A big data-based automatic score correction method

Abstract

The application discloses a kind of based on big data's music score automatic correction method, specifically related to music field, including steps S1), obtaining the audio data to be identified;S2), noise elimination;S3), audio conversion;S4), music score correction.The application is set by step S1) to step S4) compared with prior art, music source can be identified classification, and it is once and secondary noise reduction processing, so that the music source with noise reduction is obtained, make the music source with noise reduction input music score database, search high similarity music score with it, by automatic correction module, so it can be corrected, and for the music score that cannot be searched, also can be corrected by targeted search to part of node, to improve the universality of the method correction, compared with the traditional method of directly correcting music source, it is convenient for the correction of the music score, improve its correction accuracy while facilitating subsequent user's learning reference use.

Description

Technical Field

[0001] This invention relates to the field of music technology, and more specifically, to a method for automatic music score correction based on big data. Background Technology

[0002] Music is an art form and cultural activity whose medium is time-organized, regular sound waves (a type of mechanical wave). Its basic elements include dynamics, tonality, duration, and timbre. These basic elements combine to form the commonly used "formal elements" of music, such as rhythm, melody, harmony, as well as dynamics, tempo, mode, form, and texture. These formal elements constitute the means of musical expression. Different types of music may emphasize or omit certain elements. Music is performed using various instruments and vocal techniques, and is divided into instrumental music, vocal music (such as songs without instrumental accompaniment), and works that combine singing and instrumental music. In its most general form, music is described as an art form or cultural activity, including the creation of musical works (songs, melodies, symphonies, etc.), performance, evaluation of music, study of music history, and music teaching. Ancient Greek and Indian philosophers defined music as tones arranged horizontally as melody and vertically as harmony. Proverbs such as “harmony in every field” and “this is music to my ears” point to the idea that music is usually orderly and pleasant to listen to, but 20th-century composer John Cage thought that any sound could be music, saying, for example, “There is no noise, only sound.”

[0003] With the development of big data, music has become more popular in our lives. In our daily musical environment, we often encounter different kinds of music, and there are many traditional music creation methods. Composing music is an important part of the creation process. Some people learn music scores from the surrounding music during the process of music creation or learning. However, due to the limitations of the recording environment or venue for some music, the music source often has a certain amount of noise, and some audio scores need further correction and proofreading to meet the recording needs of score learners. If the score is generated by directly correcting the noisy audio, the accuracy of the score will be greatly reduced, which is not conducive to the user's subsequent music creation. Therefore, a big data-based automatic score correction method is proposed. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a method for automatic music score correction based on big data. The technical problem to be solved by the present invention is to improve the accuracy of music score correction.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for automatic music score correction based on big data, comprising the following steps;

[0006] S1) Obtain the audio data to be recognized.

[0007] S11) Identify the audio data and determine whether it is music audio or sheet music audio;

[0008] S12) If it is music audio, proceed to step S13); if it is sheet music audio, proceed directly to step S3).

[0009] S13) The noise recognition module identifies the music audio in step S12). If the music audio does not contain noise, proceed directly to step S3); if the music audio contains noise, proceed directly to step S2).

[0010] S2) Noise cancellation

[0011] S21) Identify the music audio containing noise and confirm it as noise audio in a specific environment;

[0012] S22) The classified music audio is processed by an adaptive filter to obtain the first noise reduction frequency;

[0013] S23) The music audio is interrupted at preset time intervals according to specific parameters in the corresponding noise environment to form multiple sets of interrupted audio;

[0014] S24) The noise reduction parameters corresponding to the environment in the noise reduction processing module are used to process the multiple sets of isolated audio formed after the first noise reduction to obtain the second noise reduction frequency.

[0015] S25) The multiple isolated audio segments after secondary noise reduction are spliced ​​and converted by the audio score conversion module to convert them into complete musical score audio.

[0016] S3), Audio Conversion

[0017] S31) Generate sheet music data from the audio of the sheet music in steps S12), S13), or S25) using the automatic sheet music generation module;

[0018] In steps S32 and S31, while generating the musical score data, the types of musical scores were also classified, and the pitch, notes, semitones, whole tones, altered notes, beats, and rhythms in the musical scores were also coordinated according to the corresponding musical score types.

[0019] S33) Input the music score after coordination in step S32) into the music score database via the Internet for data retrieval and comparison. If a highly similar music score is obtained, both the music score and the highly similar music score will proceed to step S41).

[0020] S4) Score Correction

[0021] S41) The automatic correction module corrects the pitch, notes, semitones, whole tones, altered notes, meter, and rhythm in the score.

[0022] S42) If there are minor differences between the corrected score and the compared score in step S41), the score will be modified based on the compared score. If there are major differences between the corrected score and the compared score in a certain node, the data of the 21 nodes before and after that node will be integrated and returned to step S33).

[0023] S43) For this node, a targeted search is performed in the music score database, and the music scores from various searches are integrated, and the music score with the highest similarity to this music score is selected as the basis for correction.

[0024] (S44) The corrected node data from step S43) is transmitted to the user via the network. The user makes a judgment. If it is qualified, the score correction is completed. If it is not qualified, the user returns to step S43) to select the second most similar score for correction until the user is satisfied.

[0025] In a preferred embodiment, the noise audio in the specific environment in step S21) includes industrial production noise, transportation noise, construction noise, and social life noise.

[0026] In a preferred embodiment, the sheet music database includes a dynamic update module for real-time updating of Internet sheet music data and a sheet music classification module for classifying different types of sheet music.

[0027] In a preferred embodiment, the sheet music database further includes a retrieval instruction module for quickly searching for corresponding sheet music and a cleaning module for identifying and removing useless sheet music data.

[0028] In a preferred embodiment, the automatic correction module includes an interlude correction module, a data comparison module, a pitch correction module, and a time control module.

[0029] In a preferred embodiment, the Internet is connected to cloud-based big data.

[0030] In a preferred embodiment, the classification of sheet music types in step S32) includes simplified musical notation, guitar sheet music, piano sheet music, electronic keyboard sheet music, accordion sheet music, erhu sheet music, dizi and xiao sheet music, saxophone sheet music, guzheng sheet music, and other sheet music.

[0031] The technical effects and advantages of this invention are as follows:

[0032] Compared with existing technologies, this invention, through steps S1) to S4), can identify and classify music sources and perform primary and secondary noise reduction processing to obtain noise-reduced music sources. These noise-reduced music sources are then input into a music score database, where highly similar scores are searched. An automatic correction module can then correct these scores. Furthermore, for scores that cannot be found, targeted correction can be performed on specific nodes, thus improving the breadth of the correction method. Compared to traditional methods that directly correct the music source, this invention facilitates the correction of the music score, improves its accuracy, and is also convenient for subsequent users to learn from and reference. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the embodiments thereof. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0034] This invention provides a method for automatic music score correction based on big data, comprising the following steps: S1) acquiring audio data to be identified.

[0035] S11) Identify the audio data and determine whether it is music audio or sheet music audio;

[0036] S12) If it is music audio, proceed to step S13); if it is sheet music audio, proceed directly to step S3).

[0037] S13) The noise recognition module identifies the music audio in step S12). If the music audio does not contain noise, proceed directly to step S3); if the music audio contains noise, proceed directly to step S2).

[0038] S2) Noise cancellation

[0039] S21) Identify the music audio containing noise and confirm it as noise audio in a specific environment;

[0040] The noise audio in the specific environment in step S21) includes industrial production noise, transportation noise, construction noise and social life noise. For noise in different environments, users can set different noise cancellation parameters for the noise reduction processing module to make the noise reduction processing module have a better noise cancellation effect.

[0041] S22) The classified music audio is processed by an adaptive filter to obtain the first noise reduction frequency;

[0042] A filter is a filtering circuit composed of capacitors, inductors, and resistors. A filter can effectively filter out a specific frequency point or frequencies other than that in a power line to obtain a power signal of a specific frequency, or eliminate a power signal after a specific frequency. A filter is a frequency selective device that allows specific frequency components in a signal to pass through while greatly attenuating other frequency components. By utilizing this frequency selective function of the filter, interference noise can be filtered out or spectrum analysis can be performed.

[0043] This process is converted into a time function of voltage or current through the action of various sensors, which is called the time waveform of various physical quantities, or signal. Because the independent variable time takes continuous values, it is called a continuous time signal, and is also commonly referred to as an analog signal. Filtering is an important concept in signal processing. In a DC regulated power supply, the function of the filter circuit is to minimize the AC component in the pulsating DC voltage, retain its DC component, reduce the output voltage ripple coefficient, and make the waveform smoother.

[0044] S23) The music audio is interrupted at a preset time interval according to the specific parameters of the corresponding noise environment to form multiple sets of interrupted audio. By setting multiple sets of interrupted audio, it is not only beneficial for the secondary noise reduction of subsequent audio, but also beneficial for the targeted inspection and correction of some nodes in step S43).

[0045] S24) The noise reduction parameters corresponding to the environment in the noise reduction processing module are used to process the multiple sets of isolated audio formed after the first noise reduction to obtain the second noise reduction frequency.

[0046] S25) The multiple isolated audio segments after secondary noise reduction are spliced ​​and converted by the audio score conversion module to convert them into complete musical score audio.

[0047] S3), Audio Conversion

[0048] S31) Generate sheet music data from the audio of the sheet music in steps S12), S13), or S25) using the automatic sheet music generation module;

[0049] In steps S32 and S31, while generating the musical score data, the types of musical scores were also classified, and the pitch, notes, semitones, whole tones, altered notes, beats, and rhythms in the musical scores were also coordinated according to the corresponding musical score types.

[0050] The classification of sheet music types in step S32) includes numbered musical notation, guitar sheet music, piano sheet music, electronic keyboard sheet music, accordion sheet music, erhu sheet music, dizi and xiao sheet music, saxophone sheet music, guzheng sheet music and other sheet music;

[0051] S33) Input the music score after coordination in step S32) into the music score database via the Internet for data retrieval and comparison. If a highly similar music score is obtained, both the music score and the highly similar music score will proceed to step S41).

[0052] In step S33), if there are multiple scores similar to the score to be corrected, the best one is selected, and so on, the score with the highest similarity is selected.

[0053] The sheet music database includes a dynamic update module for real-time updates of internet sheet music data and a sheet music classification module for categorizing different types of sheet music.

[0054] The internet is connected to cloud-based big data;

[0055] The sheet music database also includes a search indicator module for quickly searching for corresponding sheet music and a cleaning module for identifying and removing useless sheet music data;

[0056] By setting up the cleaning module, useless or erroneous scores in the score database can be cleaned up in a timely manner. This avoids comparing and correcting invalid scores with scores that need to be corrected, which would greatly reduce the accuracy of the automatic correction module.

[0057] The automatic correction module includes an interlude correction module, a data comparison module, a pitch correction module, and a time control module;

[0058] S4) Score Correction

[0059] S41) The automatic correction module corrects the pitch, notes, semitones, whole tones, altered notes, meter, and rhythm in the score.

[0060] S42) If there are minor differences between the corrected score and the compared score in step S41), the score will be modified based on the compared score. If there are major differences between the corrected score and the compared score in a certain node, the data of the 21 nodes before and after that node will be integrated and returned to step S33).

[0061] S43) For this node, a targeted search is performed in the music score database, and the music scores from various searches are integrated, and the music score with the highest similarity to this music score is selected as the basis for correction.

[0062] (S44) The corrected node data from step S43) is transmitted to the user via the network. The user makes a judgment. If it is qualified, the score correction is completed. If it is not qualified, the user returns to step S43) to select the second most similar score for correction until the user is satisfied.

[0063] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort. Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware.

[0064] Based on this understanding, the above technical solution, in essence or the part that contributes to the prior art, can be embodied in the form of a product. This computer product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments.

[0065] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for automatic music score correction based on big data, characterized in that: Includes the following steps; S1) Obtain the audio data to be recognized. S11) Identify the audio data and determine whether it is music audio or sheet music audio; S12) If it is music audio, proceed to step S13); if it is sheet music audio, proceed directly to step S3). S13) The noise recognition module identifies the music audio in step S12). If the music audio does not contain noise, proceed directly to step S3); if the music audio contains noise, proceed directly to step S2). S2) Noise cancellation S21) Identify the music audio containing noise and confirm it as noise audio in a specific environment; S22) The classified music audio is processed by an adaptive filter to obtain the first noise reduction frequency; S23) The music audio is interrupted at preset time intervals according to specific parameters in the corresponding noise environment to form multiple sets of interrupted audio; S24) The noise reduction parameters corresponding to the environment in the noise reduction processing module are used to process the multiple sets of isolated audio formed after the first noise reduction to obtain the second noise reduction frequency. S25) The multiple sets of segmented audio after secondary noise reduction are spliced ​​and converted by the audio sonic notation conversion module to make them into complete input audio; S3), Audio Conversion S31) Generate sheet music data from the audio of the sheet music in steps S12), S13), or S25) using the automatic sheet music generation module; In steps S32 and S31, while generating the musical score data, the types of musical scores were also classified, and the pitch, notes, semitones, whole tones, altered notes, beats, and rhythms in the musical scores were also coordinated according to the corresponding musical score types. Sheet music can be categorized into various types, including numbered musical notation, guitar sheet music, piano sheet music, electronic keyboard sheet music, accordion sheet music, erhu sheet music, dizi and xiao sheet music, saxophone sheet music, guzheng sheet music, and other sheet music. S33) Input the coordinated musical scores into the musical score database via the Internet for data retrieval and comparison. If highly similar musical scores are obtained, both the musical scores and the highly similar musical scores will proceed to step S41). S4) Score Correction S41) The automatic correction module corrects the pitch, notes, semitones, whole tones, altered notes, meter, and rhythm in the score. S42) If there are minor differences between the corrected score and the compared score in step S41), the score will be modified based on the compared score. If there are major differences between the corrected score and the compared score in a certain node, the data of the 21 nodes before and after that node will be integrated and returned to step S33). S43) For this node, a targeted search is performed in the music score database, and the music scores from various searches are integrated, and the music score with the highest similarity to this music score is selected as the basis for correction. S44) The corrected node data from step S43) is transmitted to the user via the network. The user makes a judgment. If it is qualified, the score correction is completed. If it is not qualified, the user returns to step S43) to select the second most similar score for correction until the user is satisfied. The noise audio in the specific environment mentioned in step S21) includes industrial production noise, transportation noise, construction noise, and social life noise. The music score database includes a dynamic update module for real-time updates of internet music score data and a music score classification module for classifying different types of music scores. The sheet music database also includes a retrieval instruction module for quickly searching for corresponding sheet music and a cleaning module for identifying and removing useless sheet music data. By setting up the cleanup module, useless or erroneous sheet music in the sheet music database can be cleaned up in a timely manner. The automatic correction module includes an interlude correction module, a data comparison module, a pitch correction module, and a time control module.

2. The method for automatic music score correction based on big data according to claim 1, characterized in that: The internet is connected to cloud-based big data.