Construction method of music online teaching service platform

Abstract

The invention belongs to the technical field of music online teaching, and discloses a method for constructing a music online teaching service platform. Comprising the steps of collecting student behavior data, student recording data and standard music data; processing the student recording data to obtain enhanced recording data, and performing feature extraction on the enhanced recording data and the standard music data to obtain ontology features of the enhanced recording data and the standard music data; comparing the ontology characteristics of the enhanced recording data with the ontology characteristics of the standard music data, obtaining a music level rating coefficient according to a comparison result, and obtaining a music level grade of the student according to the music level rating coefficient; pushing corresponding online music courses to the trainees according to the trainee behavior data and the music level grades of the trainees; and corresponding online music courses are pushed to the students according to the preferences of the students and the music levels, so that the efficiency of music online teaching and the satisfaction of the students are further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of online music teaching, and more specifically, to a method for constructing an online music teaching service platform. Background Art

[0002] The patent application publication number CN109191952A discloses a network-based teacher-student platform interactive musical instrument learning system and a method for producing micro-MOOC courseware, including a teacher teaching classroom end, used to obtain multimedia video signals and / or micro-MOOC courseware; a server, including a memory, security controls and a processor; an online music teaching interface platform, including an APP and a website, used to obtain content including courseware interfaces from the server and push them to student terminals via the network; student terminals, including PADs, computers and / or mobile phones, are connected to the online music teaching interface platform via the network; it effectively improves students' learning efficiency and interest; effectively controls the learning process; effectively solves problems such as the shortage of excellent music teachers, high social training costs, and limited training venues. Through platform classroom learning, the entire teaching process is clarified, and lesson plans, learning progress, learning outcomes, attendance records, etc. are effectively and completely displayed to the public.

[0003] However, when the content including the courseware interface is obtained from the server and pushed to the student's terminal through the network, it is not pushed according to the student's hobbies and musical level, which leads to students' lack of interest in the pushed courseware interface and other content, making students feel that the pushed courseware interface and other content are difficult, which reduces students' satisfaction.

[0004] In view of this, the present invention proposes a method for constructing an online music teaching service platform to solve the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for constructing an online music teaching service platform, comprising: S1, collecting student behavior data, student recording data and standard music data;

[0006] S2. Process the student's recording data to obtain enhanced recording data, perform feature extraction on the enhanced recording data and the standard music data, and obtain ontological features of the enhanced recording data and the standard music data;

[0007] S3. Comparing the ontological features of the enhanced recording data with the ontological features of the standard music data, obtaining a music level rating coefficient based on the comparison result, and obtaining the student's music level grade based on the music level rating coefficient;

[0008] S4. Push corresponding online music courses to students based on their behavior data and their music proficiency levels.

[0009] Furthermore, the process of collecting student behavior data, student recording data and standard music data includes:

[0010] The student behavior data includes the number of times the student listens to a song and the frequency of the student listening to the song. The number of times the student listens to a song is the number of times the same piece of music is played, and the frequency of the student listening to the song is the frequency of the same piece of music being played. The standard music data is the music played by the composer.

[0011] The students select a piece of music and sing it, and the students' singing is recorded to obtain the students' recording data;

[0012] The number of times students listen to songs, the frequency of students listening to songs and standard music data are obtained through the music software in the students' mobile phones.

[0013] Furthermore, the process of processing the student's recorded data to obtain enhanced recorded data includes:

[0014] The student recording data is recorded as y(t ) , t is time, and the student recording data includes student voice data and background noise data;

[0015] Perform short-time Fourier transform on the student's recording data to obtain the student's recording spectrum data Y9f,t), where f is the frequency;

[0016] Estimating the background noise of the students' recorded data to obtain estimated background noise data;

[0017] Obtain estimated background noise spectrum data B based on the estimated background noise data 预估 (f, t);

[0018] Obtain the speech noise signal-to-noise ratio ε(f, t) and the recording noise signal-to-noise ratio γ(f, t) based on the obtained student recording spectrum data and the estimated background noise spectrum data;

[0019] in, E[*] means calculating the expected value of *;

[0020] in,

[0021] Obtaining an enhanced recording coefficient G(f, t) according to the obtained speech noise signal-to-noise ratio and the recording noise signal-to-noise ratio;

[0022] in, I0 and I1 are modified Bessel functions;

[0023] Obtain enhanced recording spectrum data U(f, t) according to the obtained enhanced recording coefficient;

[0024] Among them, U(f, t)=G(f, t)×Y(f, t);

[0025] Perform inverse short-time Fourier transform on the obtained enhanced recording spectrum data to obtain enhanced recording data, and record the obtained enhanced recording data as u(t).

[0026] Furthermore, the process of performing short-time Fourier transform on the student's recording data to obtain the student's recording spectrum data includes:

[0027] Set a window function w(t), which intercepts the student's recording data from the starting position of the student's recording data to obtain a section of the student's recording data. The window function moves until it moves to the end of the student's recording data. Each time it moves, a section of the student's recording data is intercepted to obtain g sections of the student's recording data, where g is a positive integer. Perform a short-time Fourier transform on each section of the student's recording data to obtain the student's recording spectrum data;

[0028] in, τ is a time shift parameter, e is the base of the natural logarithm, j is the imaginary unit, π is the ratio of pi, y(τ) is the value of y(t) when t = τ, and w(τ-t) is the value of w(t) when t = τ-t.

[0029] Furthermore, the process of estimating the background noise of the student's recorded data to obtain estimated background noise data includes:

[0030] The student recording data is divided into frames to obtain T frames of student recording data. The obtained frames of student recording data are labeled as i, i = 1, 2, 3, ..., T, T is a positive integer, and the T frame of student recording data is recorded as y i (t);

[0031] Sampling is performed on each frame of student recording data to obtain the set y of values ​​corresponding to the sampling points of each frame of student recording data i (t ) =[y i，1 ,y i，2 ,y i，3 ,……,y i，N ]; N is the number of sampling points of each frame of student recording data, y i，1 ,y i，2 ,y i，3 ,……,y i，N The value corresponding to the sampling point of each frame of student recording data;

[0032] The energy E corresponding to each frame of student recording data is obtained according to the set of values ​​corresponding to the sampling points of each frame of student recording data. i ;

[0033] in, k is the number of the sampling point;

[0034] According to the set of values ​​corresponding to the sampling points of each frame of student recording data, the zero value frequency P corresponding to each frame of student recording data is obtained. i ;

[0035] in, y i，k+1 y i，k The value corresponding to the next sampling point of the corresponding sampling point;

[0036] Preset the energy threshold E0 and zero frequency threshold P0 according to actual conditions;

[0037] When E i <E0,P i ≥P0 or E i <E0,P i <P0 or E i ≥E0,P i When <P0, the frame of student recording data is marked as estimated background noise data, and the estimated background noise data is recorded as b(t);

[0038] Obtain the number of student recording data frames marked as estimated background noise data, record the number of student recording data frames marked as estimated background noise data as M, label the student recording data frames marked as estimated background noise data as z, z = 1, 2, 3, ..., M, and record the student recording data frames marked as estimated background noise data as x z (t);

[0039] Where, b(t)=[x1(t),x2(t),x3(t),……,x M (t)].

[0040] Furthermore, the process of obtaining estimated background noise spectrum data according to the estimated background noise data includes:

[0041] Perform short-time Fourier transform on the estimated background noise data to obtain initial estimated background noise spectrum data B(f, t);

[0042] Wherein, B(f, t)=[X1(f, t), X2(f, t), X3(f, t), ..., X M (f,t)];X1(f,t),X2(f,t),X3(f,t),…,X M (f, t) is the result of short-time Fourier transform of the student’s recording data frame marked as estimated background noise data;

[0043] Averaging the obtained initial estimated background noise spectrum data to obtain estimated background noise spectrum data;

[0044] in,

[0045] Furthermore, the process of extracting features from the enhanced recording data and the standard music data to obtain the ontological features of the enhanced recording data and the standard music data includes:

[0046] The intrinsic characteristics of the enhanced recording data include the wavelength and amplitude of the enhanced recording data;

[0047] The extraction cycle is set according to the actual situation, and the duration of the set extraction cycle is recorded as t0 (ms). The enhanced recording data is intercepted once every extraction cycle to obtain n enhanced recording sub-data. The obtained enhanced recording sub-data are labeled and recorded as a, a = 1, 2, 3, ..., n, where n is a positive integer. The obtained n enhanced recording sub-data are recorded as u a( t ) ;

[0048] The enhanced recording sub-data is sampled and processed to obtain a set u of values ​​corresponding to the sampling points of the enhanced recording sub-data a (t)=[u a，1 ,u a，2 ,u a，3 ,……,u a，A ]; A is the number of sampling points of the enhanced recording sub-data, u a，1 ,u a，2 ,u a，3 ,……,u a，A To enhance the value corresponding to the sampling point of the recording sub-data, the amplitude ZF of the enhanced recording sub-data is obtained according to the set of values ​​corresponding to the sampling points of the enhanced recording sub-data. a ;

[0049] Among them, ZF a =max{u a，1 ,u a，2 ,u a，3 ,……,u a，A};

[0050] Perform fast Fourier transform on the enhanced recording sub-data to obtain the enhanced recording sub-spectrum data U a (f a , t);

[0051] The wavelength BC of the enhanced recording data is obtained according to the obtained enhanced recording sub-spectrum data a ;

[0052] in, c is the speed of sound in the medium;

[0053] The same method as that used to obtain the ontological features of the enhanced recording data is used to synchronously obtain the ontological features of the standard music data, and the amplitude of the obtained standard music data is recorded as BZF a , the wavelength of the acquired standard music data is recorded as BBC a .

[0054] Furthermore, the ontological features of the enhanced recording data are compared with the ontological features of the standard music data, and a process of obtaining a music level rating coefficient based on the comparison results includes:

[0055] Get the amplitude difference CZF between the enhanced recording data and the standard music data a and wavelength difference CBC a ;

[0056] Among them, CZF a =|ZF a -BZF a |CBC a =|BC a -BBC a |;

[0057] Obtaining a music level rating coefficient PJ according to the amplitude difference and wavelength difference between the acquired enhanced recording data and the standard music data;

[0058] in, kp_1 is the adjustment coefficient of the amplitude difference, and kp_2 is the adjustment coefficient of the wavelength difference.

[0059] Furthermore, the process of obtaining the student's music level grade according to the music level rating coefficient includes:

[0060] The music level rating coefficient threshold range is set according to the actual situation, and the set music level rating coefficient threshold range is recorded as (PJ0, PJ1);

[0061] Obtaining the student's music level grade based on a comparison result between the obtained music level rating coefficient and a set music level rating coefficient threshold range;

[0062] The music level grades include first-level music level, second-level music level and third-level music level;

[0063] When PJ≤PJ0, the student’s music proficiency level is Grade 3 music proficiency;

[0064] When PJ0<PJ<PJ1, the student’s music level is level 2;

[0065] When PJ≥PJ1, the student's music level is the first-level music level.

[0066] Furthermore, the process of pushing corresponding online music courses to students based on student behavior data and student music proficiency levels includes:

[0067] Obtain the music types corresponding to the music that the student listens to the most times and the music that the student listens to most frequently, record the music type corresponding to the music that the student listens to the most times as the first music type, and record the music type corresponding to the music that the student listens to most frequently as the second music type;

[0068] The online music courses are divided into difficulty levels and types, and the difficulty levels of the online music courses include level one difficulty, level two difficulty, and level three difficulty;

[0069] When the student's music proficiency level is level three, online music courses of level three difficulty of the same type as the first music genre and online music courses of level three difficulty of the same type as the second music genre are pushed to the student;

[0070] When the student's music proficiency level is level two, online music courses of level two difficulty of the same type as the first music genre and online music courses of level two difficulty of the same type as the second music genre are pushed to the student;

[0071] When the student's music proficiency level is level one, online music courses of level one difficulty of the same type as the first music type and online music courses of level one difficulty of the same type as the second music type are pushed to the student.

[0072] The technical effects and advantages of the method for constructing an online music teaching service platform of the present invention are as follows:

[0073] Collect student behavior data, student recording data and standard music data, process the student recording data to obtain enhanced recording data, extract features from the enhanced recording data and the standard music data, obtain ontological features of the enhanced recording data and the standard music data, compare the ontological features of the enhanced recording data with the ontological features of the standard music data, obtain a music level rating coefficient based on the comparison result, obtain the student's music level grade based on the music level rating coefficient, push corresponding online music courses to students based on the student behavior data and the student's music level grade; push corresponding online music courses to students based on the student's preferences, so that the students are interested in the pushed online music courses, and push corresponding online music courses to students based on the student's music level grade, so that the pushed online music courses are of appropriate difficulty for the students, and students with higher music level grades can directly learn online music courses with higher difficulty, further improving the efficiency of online music teaching and student satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 This is a schematic diagram of a method for constructing an online music teaching service platform of the present invention. DETAILED DESCRIPTION

[0075] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0076] Example 1

[0077] See also Figure 1 As shown, the method for constructing a music online teaching service platform described in this embodiment includes: S1, collecting student behavior data, student recording data and standard music data;

[0078] S2. Process the student's recording data to obtain enhanced recording data, perform feature extraction on the enhanced recording data and the standard music data, and obtain ontological features of the enhanced recording data and the standard music data;

[0079] S3. Comparing the ontological features of the enhanced recording data with the ontological features of the standard music data, obtaining a music level rating coefficient based on the comparison result, and obtaining the student's music level grade based on the music level rating coefficient;

[0080] S4. Push corresponding online music courses to students based on their behavior data and music proficiency levels.

[0081] It should be further explained that, in the specific implementation process, the process of collecting student behavior data, student recording data and standard music data includes:

[0082] The student behavior data includes the number of times the student listens to a song and the frequency of the student listening to the song. The number of times the student listens to a song is the number of times the same piece of music is played, and the frequency of the student listening to the song is the frequency of the same piece of music being played. The standard music data is the music played by the composer.

[0083] The students select a piece of music and sing it, and the students' singing is recorded to obtain the students' recording data;

[0084] The number of times students listen to songs, the frequency of students listening to songs and standard music data are obtained through the music software in the students' mobile phones.

[0085] It should be further explained that, in a specific implementation process, the process of processing the student recording data to obtain enhanced recording data includes:

[0086] The student recording data is recorded as y(t), where t is time. The student recording data includes the student's voice data and background noise data.

[0087] Perform short-time Fourier transform on the student's recording data to obtain the student's recording spectrum data;

[0088] It should be further explained that, in a specific implementation process, the process of performing short-time Fourier transform on the student's recording data to obtain the student's recording spectrum data includes:

[0089] Set the window function w(t), which intercepts the student recording data from the starting position of the student recording data to obtain a section of the student recording data. The window function moves until it moves to the end of the student recording data. Each time it moves, a section of the student recording data is intercepted to obtain g sections of the student recording data, where g is a positive integer. Perform short-time Fourier transform on each section of the student recording data to obtain the student recording spectrum data Y(f, t); f is the frequency

[0090] in, τ is the time shift parameter, f is the frequency, e is the base of the natural logarithm, j is the imaginary unit, π is the circumference of pi, y(τ) is the value of y(t) when t = τ, and w(τ-t) is the value of w(t) when t = τ-t;

[0091] Estimating the background noise of the students' recorded data to obtain estimated background noise data;

[0092] It should be further explained that, in the specific implementation process, the process of estimating the background noise of the student's recording data and obtaining the estimated background noise data includes:

[0093] The student recording data is divided into frames to obtain T frames of student recording data. The obtained frames of student recording data are labeled as i, i = 1, 2, 3, ..., T, T is a positive integer, and the T frame of student recording data is recorded as y i (t);

[0094] Sampling is performed on each frame of student recording data to obtain the set y of values ​​corresponding to the sampling points of each frame of student recording data i (t) = [y i，1 ,y i，2 ,y i，3 ,……,y i，N ]; N is the number of sampling points of each frame of student recording data, y i，1 ,y i，2 ,y i，3 ,……,y i，N The value corresponding to the sampling point of each frame of student recording data;

[0095] The energy E corresponding to each frame of student recording data is obtained according to the set of values ​​corresponding to the sampling points of each frame of student recording data. i ;

[0096] in, k is the number of the sampling point;

[0097] According to the set of values ​​corresponding to the sampling points of each frame of student recording data, the zero value frequency P corresponding to each frame of student recording data is obtained. i ;

[0098] in, y i，k+1 y i，k The value corresponding to the next sampling point of the corresponding sampling point;

[0099] Preset the energy threshold E0 and zero frequency threshold P0 according to actual conditions;

[0100] When E i <E0,P i ≥P0 or E i <E0,P i <P0 or E i ≥E0,P i When <P0, the frame of student recording data is marked as estimated background noise data, and the estimated background noise data is recorded as b(t);

[0101] Obtain the number of student recording data frames marked as estimated background noise data, record the number of student recording data frames marked as estimated background noise data as M, label the student recording data frames marked as estimated background noise data as z, z = 1, 2, 3, ..., M, and record the student recording data frames marked as estimated background noise data as x z (t);

[0102] Where, b(t)=[x1(t),x2(t),x3(t),……,x M (t)];

[0103] Perform short-time Fourier transform on the estimated background noise data to obtain initial estimated background noise spectrum data B(f, t);

[0104] Wherein, B(f, t)=[X1(f, t), X2(f, t), X3(f, t), ..., X M (f,t)];X1(f,t),X2(f,t),X3(f,t),…,X M (f, t) is the result of short-time Fourier transform of the student’s recording data frame marked as estimated background noise data;

[0105] Averaging the obtained initial estimated background noise spectrum data to obtain estimated background noise spectrum data;

[0106] in,

[0107] Obtain the speech noise signal-to-noise ratio ε(f, t) and the recording noise signal-to-noise ratio γ(f, t) based on the obtained student recording spectrum data and the estimated background noise spectrum data;

[0108] in, E[*] means calculating the expected value of *;

[0109] in,

[0110] Obtaining an enhanced recording coefficient G(f, t) according to the obtained speech noise signal-to-noise ratio and the recording noise signal-to-noise ratio;

[0111] in, I0 and I1 are modified Bessel functions;

[0112] Obtain enhanced recording spectrum data U(f, t) according to the obtained enhanced recording coefficient;

[0113] Among them, U(f, t)=G(f, t)×Y(f, t);

[0114] Perform inverse short-time Fourier transform on the obtained enhanced recording spectrum data to obtain enhanced recording data, and record the obtained enhanced recording data as u(t0;

[0115] Wherein, u(t)=ISTFT(U(f,t)).

[0116] It should be further explained that, in a specific implementation process, the process of extracting features from the enhanced recording data and the standard music data to obtain the ontological features of the enhanced recording data and the standard music data includes:

[0117] The intrinsic characteristics of the enhanced recording data include the wavelength and amplitude of the enhanced recording data;

[0118] The extraction cycle is set according to the actual situation, and the duration of the set extraction cycle is recorded as t0 (ms). The enhanced recording data is intercepted once every extraction cycle to obtain n enhanced recording sub-data. The obtained enhanced recording sub-data are labeled and recorded as a, a = 1, 2, 3, ..., n, where n is a positive integer. The obtained n enhanced recording sub-data are recorded as u a( t ) ;

[0119] The enhanced recording sub-data is sampled and processed to obtain a set u of values ​​corresponding to the sampling points of the enhanced recording sub-data a (t)=[u a，1 ,u a，2 ,u a，3 ,……,u a，A ]; A is the number of sampling points of the enhanced recording sub-data, u a，1 ,u a，2 ,u a，3 ,……,u a，A To enhance the value corresponding to the sampling point of the recording sub-data, the amplitude ZF of the enhanced recording sub-data is obtained according to the set of values ​​corresponding to the sampling points of the enhanced recording sub-data. a ;

[0120] Among them, ZF a =max{u a，1 ,u a，2 ,u a，3 ,……,u a，A};

[0121] Perform fast Fourier transform on the enhanced recording sub-data to obtain the enhanced recording sub-spectrum data U a (f a , t);

[0122] Among them, U a (f a , t)=FFT(u a(t));

[0123] The wavelength BC of the enhanced recording data is obtained according to the obtained enhanced recording sub-spectrum data a ;

[0124] in, c is the speed of sound in the medium;

[0125] The same method as that used to obtain the ontological features of the enhanced recording data is used to synchronously obtain the ontological features of the standard music data, and the amplitude of the obtained standard music data is recorded as BZF a , the wavelength of the acquired standard music data is recorded as BBC a .

[0126] It should be further explained that, in the specific implementation process, the ontological features of the enhanced recording data are compared with the ontological features of the standard music data, and the process of obtaining the music level rating coefficient based on the comparison results includes:

[0127] Get the amplitude difference CZF between the enhanced recording data and the standard music data a and wavelength difference CBC a ;

[0128] Among them, CZF a =|ZF a -BZF a |CBC a =|BC a -BBC a |;

[0129] Obtaining a music level rating coefficient PJ according to the amplitude difference and wavelength difference between the acquired enhanced recording data and the standard music data;

[0130] in, kp_1 is the adjustment coefficient of the amplitude difference, and kp_2 is the adjustment coefficient of the wavelength difference.

[0131] It should be further explained that, in a specific implementation process, the process of obtaining the student's music level grade according to the music level rating coefficient includes:

[0132] The music level rating coefficient threshold range is set according to the actual situation, and the set music level rating coefficient threshold range is recorded as (PJ0, PJ1);

[0133] Obtaining the student's music level grade based on a comparison result between the obtained music level rating coefficient and a set music level rating coefficient threshold range;

[0134] The music level grades include first-level music level, second-level music level and third-level music level;

[0135] It should be further explained that, in the specific implementation process, the music level of students with a third-level music level is higher than that of students with a second-level music level, which is higher than that of students with a first-level music level;

[0136] When PJ≤PJ0, the student’s music proficiency level is Grade 3 music proficiency;

[0137] When PJ0<PJ<PJ1, the student’s music level is level 2;

[0138] When PJ≥PJ1, the student's music level is the first-level music level.

[0139] It should be further explained that, in a specific implementation process, the process of pushing corresponding online music courses to students based on student behavior data and student music level includes:

[0140] Obtain the music that the student listens to the most times and the music genre corresponding to the music that the student listens to most frequently, wherein the music genre includes classical, pop, rock, etc. The music genre corresponding to the music that the student listens to the most times is recorded as the first music genre, and the music genre corresponding to the music that the student listens to most frequently is recorded as the second music genre;

[0141] The online music courses are divided into difficulty levels and types. The difficulty levels of the online music courses include level one, level two, and level three. The types of the online music courses include classical, pop, rock, etc.

[0142] It should be further explained that, in the specific implementation process, the difficulty of the online music course of level 3 difficulty is greater than the difficulty of the online music course of level 2 difficulty is greater than the difficulty of the online music course of level 1 difficulty;

[0143] When the student's music proficiency level is level three, online music courses of level three difficulty of the same type as the first music genre and online music courses of level three difficulty of the same type as the second music genre are pushed to the student;

[0144] When the student's music proficiency level is level two, online music courses of level two difficulty of the same type as the first music genre and online music courses of level two difficulty of the same type as the second music genre are pushed to the student;

[0145] When the student's music proficiency level is level one, online music courses of level one difficulty of the same type as the first music type and online music courses of level one difficulty of the same type as the second music type are pushed to the student.

[0146] In this embodiment, corresponding online music courses are pushed to students based on their preferences, so that students are interested in the pushed online music courses, and corresponding online music courses are pushed to students based on their music proficiency levels, so that the pushed online music courses are of appropriate difficulty for students, and students with higher music proficiency levels can directly learn online music courses with higher difficulty, further improving the efficiency of online music teaching and students' satisfaction.

[0147] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0148] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only one type. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0149] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

[0150] Finally: The above description is only a preferred embodiment of the present invention and is 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 in the scope of protection of the present invention.

Claims

1. A method for constructing an online music teaching service platform, characterized in that: The method for constructing the online music teaching service platform includes: S1, collecting student behavior data, student recording data and standard music data; S2, processing the student's recording data to obtain enhanced recording data, extracting features from the enhanced recording data and standard music data to obtain the ontological features of the enhanced recording data and the standard music data; S3, comparing the ontological features of the enhanced recording data with the ontological features of the standard music data, obtaining a music level rating coefficient according to the comparison result, and obtaining the music level grade of the student according to the music level rating coefficient; S4. Push corresponding online music courses to students based on their behavior data and their music proficiency levels.

2. The method for constructing an online music teaching service platform according to claim 1, characterized in that: The process of collecting student behavior data, student recording data and standard music data includes: The student behavior data includes the number of times the student listens to a song and the frequency of the student listening to the song. The number of times the student listens to a song is the number of times the same piece of music is played, and the frequency of the student listening to the song is the frequency of the same piece of music being played. The standard music data is the music played by the music composer; The students select a piece of music and sing it, and the students' singing of the music is recorded to obtain the students' recording data; The number of times students listen to songs, the frequency of students listening to songs and standard music data are obtained through the music software in the students' mobile phones.

3. The method for constructing a music online teaching service platform according to claim 2, characterized in that: The process of processing the student's recorded data to obtain enhanced recorded data includes: The student recording data is recorded as y(t ) , t is time, and the student recording data includes student voice data and background noise data; Perform short-time Fourier transform on the student's recording data to obtain the student's recording spectrum data Y(f, t), where f is the frequency; Estimating the background noise of the students' recorded data to obtain estimated background noise data; Obtain estimated background noise spectrum data B based on the estimated background noise data 预估 (f, t); The speech noise signal-to-noise ratio ε(f, t) and the recording noise signal-to-noise ratio γ(f, t) are obtained according to the obtained student recording spectrum data and the estimated background noise spectrum data; in, E[*] means calculating the expected value of *; in, Obtaining an enhanced recording coefficient G(f, t) according to the obtained speech noise signal-to-noise ratio and the recording noise signal-to-noise ratio; in, I0 and I1 are modified Bessel functions; Obtain enhanced recording spectrum data U(f, t) according to the obtained enhanced recording coefficient; Among them, U(f, t)=G(f, t)×Y(f, t); The obtained enhanced recording spectrum data is subjected to inverse short-time Fourier transform to obtain enhanced recording data, and the obtained enhanced recording data is recorded as u(t).

4. The method for constructing a music online teaching service platform according to claim 3, characterized in that: The process of performing short-time Fourier transform on the student's recorded data to obtain the student's recorded spectrum data includes: Set a window function w(t), the window function intercepts the student recording data from the starting position of the student recording data to obtain a section of the student recording data, the window function moves until it moves to the end of the student recording data, intercepts a section of the student recording data each time it moves, and obtains g sections of the student recording data, where g is a positive integer, performs short-time Fourier transform on each section of the student recording data to obtain the student recording spectrum data; in, τ is a time shift parameter, e is the base of the natural logarithm, j is the imaginary unit, π is the circumference of a circle, y(τ) is the value of y(t) when t=τ, and w(τ-t) is the value of w(t) when t=τ-t.

5. The method for constructing an online music teaching service platform according to claim 4, characterized in that: The process of estimating the background noise of the student's recorded data to obtain estimated background noise data includes: The student recording data is divided into frames to obtain T frames of student recording data. The obtained frames of student recording data are labeled as i, i = 1, 2, 3, ..., T, T is a positive integer, and the T frame of student recording data is recorded as y i (t); Sampling is performed on each frame of student recording data to obtain a set of values ​​y corresponding to the sampling points of each frame of student recording data. i (t) = [y i，1 ,y i，2 ,y i，3 , ..., y i，N ]; N is the number of sampling points of each frame of student recording data, y i，1 ,y i，2 ,y i，3 , ..., y i，N The value corresponding to the sampling point of each frame of student recording data; The energy E corresponding to each frame of student recording data is obtained according to the set of values ​​corresponding to the sampling points of each frame of student recording data. i ; in, k is the number of the sampling point; According to the set of values ​​corresponding to the sampling points of each frame of student recording data, the zero value frequency P corresponding to each frame of student recording data is obtained. i ; in, y i，k+1 for y i，k The value corresponding to the next sampling point of the corresponding sampling point; Preset the energy threshold E0 and the zero-value frequency threshold P0 according to the actual situation; When E i <E0,P i ≥P0 or E i <E0,P i <P0 or E i ≥E0,P i When <P0, the frame of student recording data is marked as estimated background noise data, and the estimated background noise data is recorded as b(t); Obtain the number of student recording data frames marked as estimated background noise data, record the number of student recording data frames marked as estimated background noise data as M, label the student recording data frames marked as estimated background noise data as z, z = 1, 2, 3, ..., M, and record the student recording data frames marked as estimated background noise data as x z (t); Where, b(t)=[x1(t), x2(t), x3(t), …, x M (t)].

6. The method for constructing an online music teaching service platform according to claim 5, characterized in that: The process of obtaining estimated background noise spectrum data according to the estimated background noise data comprises: Performing short-time Fourier transform on the estimated background noise data to obtain initial estimated background noise spectrum data B(f, t); Where, B(f, t) = [X1(f, t), X2(f, t), X3(f, t), ..., X M (f,t)];X1(f,t),X2(f,t),X3(f,t),…,X M (f, t) is the result of short-time Fourier transform of the student's recording data frame marked as estimated background noise data; Averaging the obtained initial estimated background noise spectrum data to obtain estimated background noise spectrum data; in, 7. The method for constructing an online music teaching service platform according to claim 6, characterized in that: The process of extracting features from the enhanced recording data and the standard music data to obtain the ontological features of the enhanced recording data and the standard music data includes: The main features of the enhanced recording data include the wavelength and amplitude of the enhanced recording data; The extraction cycle is set according to the actual situation, and the duration of the set extraction cycle is recorded as t0 (ms). The enhanced recording data is intercepted once every extraction cycle to obtain n enhanced recording sub-data. The obtained enhanced recording sub-data are labeled and recorded as a, where a=1, 2, 3, ..., n, where n is a positive integer. The obtained n enhanced recording sub-data are recorded as u a( t ) ; The enhanced recording sub-data is sampled to obtain a set u of values ​​corresponding to the sampling points of the enhanced recording sub-data a (t)=[u a，1 ,u a，2 ,u a，3 ,……,u a，A ]; A is the number of sampling points of the enhanced recording sub-data, u a，1 ,u a，2 ,u a，3 ,……,u a，A To enhance the value corresponding to the sampling point of the recording sub-data, the amplitude ZF of the enhanced recording sub-data is obtained according to the set of values ​​corresponding to the sampling points of the enhanced recording sub-data. a ; Among them, ZF a = max{u a，1 , u a，2 , u a，3 , …, u a，A }; Perform fast Fourier transform on the enhanced recording sub-data to obtain enhanced recording sub-spectrum data U a (f a , t); The wavelength BC of the enhanced recording data is obtained according to the obtained enhanced recording sub-spectrum data a ; in, c is the speed of sound in the medium; The same method as that used to obtain the ontological features of the enhanced recording data is used to synchronously obtain the ontological features of the standard music data, and the amplitude of the obtained standard music data is recorded as BZF a , the wavelength of the acquired standard music data is recorded as BBC a .

8. The method for constructing an online music teaching service platform according to claim 7, characterized in that: The process of comparing the ontological features of the enhanced recording data with the ontological features of the standard music data and obtaining the music level rating coefficient according to the comparison results includes: Get the amplitude difference CZF between the enhanced recording data and the standard music data a and wavelength difference CBC a ; Among them, CZF a = | ZF a -BZF a| ;CBC a = | BC a -BBC a| ; Obtaining a music level rating coefficient PJ according to the amplitude difference and wavelength difference between the acquired enhanced recording data and the standard music data; in, kp_1 is the adjustment coefficient of the amplitude difference, and kp_2 is the adjustment coefficient of the wavelength difference.

9. The method for constructing an online music teaching service platform according to claim 8, characterized in that: The process of obtaining the student's music level grade according to the music level rating coefficient includes: The music level rating coefficient threshold range is set according to the actual situation, and the set music level rating coefficient threshold range is recorded as (PJ0, PJ1); Obtaining the student's music level grade according to a comparison result between the obtained music level rating coefficient and a set music level rating coefficient threshold range; The music level grades include first-level music level, second-level music level and third-level music level; When PJ≤PJ0, the student’s music level is at the third-level music level; When PJ0<PJ<PJ1, the student’s music level is level 2; When PJ≥PJ1, the student's music level is the first-level music level.

10. The method for constructing an online music teaching service platform according to claim 9, characterized in that: The process of pushing corresponding online music courses to students according to student behavior data and student music level includes: Obtain the music types corresponding to the music that the student listens to most times and the music that the student listens to most frequently, record the music type corresponding to the music that the student listens to most times as the first music type, and record the music type corresponding to the music that the student listens to most frequently as the second music type; The online music courses are divided into difficulty levels and types, and the difficulty levels of the online music courses include level one difficulty, level two difficulty and level three difficulty; When the student's music proficiency level is level three, online music courses of level three difficulty of the same type as the first music type and online music courses of level three difficulty of the same type as the second music type are pushed to the student; When the student's music proficiency level is level 2, online music courses of level 2 difficulty of the same type as the first music type and online music courses of level 2 difficulty of the same type as the second music type are pushed to the student; When the student's music proficiency level is level one, online music courses of level one difficulty of the same type as the first music type and online music courses of level one difficulty of the same type as the second music type are pushed to the student.

Images