, but may be slightly lacking in professional settings. Lilypond is a text-based notation software that generates high-quality musical notation through code. Its output is extremely beautiful, nearing publication quality, and it is completely free. It supports complex musical notation and highly automated operations, but requires certain programming knowledge, is less intuitive to operate, and lacks real-time feedback. Current digital notation processing technology for simplified and five-line musical notation is relatively mature, but it lacks the ability to display a comparison between simplified and five-line musical notation. These notation software programs have limited support for traditional vernacular notation and cannot directly compare or convert between the three.

[0007] In summary, the existing technology has the following obvious defects and deficiencies when processing Xi'an drum music folk notation:

[0008] (1) Insufficient data on Suzipu: The problem of insufficient Suzipu data has brought great challenges to the research and inheritance of Xi'an drum music. First, the number of existing scores is limited and the quality is uneven. Some scores are very old, the handwriting is blurred, and they are difficult to recognize, which increases the difficulty of sorting and research. Secondly, among the scores that have been sorted out, the proportion of songs that can actually be played is relatively low, which limits researchers' comprehensive understanding of the musical form and style characteristics of Xi'an drum music. In addition, Suzipu itself is a special way of notation. Its symbol system is complex and lacks systematicity, which places high demands on the professional quality of researchers. At the same time, due to the particularity of the inheritance of Xi'an drum music, the tradition of oral transmission has prevented many precious musical information from being preserved in the form of scores, further exacerbating the problem of insufficient data. These factors work together to make the research on Xi'an drum music face multiple difficulties such as lack of data, lack of information, and inheritance gaps, and effective measures are urgently needed to address them.

[0009] (2) Difficulty in recognizing the common characters: There are many difficulties in recognizing the common characters of Xi'an drum music, mainly due to the high variability of their forms. Due to differences in regions and styles of inheritors, the characters have gradually evolved into many variants, and their forms vary from person to person and lack standardized and unified standards. Therefore, the same symbol may be written in multiple ways, and even the meaning may vary in different musical scores. The personal style and unique calligraphy habits of artists make it difficult to standardize these characters. These factors have brought huge challenges to automatic recognition technology based on image processing and deep learning. Existing technologies are often unable to recognize these variants and cannot accurately restore the specific meaning of each symbol. In addition, the complex lines, subtle differences and fuzzy boundaries between variants of handwriting make it easy for the model to make misjudgments during the recognition process, affecting the accuracy of the final recognition.

[0010] (3) Insufficient support for digital notation of vernacular notation: Existing notation software and technologies do not support the notation function of vernacular notation. As a result, during the digital notation process, it is impossible to directly input or edit vernacular notation. It must rely on manual conversion or the use of other tools to achieve this indirectly. The traditional handwritten vernacular notation of Xi'an drum music is not only inefficient in production and requires a lot of manpower and material resources, but the inconsistency of the handwritten fonts leads to large differences in the display effect of the music score and uneven quality, which greatly reduces work efficiency.

[0011] (4) Unable to compare the three notations: Due to the different symbol systems and expression methods of vernacular notation, simplified notation, and five-line notation, existing technology cannot achieve automatic comparison and conversion between the three. The notation software on the market cannot even compare and display simplified notation and five-line notation. This often requires manual and tedious comparison work during the notation process, greatly increasing the difficulty of recording, interpreting, and disseminating Xi'an drum music.

[0012] The above technical deficiencies and shortcomings indicate that existing technologies are unable to automatically recognize, digitally record, edit, and disseminate vernacular notation for Xi'an drum music, let alone automatically compare and convert between vernacular notation, simplified notation, and five-line notation. Therefore, developing automatic recognition technology for vernacular notation, proposing efficient, high-quality, standardized, and flexible digital notation notation standards, and developing Xi'an drum music notation technology that compares the three notation styles are urgent technical challenges in the fields of digital music and intelligent music. Summary of the Invention

[0013] In view of the deficiencies in the prior art, the purpose of the present invention is to propose a method and related device for digitally notating drum music that can automatically identify vernacular notation and generate a comparison of three notations. The method can automatically identify vernacular notation and compare vernacular notation, five-line notation and simplified notation. It can not only automatically identify vernacular notation in various handwritten scripts efficiently and quickly, but also realize the digitization and standardization of the vernacular notation of Xi'an drum music, making the vernacular notation of drum music flexible in editability and portability. At the same time, a digital notation method for integrating three notations is proposed, which improves the efficiency and quality of music score production.

[0014] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0015] A digital drum music notation method for automatically identifying colloquial notation and generating three-notation comparison includes the following steps:

[0016] Get the common character spectrum;

[0017] The input is fed into the trained YOLOv5 network for recognition to obtain the recognized colloquial Chinese character spectrum;

[0018] The recognized colloquial Chinese characters are digitized and standardized to generate colloquial Chinese characters;

[0019] Add five-line staff lines to the generated popular notation to obtain a double-staff music score;

[0020] Add the simplified notation line to the double-notation score to get the triple-notation drum score file.

[0021] Furthermore, the trained YOLOv5 network is obtained through the following process:

[0022] The handwritten and printed notation characters in the music score collection are used as the dataset for the recognition of vernacular notation characters;

[0023] Clean, cut, and normalize the music score images in the dataset to obtain annotated files;

[0024] The sheet music image is passed through the YOLOv5 network to calculate the category and bounding box of the musical notation characters. Based on the difference between the category and bounding box of the musical notation characters and the annotation file, the loss value is calculated, and the YOLOv5 network parameters are optimized to obtain the trained YOLOv5 network.

[0025] Furthermore, the music score images in the dataset are cleaned, cut, and normalized to obtain the annotation files, which includes the following steps:

[0026] Denoise and enhance the music score images in the dataset to obtain cleaned data;

[0027] The cleaned data is cut and normalized, and labeled, and digital labels are assigned to the musical notation characters and symbols to obtain the labeled file.

[0028] Furthermore, the gradient descent algorithm is used to optimize the YOLOv5 network parameters.

[0029] Furthermore, adding stave lines to the generated popular notation to obtain a music score with double notation comparison includes the following steps:

[0030] Get the MXL file of Xi'an drum music;

[0031] Import the collected MXL files of Xi'an drum music into the database, including file name, pitch, rhythm, Suzipu, Banyan, and Hengha information;

[0032] Convert the MXL file in the database to LY file format;

[0033] Replace the lyrics of the first and third lines in the LY file while keeping the key information of the second line;

[0034] Load the command of the embedded Suzi notation font TTF file at the Suzi notation display position of the LY file to generate a music score with double notation comparison.

[0035] Furthermore, the MXL file of Xi'an drum music includes music pitch and rhythm information.

[0036] Furthermore, adding a simplified musical notation line to the music score with double-staff matching to obtain a drum music score file with triple-staff matching includes the following steps:

[0037] Automatically generate the musical notation corresponding to the note data based on the musical notation information of the LY file;

[0038] Convert the music score LY file data into simplified notation, add the simplified notation below the five-line notation, adjust the typesetting parameters, and obtain the drum score file with three-score comparison.

[0039] A digital drum music notation system with comparison between vernacular notation, five-line notation and simplified notation, comprising:

[0040] A module for obtaining the spectrum of common characters, used for obtaining the spectrum of common characters;

[0041] The vulgar character spectrum recognition module is used to input the trained YOLOv5 network for recognition and obtain the recognized vulgar character spectrum;

[0042] A vulgar character spectrum generation module is used to digitize and standardize the recognized vulgar character spectrum to generate a vulgar character spectrum;

[0043] A stave line adding module is used to add stave lines to the generated popular notation to obtain a music score with double notation comparison;

[0044] The simplified notation line adding module is used to add simplified notation lines to the music score with double notation comparison to obtain the drum music score file with triple notation comparison.

[0045] An electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a digital drum music notation method for comparing vernacular notation, five-line notation, and simplified notation is implemented.

[0046] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the digital drum music notation method that compares vernacular notation, five-line notation and simplified notation.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. Automatic recognition of common characters

[0049] This method uses computer vision technology to identify handwritten and printed ancient Chinese characters and converts them into a standardized digital format to build a database of original vernacular Chinese characters. Compared to traditional manual recognition, this method saves significant time and effort, reduces the potential for human error, and improves the efficiency of digital archiving of vernacular Chinese characters.

[0050] 2. Standardization and standardization of common character notations

[0051] The present invention creates a digital TTF font through handwritten folk musical notation symbols, so that the folk musical notation has flexible editability and portability, and realizes the normalization, standardization and digitization of the folk musical notation. It not only greatly improves the efficiency of notating folk musical notation and the quality of the musical notation, saves a lot of manpower and material resources, and ensures the display of a unified drum music notation on various digital platforms, solves the inconsistency problem caused by traditional handwritten folk musical notation in the process of inheritance, and overcomes the problem that the existing folk musical notation of drum music adopts the traditional handwriting method, which not only has low output efficiency and usually requires a lot of manpower and material resources, but also has a large difference in display effect and uneven quality due to the inconsistency of handwriting.

[0052] 3. Three-spectrum comparison function

[0053] The present invention achieves the purpose of converting five-line musical notation into simplified musical notation and proposes a method for notating music using a three-score comparison function: five-line musical notation, vernacular musical notation, and simplified musical notation. This three-score comparison function improves the completeness and uniformity of musical notation output, improves the modernization of traditional musical notation, enables readers to intuitively understand the correspondence between different musical notations, facilitates musicians to more quickly understand and perform drum music, and can greatly promote the teaching, research, and inheritance of China's excellent traditional musical culture, such as drum music. It overcomes the problem that existing musical notation software and technology cannot simultaneously display five-line musical notation, simplified musical notation, and vernacular musical notation, and cannot even achieve automatic conversion between five-line musical notation and simplified musical notation.

[0054] Furthermore, the present invention can embed snippets of popular Chinese notation for drum music and its three-spectrum comparison into LaTeX documents, enabling the output of any snippet of music at any location in a graphic publication (e.g., embedded in the text of an academic paper). This improves the accuracy and readability of academic research, greatly promotes interdisciplinary communication and integration, and facilitates the preservation and dissemination of music scores, thereby improving publishing efficiency and quality. Furthermore, the present invention supports MIDI output of music scores, facilitating music production and actual performance, expanding its application scenarios, and enhancing its practicality. This overcomes the problem that traditional drum music scores lack flexibility in display and output methods, and have very limited applications on different platforms.

[0055] Furthermore, the technology in this paper, through the development of Python scripts, can automatically process music score data in batches as needed, achieving a number of practical functions. This automated process significantly reduces manual operation time and the possibility of human error, making large-scale data processing and conversion more feasible, and has significant advantages in improving data processing efficiency and consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of the digital notation method for automatically identifying colloquial notations and generating three-notation comparisons according to the present invention;

[0057] Figure 2 It is a handwritten notation and a printed notation diagram;

[0058] Figure 3 The image before cutting and the image after segmentation (part);

[0059] Figure 4 Visualization images obtained for testing the trained model;

[0060] Figure 5 A diagram of common characters and heng-ha symbols written by a calligrapher;

[0061] Figure 6 This is an example of a three-spectrum music score;

[0062] Figure 7 Schematic diagram of a digital musical notation system that automatically recognizes colloquial notations and generates three-notation comparisons. DETAILED DESCRIPTION

[0063] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The accompanying drawings illustrate preferred embodiments of the present invention. However, the present invention may be implemented in a variety of different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the disclosure of the present invention.

[0064] See also Figure 1The present invention provides a digital drum music notation method for automatically identifying colloquial notation and generating three-notation comparisons, comprising the following steps:

[0065] 1. Select data source and data cleaning:

[0066] (1) Data source selection: The authoritative book on Xi'an drum music theory, "Xi'an Drum Music Encyclopedia", was selected, and the handwritten and printed notations in many music scores were collected. Figure 2 , as a dataset for common character font recognition.

[0067] (2) Data cleaning: Use image processing tools (OpenCV) to perform denoising and enhancement on the original images in the dataset. Remove interfering background (such as page background color and lines), improve clarity, and enhance the distinguishability of musical notation characters and symbols.

[0068] 2. Data cutting and normalization of the data set:

[0069] (1) Data Slicing: Using an image slicer, we extract each individual musical notation character or symbol from the cleaned musical notation image. Each musical notation character is sliced ​​into a separate image file, and the background is removed. If any residual background remains, we use image binarization to further optimize the image.

[0070] (2) Dataset normalization: Adjust the size of the image files after cutting so that all the word images are of the same size. Standardize the pixel value range to match the input requirements of the deep learning model, see Figure 3 .

[0071] 3. Label the data set:

[0072] Manual Annotation: Use the LabelImg toolkit to manually annotate the normalized fenzi images and assign numerical labels to each fenzi and symbol. Manually select each fenzi or symbol and record its bounding box coordinates (x, y, width, height), where x is the horizontal coordinate, y is the vertical coordinate, width is the width, and height is the height. Assign each fenzi a unique category label to generate an annotation file. The annotation file is stored in YOLO format, with each line representing the category and location of a fenzi.

[0073] 4. Train and test the model:

[0074] (1) Model training: In order to improve the model's adaptability to different fonts and symbol forms, data enhancement is performed through rotation, scaling, blurring, and noise. The dataset is divided into training, validation, and test sets in proportion to perform data segmentation. The loss function and optimizer are selected, and the optimizer, learning rate, and training rounds are set. Input the music score image and calculate the category and bounding box of the score characters through the YOLOv5 network. The loss value is calculated based on the difference between the model output and the annotation file. The model parameters are optimized using the gradient descent algorithm to reduce the loss. During the training process, the accuracy, recall rate, F1 score, and mAP (mean Average Precision) are monitored in real time to ensure that the model performance gradually improves and avoid overfitting. This model is used to automatically recognize handwritten colloquial Chinese scores.

[0075] (2) Model testing: Input unseen music sheet images and test the model’s recognition performance for handwritten and printed music sheet characters. The model returns the category label and the corresponding bounding box coordinates of the music sheet characters, thus achieving high-accuracy music sheet character recognition, with an accuracy rate of over 97%. Figure 4 .

[0076] 5. The fonts of the common character spectrum identified in step 4 are digitized and standardized to generate the common character spectrum.

[0077] (1) Calligrapher participation: Invite calligraphers to write the common folk characters and hengha symbols used in Xi'an drum music in handwritten form. Figure 5 , other people can write too.

[0078] (2) Digital processing: Use a high-precision scanner to digitize the handwritten Chinese characters and Hengha symbols to generate high-definition bitmap files;

[0079] (3) Font design: Use font design software to convert the digitized common characters and hum-ha symbols (i.e., high-definition bitmap files) into vector graphics, and standardize them according to the TTF (TrueType Font, a font file format) format to create a TTF font file for displaying common characters. The English capital letters AP represent the characters:

[0080] AB CDEFGHIJKLMNOP (After installing the font file, it will be displayed correctly:

[0081] ).

[0082] This allows for flexible editing and embedding of musical notation in most software and platforms.

[0083] (4) Verification and optimization: Through multiple tests, ensure that each glyph can be displayed correctly at different resolutions and zoom ratios, and that the glyph design conforms to traditional aesthetic standards.

[0084] 6. Convert the drum music score into a score that compares both the five-line score and the vernacular score, that is, add a five-line score row to the vernacular score generated in step 5.

[0085] (1) Data collection: Collect MXL files (MusicXML, Music eXtensible Markup Language, an open XML-based music notation file format for music exchange and distribution) of Xi'an drum music. These files should contain detailed pitch and rhythm information of the music, and input the lyrics lines aligned with the corresponding notes in the popular notation.

[0086] (2) Database establishment: Create a structured database and import the collected MXL files into the database. The fields include file name, pitch, rhythm, Suzipu, Banyan, Hengha and other information.

[0087] (3) Python programming: Write a Python script and call tools such as musicexp and musicxml to batch convert the MXL files in the database into the LY file format that can be read by the music notation software Lilypond.

[0088] (4) Conversion verification: Verify the converted LY file to ensure that the file structure is correct and can be successfully compiled in the music notation software Lilypond, and correctly display the music notation layout of Xi'an drum music staff.

[0089] (5) Lyric line replacement: According to the database information, the lyrics of the first line (vulgar words) and the third line (hengha) in the LY file are automatically replaced, while the beat information of the second line is kept.

[0090] (6) Font embedding: Load the command to embed the Suzi notation font TTF file in the Suzi notation display position of the LY file to ensure that Lilypond can correctly call and display the Suzi notation font created during typesetting. This will generate a music score that compares the five-line notation and Suzi notation.

[0091] 7. Add a line of simplified notation to the music score with double notation comparison.

[0092] (1) Simplified musical notation generation: Write the LY script "jianpu.ly" to automatically generate the simplified musical notation corresponding to the note data based on the musical notation information of the LY file.

[0093] (2) Calling the LY script: Call the written LY script "jianpu.ly" in the music score LY file to convert the music score LY file data into simplified notation, and add the simplified notation below the five-line notation. By adjusting the layout parameters in Lilypond, ensure that the display ratio of the five-line notation and simplified notation is appropriate to avoid overlapping or misalignment.

[0094] (3) Output music scores: Write a Python script "output_score.py" to batch export LY files containing the three-way music score comparison of five-line music score, simplified music score and vernacular music score into PDF music score files.

[0095] (4) Display comparison: The three-score comparison PDF score file can clearly show the differences in the scores of Xi'an drum music and their corresponding relationships, which is convenient for music research and teaching. Figure 6 .

[0096] 8. The PDF music score file with three-score comparison also has other functions

[0097] (1) LaTeX embedding: The three-score music scores or music score fragments can be displayed using LaTeX, so that the three-score music score layout of Xi'an drum music developed by this technology can be freely embedded in the text to meet professional typesetting needs.

[0098] (2) MIDI output: Given the built-in functionality of Lilypond, MIDI files can be output when using the Lilypond software to create music. Users can listen to the generated music by playing MIDI files, and the music can also be used as a MIDI database for subsequent research such as intelligent notation and AI rhyme.

[0099] The present invention collects all the popular notations of Xi'an drum music that have been preserved since 1953, and builds a digital music library of popular notations of Xi'an drum music, including music scores in PDF format and a music score library in MusicXML format containing digital music information such as pitch and rhythm.

[0100] Among them, steps 1-4 are the construction and training process of the model for identifying vulgar character notation; step 5 is used to design the font of the vulgar character notation, so that the vulgar characters identified in steps 1-4 can be displayed on the screen; step 6 adds five-line staff lines to the generated vulgar character notation; step 7 adds simplified staff lines based on the previous ones; step 8 is an additional function.

[0101] In this method, a musical score image is input into the YOLOv5 network to identify and calculate the categories and bounding boxes of the musical notation characters. The predicted results are compared with the annotation file, and a loss value is calculated. The magnitude of the loss value reflects the difference between the current model recognition result and the actual annotation, which is used to guide the parameter optimization direction during the model training process. Next, a gradient descent algorithm is used to continuously optimize the parameters of the YOLOv5 network based on the loss value, ultimately obtaining a trained model. This trained network can automatically detect and locate musical notation characters in new musical score images, achieving automatic recognition and extraction of musical notation characters, laying the foundation for subsequent music score analysis and digital processing.

[0102] See also Figure 7One embodiment of the present invention provides a digital drum music notation system that compares vernacular notation, five-line notation, and simplified notation, including:

[0103] The module for obtaining the fonts of Suzipu and MXL files is used to obtain the MXL files of Suzipu fonts and drum music;

[0104] The Suzipu font processing module is used to process the Suzipu font and obtain the Suzipu font TTF file;

[0105] A file conversion module is used to import the drum music MXL file into the database and convert the MXL file in the database into the music score LY file;

[0106] A loading module is used to load a command of a TTF file embedded with a Suzi notation font at the Suzi notation display position of the LY file, thereby forming a musical score with both the five-line notation and the Suzi notation in contrast;

[0107] The module for adding simplified musical notation lines is used to add simplified musical notation lines to the music score with double-staff notation comparison, and obtain a music score file containing the comparison of three notations: five-line notation, simplified musical notation and vernacular notation.

[0108] One embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for digitally notating drum music by comparing vernacular notation, five-line notation, and simplified notation is implemented.

[0109] One embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for digitally notating drum music by comparing vernacular notation, five-line notation, and simplified notation is implemented.

[0110] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0111] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A digital drum music notation method for automatically identifying colloquial notation and generating three-notation comparison, characterized in that: The following steps are involved: Get the common character spectrum; The input is fed into the trained YOLOv5 network for recognition to obtain the recognized colloquial Chinese character spectrum; The recognized colloquial Chinese character spectrum is digitized and standardized to generate a colloquial Chinese character spectrum; Add five-line staff lines to the generated popular notation to obtain a double-staff music score; Add the simplified notation line to the double-notation score to get the triple-notation drum score file.

2. The digital drum music notation method for automatically identifying common Chinese notation and generating three-notation comparison according to claim 1 is characterized in that: The trained YOLOv5 network is obtained through the following process: The handwritten and printed notation characters in the music score collection are used as the dataset for the recognition of vernacular notation characters; Clean, cut, and normalize the music score images in the dataset to obtain annotated files; The sheet music image is passed through the YOLOv5 network to calculate the category and bounding box of the musical notation characters. Based on the difference between the category and bounding box of the musical notation characters and the annotation file, the loss value is calculated, and the YOLOv5 network parameters are optimized to obtain the trained YOLOv5 network.

3. The digital drum music notation method for automatically identifying common Chinese notations and generating three-notation comparisons according to claim 2 is characterized in that: Cleaning, cutting, and normalizing the music score images in the dataset to obtain annotated files involves the following steps: Denoise and enhance the music score images in the dataset to obtain cleaned data; The cleaned data is cut and normalized, and labeled, and digital labels are assigned to the musical notation characters and symbols to obtain the labeled file.

4. The digital drum music notation method for automatically identifying common Chinese notations and generating three-notation comparisons according to claim 2 is characterized in that: The gradient descent algorithm is used to optimize the YOLOv5 network parameters.

5. The digital drum music notation method for automatically identifying colloquial notation and generating three-notation comparison according to claim 1 is characterized in that: Adding stave lines to the generated popular notation to obtain a double-staff musical score includes the following steps: Get the MXL file of Xi'an drum music; Import the collected MXL files of Xi'an drum music into the database, including file name, pitch, rhythm, Suzipu, Banyan, and Hengha information; Convert the MXL file in the database to LY file format; Replace the lyrics of the first and third lines in the LY file while keeping the key information of the second line; Load the command of the embedded Suzi notation font TTF file at the Suzi notation display position of the LY file to generate a music score with double notation comparison.

6. The digital drum music notation method for automatically identifying colloquial notation and generating three-notation comparison according to claim 5 is characterized in that: The MXL file of Xi'an drum music includes music pitch and rhythm information.

7. The digital drum music notation method for automatically identifying colloquial notation and generating three-notation comparison according to claim 1 is characterized in that: Adding a line of simplified notation to a double-notation score to obtain a triple-notation drum score file includes the following steps: Automatically generate the musical notation corresponding to the note data based on the musical notation information of the LY file; Convert the music score LY file data into simplified notation, add the simplified notation below the five-line notation, adjust the typesetting parameters, and obtain the drum score file with three-score comparison.

8. A digital drum music notation system that compares vernacular notation, five-line notation and simplified notation, characterized by: include: A module for obtaining the spectrum of common characters, used for obtaining the spectrum of common characters; The vulgar character spectrum recognition module is used to input the trained YOLOv5 network for recognition and obtain the recognized vulgar character spectrum; A vulgar character spectrum generation module is used to digitize and standardize the recognized vulgar character spectrum to generate a vulgar character spectrum; A stave line adding module is used to add stave lines to the generated popular notation to obtain a music score with double notation comparison; The simplified notation line adding module is used to add simplified notation lines to the music score with double notation comparison to obtain the drum music score file with triple notation comparison.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the digital notation method for drum music with comparison of vernacular notation, five-line notation and simplified notation as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the digital notation method for drum music with comparison of vernacular notation, five-line notation and simplified notation as described in any one of claims 1 to 7 is implemented.

Images