Virtual reality interactive music teaching system based on scene teaching method

Abstract

The invention discloses a virtual reality interactive music teaching system based on a scene teaching method. The method comprises the following steps: dynamically adjusting the content pushing sequence and opportunity of a teaching scene based on real-time feedback data of a student; creating a virtual reality environment to simulate different music scenes and introducing interactive elements; a multi-level teaching resource library is provided to match the requirements of different learning stages; and a personalized guide prompt is triggered to enhance the understanding and participation sense of the trainee. Through the scheme, the problem of how to intelligently regulate and control the content pushing sequence and time of the teaching scene so as to solve the problem that the course progress is not matched with the understanding degree of students can be solved.

Description

Technical Field

[0001] The present application relates to the field of virtual reality technology, and in particular to a virtual reality interactive music teaching system based on a situational teaching method. Background Art

[0002] A virtual reality interactive music teaching system based on situational teaching method is an innovative teaching method. It combines virtual reality (VR) technology with situational teaching concepts to provide students with an immersive music learning experience by constructing realistic music performance scenes and interactive links. This system not only allows students to understand music theory knowledge and performance skills more intuitively, but also improves their interest and participation in music. However, one challenge faced in the implementation process is how to intelligently control the order and timing of content push in teaching scenarios. Since each student's understanding ability and learning progress may vary, if the presentation order and time of teaching content cannot be reasonably arranged, the course progress may not match the student's actual mastery, thereby affecting the learning effect. This requires effective monitoring and intelligent algorithms to optimize the content push strategy to better meet the learning needs of different students. Summary of the invention

[0003] In view of this, the embodiments of the present disclosure provide a virtual reality interactive music teaching system based on a situational teaching method, which at least partially solves the problems existing in the prior art.

[0004] A virtual reality interactive music teaching system based on situational teaching method, comprising:

[0005] Dynamically adjust the order and timing of content delivery in teaching scenarios based on real-time feedback from students;

[0006] Creating VR environments to simulate different music scenarios and introduce interactive elements;

[0007] Provide a multi-level teaching resource library to match the needs of different learning stages;

[0008] Trigger personalized guidance prompts to enhance learner understanding and engagement.

[0009] In a specific embodiment,

[0010] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0011] Monitor and collect students' real-time responses during the learning process;

[0012] Analyze the level of understanding and emotional state reflected in real-time reactions;

[0013] Use machine learning algorithms to evaluate the progress and difficulties of students in order to optimize the pushed content;

[0014] If the student's understanding level is L and the course difficulty level is D, when L < T and the emotional stability S ≥ M, lower the course difficulty level by one level to D - 1, otherwise keep the course at the current difficulty level. Here, L represents the understanding level, T represents the preset understanding threshold, S represents the student's emotional stability score, M represents the critical value of emotional stability, and D represents the current course difficulty level.

[0015] In a specific embodiment,

[0016] The steps of dynamically adjusting the content push sequence and timing of the teaching scenario based on the real-time feedback data of students further include:

[0017] Collect the heart rate and facial expression data of students as a reference for psychological reactions;

[0018] Analyze the heart rate variability HRV and facial expressions E, and calculate the instant engagement I through a model;

[0019] Estimate the best interaction element Ei for the next display before the teaching scenario progresses;

[0020] If the influence value of the current interaction element is Ci and the predicted influence value of the next interaction element is Cn + 1, when Ci < Cn + 1+ ε and I > β, add this element to the scenario; where HRV refers to the heart rate variability index, E represents facial expression features, I is the instant engagement evaluation score, Ci and Cn + 1 respectively refer to the existing and predicted influence values, ε is the allowable error range, and β represents the minimum demarcation line for effectively improving engagement.

[0021] In a specific embodiment,

[0022] The steps of dynamically adjusting the content push sequence and timing of the teaching scenario based on the real-time feedback data of students further include:

[0023] Generate a personalized progress chart by combining long-term tracking records;

[0024] Apply reinforcement learning to update the reward and punishment functions F(R) for different types;

[0025] Conduct a summative evaluation A on the results of each learning cycle;

[0026] If the evaluation result shows that the correct answer rate is P and the number of correct questions is Q, if (P / N >= g and P - Q > z), increase the teaching speed v to v + Δv, where P represents the proportion of the number of questions answered correctly by the student, N is the total number of questions answered, z is the difficulty difference metric between questions, g is the passing rate standard, V is the original teaching rhythm, and Δv is the adjustment coefficient.

[0027] In a specific embodiment,

[0028] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0029] Track the memory effect after repeated appearance of specific knowledge points R record each student's memory decay curve;

[0030] Use deep learning technology to simulate and predict the long-term effects Y of the above records;

[0031] Develop a review strategy so that the same or similar concepts are re-presented at predetermined intervals;

[0032] If the historical error frequency Hf corresponds to an important knowledge point K, and when the prediction accuracy Py is less than h, a special intensive training is arranged. Hf represents the historical error rate of a specific knowledge item, r_i determines the review interval, K refers to an important subject point, Py is the prediction accuracy of the subject, and h sets the minimum acceptable accuracy threshold.

[0033] In a specific embodiment,

[0034] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0035] The attention allocation pattern model AM is introduced to measure the attention distribution of each student;

[0036] Calculate the attention focus migration path and its frequency PF;

[0037] Determine the key attraction factor Fk according to statistical laws;

[0038] If the number of times of attention focus is Fk and the continuous distraction exceeds the time limit τ, once (PF / Td)>δ, the content complexity C of this part is reduced to an appropriate level C-1, where Fk is the attraction factor of the key point, PF is the total frequency of attention flow, Td is the total viewing time, δ sets the transfer efficiency threshold, and C is the comprehensiveness of the material in the unit.

[0039] In a specific embodiment,

[0040] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0041] The eye tracking device is used to capture the position POA and trajectory of the visual attention point;

[0042] Estimate the effective recognition probability Pr for each gaze stop;

[0043] Focus on those objects with high Pr value areas to provide targeted assistance Ohs;

[0044] If the point of interest of an object is PA, the stay time is ts, and Pr>pmin when ts / TA≥λ and PA belongs to the interest area IA, increase the teaching interaction intensity SI of this position. Here PA locates the focus seat Pr to evaluate the recognition possibility, IA is the attraction area, TA represents the cumulative visual time, and λ judges the significant stay limit value.

[0045] In a specific embodiment,

[0046] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0047] Implement a multi-round question-answering system based on natural language processing for real-time communication;

[0048] Understand the semantic context and analyze the structure and content of students' answers;

[0049] Extract possible comprehension barriers and mark them out as OBs;

[0050] If the contextual coherence score CC lags behind the ideal value CL, CCΔCC≤CL, the lecturer is prompted to intervene to explain or change to a more concise example to supplement, CC speech consistency index, ΔCC acceptable fluctuation range, OBs difficult mark.

[0051] In a specific embodiment,

[0052] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0053] Enhance the sound feature extraction at the audio input, especially the pitch TP change pattern;

[0054] Observe the emotional fluctuations ET that accompany the tone shift;

[0055] Form an emotional map ME that matches the current scene;

[0056] If the emotion score SF reaches the threshold SFT, the current situation CT is switched to a state NT that better matches the emotion expression; among them, TP characteristic pitch information, ET auditory emotion trend, CT current educational link, NT is more suitable for the new situation of emotion reflection, SF thought score, ME psychological map, and SFT drastic change boundary.

[0057] In a specific embodiment,

[0058] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0059] Optimize the image presentation mode VI to match the best visual experience for different device environments;

[0060] Test the image rendering quality Q under various resolution settings Rg;

[0061] Select the screen configuration PCfg that best suits the hardware carrying capacity and does not affect the teaching quality;

[0062] If the display resolution is set to Rs and the image clarity evaluation score is Qd, then if ((Q-ε) / Rs)>=η, this resolution is locked as the recommended configuration Rc; RS display parameters, Q clarity score results, RCfg recommendation plan.

[0063] In a specific embodiment,

[0064] The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes:

[0065] Design learning paths that are consistent with cognitive psychology theory to suit students of different levels;

[0066] Build a recurrent neural network architecture to handle time series representation learning tasks (STTs);

[0067] Dynamically fine-tune the transition processes TMfs between modules based on student feedback;

[0068] If the students' mastery status WS in a certain learning period t reaches the expected benchmark EB, they are allowed to continue to the next sequence SS, that is, (if (ΣWS[t]) ≥ EB), and start the next task. In the t time segment, ΣWS students' cumulative completion status, SS next stage learning arrangement.

[0069] The disclosed embodiment provides a virtual reality interactive music teaching system based on the scenario teaching method, including: S1, dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data of the students; S2, creating a virtual reality environment to simulate different music scenes and introduce interactive elements; S3, providing a multi-level teaching resource library to match the needs of different learning stages; S4, triggering personalized guidance prompts to enhance students' understanding and sense of participation. Through the solution of the disclosed embodiment, it is possible to solve the problem of how to intelligently adjust the content push order and timing of the teaching scenario to solve the problem of mismatch between course progress and students' understanding level. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a connection diagram of the system modules of the present invention. DETAILED DESCRIPTION

[0071] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0072] Next, a virtual reality interactive music teaching system based on the scenario teaching method of the present invention is described. This system enhances the learning effect by integrating the real-time feedback data of the students into the teaching process and dynamically adjusting the content push order and timing of the teaching scenario. At the same time, a variety of scenes are created in virtual reality to stimulate students' interest in music, provide a teaching resource library matching different learning stages, and trigger personalized prompts to ensure that each student obtains maximum participation based on understanding.

[0073] Dynamically adjust the content push order and timing of teaching scenarios based on real-time feedback data from students. This means that teachers can optimize teaching strategies in a timely and appropriate manner according to the specific circumstances of students. In one embodiment, suppose a student has a weak perception of rhythm and encounters difficulties in the jazz piano learning module. At this time, the biometric sensor worn by the student will detect changes in their heartbeat or eye movement frequency, and these data are analyzed to confirm the student's troubles. Then, through the algorithm to calculate the optimal adjustment, the system automatically pushes a simpler and more targeted exercise - that is, rhythm introduction, to help students better transition to complex movement performance. This not only improves the effectiveness of the course arrangement, but also provides personalized path selection and support for each participant, solving the problem of differences between course progress and students' personal mastery. For example, when a child is stuck at a certain level for too long, he can receive a prompt message immediately: It seems a bit difficult for you now? Let's try the easier part first!

[0074] Establish a complete virtual reality (VR) ecosystem to simulate different music occasions and import operable objects. Specifically, such an immersive platform allows students to experience concerts of various styles, such as inside a symphony hall or at an open-air music festival. In a real example, beginners often find it difficult to establish the complex and coordinated relationship between the parts of musical instruments due to lack of intuitive understanding; therefore, reproducing the actual performance venue environment through VR equipment and placing corresponding virtual objects such as piano overtones and band conductor images can allow students to understand the concept of multi-part collaboration more quickly and vividly. At the same time, the design of interactive functions allows students to manipulate certain key elements by themselves, such as changing the melody rate to intuitively feel the difference between allegro and adagio, or giving children the opportunity to stand in front of the stage as a conductor to lead the entire virtual orchestra to perform practice.

[0075] In order to meet the needs of new users of different ages and backgrounds, the teaching materials mentioned here include knowledge explanation text files from basic knowledge to high-end skills improvement, as well as video demonstration clips and audio sample packages; in addition, they also include game task-based training sessions and regular live classroom special activities. For example, if a young adult wants to obtain a professional certificate through self-study but has no classical music foundation, he will first be guided to contact the entry-level textbook "Easy to Read Simple Music Notation ABC" suitable for adults under the support of this structured content. As his skills gradually accumulate, he can also participate in online seminars to listen to industry experts analyze classic composition techniques and share their creative experiences, etc.

[0076] Appropriate guidelines and incentives are initiated based on the user's current learning situation to strengthen the understanding and memory process. Specifically, the system can evaluate the student's current level based on the data information collected in the previous steps and give positive encouragement in time, or play a demonstration operation video recorded in advance by a real teacher to help clarify misunderstandings and strengthen understanding, and even allow players to unlock some special gadgets to assist their own progress and growth. In a specific scenario, when a student successfully completes the first bass guitar chord change challenge, a line of words may pop up on the screen saying: Wow, great job! Then it immediately switches to a video explanation on how to further improve the fingering details, and at the same time issues bonus points to players to unlock cooler background skins or other advanced features, which greatly promotes learning enthusiasm.

[0077] In summary, a virtual reality interactive music teaching system based on situational teaching method forms a closed loop chain through the above four key links. The most critical point in this process is to use technical means to finely regulate the logical chain of teaching situation content presentation, and fully consider the characteristics of demand changes brought about by individual differences while ensuring the overall coherence of the course. By analyzing a large amount of real-time physiological and psychological state data, accurately predicting possible problem nodes and quickly making responsive adjustments, it not only ensures that the collective teaching goals can be achieved smoothly, but also respects and recognizes the development potential of each learner; especially when facing target groups of different types and levels, it shows particularly obvious superiority and efficiency.

[0078] Next, the present invention is described.

[0079] The virtual reality interactive music teaching system based on the scenario teaching method dynamically adjusts the content push order and timing of the teaching scenario through the real-time feedback data of the students. Specifically:

[0080] First, monitor and collect the real-time reactions of students during the learning process. This step involves using a variety of sensors and user input mechanisms, such as eye trackers, gesture recognition cameras, and speech analysis devices, to capture data on students' attention, engagement, and immediate behavior patterns during learning. The collected data will be recorded for subsequent in-depth analysis. This step ensures the collection of comprehensive, timely, and diverse student reaction information.

[0081] Next, conduct an in-depth analysis of the understanding level and emotional state reflected in the above-mentioned real-time reactions. The analysis module will evaluate the collected student actions, speech characteristics, and facial expression changes to deduce the understanding of specific course materials (understanding level) and the emotional stability and pleasure of the students (emotional state). This module uses a pre-trained neural network model for rapid and efficient evaluation, enabling the system to make flexible adaptive responses based on individual performance.

[0082] Subsequently, use machine learning algorithms to evaluate the learning progress and difficulties of students, and accordingly optimize the subsequent teaching content push strategy. This process involves the application of the previous monitoring results - intelligent comparison and prediction in combination with the accumulated large-scale learning dataset, aiming to determine which knowledge points are more difficult for a certain student to understand. Then, by implementing personalized processing on these content modules with different levels of difficulty, new topics or supplementary exercises tailored to students with different progress are provided while maintaining coherence. This precise push effectively reduces the cognitive pressure caused by unnecessary repetition or overly fast skipping.

[0083] If it is determined through the foregoing steps that the current understanding level L of the student does not match the course difficulty level D set for the current stage, then according to the established logic, it is judged whether the course difficulty needs to be adjusted, that is, when the understanding level is lower than a certain set threshold T and the emotional stability S is not lower than the minimum critical value M (i.e., L < T and S ≥ M), then the system will lower the course difficulty by one level to D - 1. The range of the parameter L can be from 0 to 100, and the optimal setting of T depends on the nature of the subject and the general cognitive ability of the target age group; similarly, the value of the emotional stability S is usually also limited to the same range, and the critical line M is generally set to 45 - 65. A more practical number that does not affect the teaching quality can be selected according to experimental statistics. The reason for formulating such a combination of conditions is to ensure that even in the face of relatively complex information, students will not have an excessive sense of burden or frustration, but at the same time, ensure sufficient challenge to stimulate learning interest.

[0084] For example, in a specific embodiment, a student named Xiao Ming who is new to electronic piano playing skills has just started to practice fingering. In the early stage, due to his unfamiliarity with finger position and strength control, monitoring found that his understanding score was about 30, and his emotional fluctuations were significantly higher, reaching 75, which obviously did not meet the adjustment standards; however, after several classes, he gradually mastered the basic operation points, and his corresponding indicators improved to L=50 and S=58. He is still in a relatively inefficient learning stage, but his emotional state is relatively stable. Therefore, according to the set rules, the system will slightly reduce the difficulty of this part of the course on scale conversion to D=3, which is suitable for beginners, to help consolidate existing skills so as to better meet the next part of advanced learning. In this way, the entire teaching process is continuously optimized until the ideal learning results are achieved.

[0085] Next, the present invention is described. In a virtual reality interactive music teaching system based on a situational teaching method, first, the heart rate and expression data of the students are collected in real time as a reference for psychological reactions. This step includes monitoring the physiological and emotional changes of the students during the interactive process, and using these data to understand the emotional state and concentration level of the participants.

[0086] Subsequently, the acquired information is processed by a professional algorithm, and the heart rate variability HRV and facial expression features E are analyzed, and then an immediate engagement level I that can accurately express the level of engagement of the participants at the current moment is calculated. For example, suppose in a virtual scene, a student is following a teacher to learn to play a musical instrument. At this time, the system detects that the student's heart rate is relatively stable and his face is focused, indicating that he is in a good state and actively participating in learning activities. At this time, the I value should be high. Here, HRV measures the difference between heartbeat intervals, ranging from less than 30 milliseconds to more than 150 milliseconds, and lower values ​​are usually associated with a more tense state; for E, different emotions, such as joy or worry, can be identified through facial muscle movements; and the I score ranges from 0 to 10 points, and a score close to 8 in this range is considered an ideal classroom response level.

[0087] The next step is to estimate the next interactive content that is most likely to trigger the best effect. Specifically, before each new teaching situation appears, determine whether the material to be used can cause the expected enthusiastic response. When the influence of the existing element Ci is judged to be less than the predicted impact of the next material plus a small range of fluctuations ε, and the immediate participation score just obtained is higher than a certain set threshold β, the new factor will be included in the display sequence. Ci and Cn+1 represent the size of the influence that may exist twice in a row. The former is the existing material used, and the latter refers to the items to be added in the future. The parameter ε is used to define the error space to adapt to uncertain factors. As for β, it is to ensure that only those things that can bring positive improvements are selected. It is generally defined as being greater than 7.

[0088] Operating under this framework can make the system more sensitive to the learning progress and changing trends of each student's inner experience, thereby providing customized and more attractive educational programs. In one embodiment, specifically, when a primary school student is tired while experiencing a guzheng performance in a simulated environment, it is determined based on his physiological characteristics that he needs to switch to a relaxing and lively serenade practice to boost his spirits. This approach ensures that each child can get the most appropriate guidance for his or her own situation, making the entire teaching process lively and orderly.

[0089] Next, the present invention is described. The virtual reality interactive music teaching system based on the scenario teaching method, in the process of dynamically adjusting the teaching content and sequence according to the real-time feedback data of the students, includes the following steps:

[0090] First, the system generates a personalized progress chart based on the learner's long-term tracking record. Long-term tracking refers to the collection of all learning-related parameters and activity information from the first use to the present. The significance of this step is to make the system clear about individual abilities, preferences, and progress. Personalized progress charts are not only a visual presentation tool for historical performance, but also provide an intuitive way to show improvement directions and development trajectories. For example, in one embodiment, after completing a section of instrument playing exercises with different difficulty levels, the system updates the chart to show the degree of skill improvement or areas that need to be consolidated.

[0091] Then, reinforcement learning technology is used to optimize the reward and punishment mechanism F(R), which is adjusted to adapt to individual differences and motivate continuous participation. Here, R represents the quantitative value of various positive or negative results; F is a process of mapping these values ​​into new reward signals. In music education, it is specifically manifested as giving immediate positive affirmation when the preset goal is achieved, such as accurately hitting the rhythm point or completing a complex score. Conversely, if a mistake is made, a certain amount of points will be deducted. This system can ensure that students' interest is continuously stimulated and positive learning motivation is formed. The optimal F function should be able to effectively distinguish different levels of achievement and encourage students to actively pursue higher levels of success experience.

[0092] Furthermore, a summary evaluation A is conducted for each complete practice cycle. In this process, all training tasks and their results involved in this round are comprehensively reviewed, and the final score is converted into a measurable evaluation indicator. P is the ratio of the total number of correctly answered questions to the total number of questions N, which reflects the current level and stability; Q refers to the specific number of questions actually answered correctly under a specific knowledge point, which can better reveal whether a specific field has been mastered. In one example, after completing a set of music theory exercises, P = 0.9 (90% of the questions were answered correctly), and Q = 18 means that 18 questions in this test paper were completed correctly.

[0093] Then, based on the result of evaluation A, decide whether to speed up or maintain the current teaching pace v. If the current learning status reaches a good level or above (i.e., P / N ≥ the given qualification standard g and PQ > the set difficulty increment z), the teaching rate will be moderately increased by Δv. P / N is a benchmark value for measuring the overall completion, which should be greater than or equal to the minimum threshold g to ensure the solidity of basic knowledge; and (PQ) is an additional consideration factor to detect the challenge gap between consecutive answers to questions of the same nature. Its ideal value should be able to reflect a stable growth trend rather than a large fluctuation. The basis for choosing z is to maintain a reasonable balance between challenge and sense of achievement. Once the above formula logical conditions are met, the teaching strategy is adjusted from the original v to a faster speed v+Δv, which enables students who show strong adaptability to acquire more diversified knowledge and skills training at a faster speed.

[0094] In short, it is crucial to dynamically adjust the configuration of elements and rhythm in the teaching scenario in this process, aiming to achieve a win-win situation of efficient and accurate knowledge transfer and personalized development of students. By continuously iterating and improving the relevant parameter settings and optimizing the algorithm selection, we ensure that each adjustment is scientific and reasonable and conducive to long-term development.

[0095] Next, the present invention is described. First, in the step of tracking the memory effect after the repetition of a specific knowledge point, the system records each student's understanding and memory changes of the important knowledge points during the learning process, forming a memory effect R. For example, the system can monitor the student's memory performance of a certain musical symbol or theory, thereby establishing a unique memory decay curve for each student. In this example, various performance situations are simulated in virtual reality, and the success rate of the student's recognition and application of the musical symbol is observed and recorded, and its change over time is plotted as a line.

[0096] Furthermore, based on the above data, deep learning technology is used to make a simulated prediction Y of the long-term effect. Here, the input parameters are the aforementioned memory curve and error history and other personal learning dynamic information of the students. After sufficient training, the prediction model can output the result Y about how much mastery the students can retain for a certain knowledge in the future. The Y value is usually a set of probability values ​​between [0,1]. Ideally, the estimated value should be as close to the real value as possible to achieve more accurate teaching planning. The optimal setting is the best performance point where the model reaches the minimum error of the validation set.

[0097] Afterwards, based on the obtained long-term effect simulation prediction Y and memory characteristics, a personalized review strategy is formulated for the same or similar concepts. The review time interval r_i will be optimized and adjusted according to the different responses of different students, so that important knowledge points can be effectively reviewed in the time window that is most easily forgotten, thereby strengthening the stability of long-term memory. Assuming that this system is applied to music teaching in an embodiment, if it is found that the pronunciation of a certain note is often forgotten, a corresponding VR teaching environment is created for this knowledge point according to the time period indicated by r_i, such as 3 days, 7 days, and 15 days, so that it can be reproduced.

[0098] Finally, the system checks whether the historical error frequency Hf corresponds to a particular key content, namely the key knowledge point K, and evaluates the prediction accuracy Py (also expressed in the form of a ratio between [0,1], with a higher limit value being preferred under high precision). If the predicted value fails to exceed the set threshold h (generally set by reference to industry or institutional standards, or may be determined based on actual experience), it means that the current review plan is insufficient to support the stable acquisition of knowledge. Specifically, in a special training, if it is observed that the error rate of students in determining the position of the C-tone on the stave remains high and the expected improvement is poor, the special education intervention mechanism is triggered to arrange intensive guidance courses to ensure that the student receives sufficient support in such basic skills until the key point can be mastered smoothly.

[0099] Next, the present invention is described. This system is a virtual reality interactive technology based on a situational teaching method for music teaching. In this system, by analyzing the real-time feedback data of students in a specific learning session, the push content and its order and timing are dynamically adjusted to optimize the teaching interactivity and effect. The specific implementation process can be divided into the following main steps for introduction.

[0100] First, the attention distribution of the trainees is measured. In this process, the attention allocation model AM is introduced. This is a specially designed model to evaluate the attention distribution of different trainees throughout the immersive experience. AM determines in which situation each student is more easily distracted or puts the most effort based on the eye movement, head movement and other body posture signals obtained by the sensor.

[0101] Secondly, the PF and frequency of the attention focus transfer path are calculated. This operation is intended to record the number and time sequence of eye gaze point changes from one activity to another. The specific method is to continuously monitor the changes in the students' viewpoint from the beginning of learning to the completion of the learning, so as to obtain a complete transition chart.

[0102] Then, the key attraction factor Fk is determined according to statistical laws. The so-called key attraction factors refer to those teaching elements that can best capture the attention of participants and have a significant impact on them. For example, in a music course, it may be a musical instrument performance or a particularly vivid and interesting sound effect display. The system will screen the locations that appear multiple times and stay for a long time, identify these high-heat areas, and use them as a reference to improve subsequent teaching arrangements.

[0103] When it is detected that some parts of attention are interrupted too frequently by unexpected things, and the cumulative distraction exceeds the predefined time limit τ (τ can be set to a few minutes to more than ten minutes, depending on the actual situation), if the average migration rate (PF / Td) is found to be greater than the predetermined value δ, it indicates that the current part of the content is difficult to attract sustained attention and the complexity C needs to be reduced. Among them, the total viewing time Td refers to a total period accumulated from the start of the course to the end of the entire period; δ is used as a transfer efficiency threshold to indicate the scale under which it is considered an excessive jump, and the experience value setting range is generally in the range of 0.3 to 1; the original value of C represents the knowledge capacity and the richness of expression covered in the current module. Once the above conditions are exceeded, it will be lowered to C-1, making the material easier to accept and understand. Take the instrument fingering demonstration unit as an example. Specifically, in one embodiment, assuming there is a set of difficult classical piano exercises, if the above analysis finds that many viewers are distracted and unable to keep up with the progress of the textbook due to the fingers sliding too quickly in the picture, and this abnormal interruption is very serious, when it reaches a certain level, the detailed instructions here will be simplified, such as reducing the rhythm of the finger displacement speed changes and refining the language instructions, so that the audience can learn new pieces more comfortably and smoothly.

[0104] Next, the present invention is described. The system dynamically adjusts the content push order and timing of the teaching scenario according to the real-time feedback data of the students, including capturing the students' sight to obtain the focus point and path data; estimating the probability according to the students' visual dwell time, focusing on the highly concerned objects, and dynamically adjusting the interaction intensity with the highly concerned areas. The following are the specific steps and their detailed description.

[0105] First, an eye tracking device is used to capture the trainee’s point of attention (POA) and its movement trajectory. This means that the device uses sensors to accurately track and record the specific content and viewing order of the trainee in the VR environment.

[0106] Secondly, estimate the effective recognition probability (Pr) for each stop in the line of sight. Pr is a probability evaluation parameter for whether the student can correctly understand the music information in a specific area after fixating on it. Its value range is between 0 and 1 (including 0 and 1), and the optimal value depends on the actual application scenario. This step aims to deduce the student's cognition of music concepts through his visual behavior, so as to provide a reference for the subsequent formulation of teaching strategies.

[0107] Then, targeted auxiliary elements Ohs are added around objects with higher Pr values. For example, in one embodiment, if a student focuses on a certain musical symbol for a long time and fails to quickly identify its function, the system will pop up a note in time to help the student understand and learn this knowledge point, avoiding progress blockage due to cognitive gaps.

[0108] If the dwell time of an object PA is ts, and when Pr>pmin (preset probability lower limit), the ratio of ts / TA (single dwell time divided by cumulative fixation time period) is equal to or exceeds the given ratio λ, and at the same time, PA is located in the interest area IA. Increase the teaching interaction intensity SI at this location. Specifically in this case, if it is detected that the student shows continuous interest and attention to a virtual key - whether it is difficult to read or curious - then the system will increase the form of interaction here: it may add animation, explanation, or invite simulated trial play and other enhanced interactions, so that the student can better immerse himself in the process of music exploration.

[0109] This formula design ensures that each student is accurately responded to, taking into account the absolute length of visual focus and the weight of interest stimulation. The λ proportional coefficient is used to balance the two considerations, and the optimal value range is set to [0.3, 0.7] to ensure that the system will neither overly sensitively trigger all gazes nor lag behind and miss key teaching opportunities. In addition, a reasonable and effective pmin is set to filter out occasional unintentional gazes.

[0110] In an application example, in an immersive virtual piano class, the students' learning efficiency of staff notation was monitored through the above steps and found to be significantly improved, because appropriate support and incentives were provided for students with different foundations, making the experience of each learner smoother, more efficient and interesting.

[0111] Next, the present invention is described.

[0112] The content push order and timing steps of dynamically adjusting the teaching scenario based on the real-time feedback data of the students are divided into multiple steps: the first is to conduct real-time communication between teachers and students through a multi-round question-and-answer system that implements natural language processing technology. In this process, the introduction of the multi-round question-and-answer system allows music courses to be continuously interactive rather than a one-way transmission of knowledge. For example, in a virtual piano teaching embodiment, this system allows students to understand music theory concepts and guide learners to think and express by answering continuous questions such as "Please tell me what this symbol represents".

[0113] Then, we analyze the semantic context, the structure and content of the students' answers (ACt: analysis of the answer content), in order to understand the students' understanding of the current music knowledge points, including the use of terms, whether the logical reasoning is accurate and coherent, etc., and identify possible cognitive gaps and make annotations (OBs: difficult to understand knowledge points or doubts). Taking a specific explanation as an example, when it comes to classical music appreciation, if the answer lacks a sense of rhythm or melody structure, it is considered that there is an unclear concept and is marked.

[0114] If the contextual coherence score (CC) is found to be lower than the expected ideal score (CL) in the above process, and this difference reaches or exceeds the maximum acceptable fluctuation (ΔCC≤0 to a certain positive integer range, set according to the educational objectives and evaluation system; the optimal configuration is the optimal deviation threshold determined when the confidence interval is statistically above 95%), it implies that the students may have a deep cognitive dilemma that needs to be solved, thus prompting the instructor to participate in further elaborating on the difficulties or turn to more direct and specific examples to assist in clarification. This link ensures that the personalized learning effect is maximized while ensuring that chaos will not occur due to excessive deviation from the original course framework.

[0115] Next, the present invention is described. This step is intended to dynamically adjust the order and timing of pushing teaching scenario content based on real-time feedback data from students to ensure the most effective learning experience. First, the process extracts sound features by enhancing the audio input end, especially focusing on identifying pitch change patterns. This means that the specific changes in the sounds made by students are accurately captured through algorithms or machine learning techniques. For example, subtle changes in pitch, speaking speed, etc., because these are one of the important clues to reflect individual emotions and understanding. Among the parameters, pitch TP represents characteristic pitch information. This value varies according to the specific value of the pitch and is not limited to a fixed frequency band. Its optimal value range should cover the vocal range that most people can comfortably reach.

[0116] Secondly, observe the emotional fluctuations ET that accompany the change in tone. The focus here is on whether the rhythm between the students' speech contains factors such as positivity, anxiety, or relaxation, in order to measure the tendency of emotional trends. The emotional score SF is set as a quantitative standard to measure the degree of the student's current emotional state; for example, 0 means complete calmness, while 100 corresponds to an extremely strong emotional outburst, and the intermediate values ​​represent various forms of feelings at different intensities. The drastic change boundary SFT is a threshold indicator for determining the intensity of emotional fluctuations. If SF reaches and exceeds SFT, it implies that it is necessary to make timely changes to the current teaching scenario. In one embodiment, SFT can be set to around 60-70 points to ensure that significant emotional transitions can be accurately captured in most cases.

[0117] Furthermore, based on the information obtained from the previous analysis, an emotion map ME that matches the current scene is formed. This step requires the integration of the sound characteristics collected in the early stage and the emotional information behind them, so as to generate a psychological map that can describe the characteristics of the user's psychological activities. This map not only records the immediate emotions, but also includes the evolution trajectory and development trend over a period of time, so that the education link can provide personalized adjustment plans for the different characteristics of individuals. ME parameters cover many aspects such as historical emotional curves and reactions to specific events, but there is no concept of absolute optimal value because it is highly dependent on the unique data set accumulated by each individual user's long-term behavioral habits.

[0118] Finally, when the calculated emotion score SF does reach the preset drastic change boundary SFT, the system will transfer from the current teaching link CT to a new environment NT that is more suitable for the current emotional expression to carry out subsequent work. For example, in one embodiment, when a student is practicing a difficult piece of music and begins to make frequent mistakes or show great frustration (i.e., a high emotion score SF is detected), the teaching program will analyze the psychological map and believe that switching to a more supportive and encouraging learning scenario will be more conducive to relieving stress and rebuilding self-confidence, so the original rigorous technical training course will be quickly transformed into an interactive communication course focusing on appreciating music works, so as to re-stimulate interest in the new atmosphere.

[0119] The above process reflects how the virtual reality interactive music teaching system uses real-time monitoring mechanisms and intelligent decision-making support systems to achieve flexible and efficient adjustments to teaching effects, so that each educational experience can closely fit the actual situation and development needs of the trainees. Specifically, by highly grasping details such as pitch TP and auditory emotional trends ET, a scientific emotion scoring SF system is constructed, and the best scenario change strategy is determined based on this, ultimately achieving the goal of optimizing the quality of music skill training.

[0120] Next, the specific steps of dynamically adjusting the content push order and timing of the teaching scenario according to the real-time feedback data of the students are described as follows:

[0121] In order to achieve the best visual experience in the virtual reality environment, the image presentation method of different device environments is optimized, such as using different rendering strategies for high-resolution displays and low-configuration mobile devices. This process ensures that music teaching scenarios on different devices can achieve high-quality visual effects while taking into account stable operating performance.

[0122] The rendering quality of the picture at different resolutions is tested and evaluated. Specifically, the resolution parameter Rg is changed to observe the clarity of the picture, and a clarity score Q is quantified for comparison. This step aims to identify which settings are most conducive to visual clarity while allowing smooth operation without stuttering or frame drops. In this process, the resolution Rg range usually covers the mainstream specifications on the market from 720P to 4K and even higher levels. For music classrooms, the best may be a point within the setting range that allows clear music score markings and fine and distortion-free instrument demonstrations.

[0123] According to the conclusions of the first two steps, the solution that best suits the hardware carrying capacity and can guarantee the teaching quality is selected, which is the PCfg selection stage. If the actual resolution score measured when the original resolution of the display is Rs is Qd, then the formula ((Q-ε) / Rs)≥η will be used to decide whether to accept this parameter combination as the quasi-recommended configuration RC.

[0124] Among them, Rs is the physical parameter of the specific display device, which can be the standard value provided by the screen manufacturer; Qd is the comprehensive picture quality score obtained through a series of algorithms to measure whether the image meets the standard; the symbol η is the preset threshold used as one of the judgment criteria, and ε belongs to the allowable error range. The setting of the above formula is intended to exclude situations that are not suitable for the carrying capacity of the equipment to avoid problems such as excessive demand exceeding the processing limit or too low picture quality affecting the user's immersive experience, so as to ensure that the entire virtual reality interactive music classroom system runs well and the teaching effect is ideal.

[0125] In one embodiment, at the beginning of a music course, if it is detected that the student's environment is a high-end personal computer, a set of content presentation formats suitable for the current conditions is generated in accordance with the principle of higher definition but still smooth, and the students' participation and reactions are continuously monitored as the class progresses, so that the user is ready to switch to an alternative scenario combination mode at any time to ensure optimal teaching efficiency.

[0126] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the embodiments of the present disclosure. It should be understood that the above description is only the specific implementation method of the embodiments of the present disclosure and is not intended to limit the protection scope of the embodiments of the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the protection scope of the embodiments of the present disclosure.

Claims

1. A virtual reality interactive music teaching system based on situational teaching method, characterized in that: The system includes: A real-time feedback module for dynamically adjusting the content pushing sequence and timing of the teaching scenario based on the real-time feedback data of the trainee; A virtual reality environment module for creating a virtual reality environment to simulate different music scenes and introduce interactive elements; A teaching resource library module for providing multi-level teaching resources to match the needs of different learning stages; A guiding prompt module for triggering personalized guiding prompts to enhance the trainee's understanding and sense of participation.

2. The virtual reality interactive music teaching system based on the scenario teaching method according to claim 1, wherein The steps of dynamically adjusting the content pushing sequence and timing of the teaching scenario based on the real-time feedback data of the trainee further include: Monitoring and collecting the real-time reactions of the trainee during the learning process; Analyzing the understanding level and emotional state reflected in the real-time reactions; Using machine learning algorithms to evaluate the progress and difficulties of the trainee to optimize the pushed content; If the understanding level of the trainee is L and the course difficulty level is D, when L < T and the emotional stability S ≥ M, lower the course difficulty by one level to D_1, otherwise keep the course at the current difficulty level. Where L represents the understanding level, T represents the preset understanding threshold, S represents the emotional stability score of the trainee, M represents the critical value of emotional stability, and D represents the current course difficulty level.

3. The virtual reality interactive music teaching system based on the scenario teaching method according to claim 2, wherein The steps of dynamically adjusting the content pushing sequence and timing of the teaching scenario based on the real-time feedback data of the trainee further include: Collecting the heart rate and facial expression data of the trainee as a reference for psychological reactions; Analyzing the heart rate variability HRV and facial expression E, and calculating the instant engagement I through a model; Estimating the best interactive element Ei to be displayed next before the teaching scenario progresses; If the influence value of the current interactive element is Ci and the predicted influence value of the next interactive element is Cn+1, when Ci < Cn+1 + ε and I > β, add this element to the scene; where HRV refers to the heart rate variability index, E represents the facial expression feature, I is the instant engagement evaluation score, Ci and Cn+1 respectively refer to the existing and predicted influence values, ε is the allowable error range, and β represents the minimum dividing line for effectively improving the engagement.

4. The virtual reality interactive music teaching system based on the scenario teaching method according to claim 3, wherein The steps of dynamically adjusting the content pushing sequence and timing of the teaching scenario based on the real-time feedback data of the trainee further include: Generating a personalized progress chart by combining long-term tracking records; Applying reinforcement learning to update the reward and punishment functions F(R) for different types; Conducting a summative evaluation A on the results of each learning cycle; If the evaluation result shows that the correct answer rate is P and the number of correct questions is Q, if (P / N >= g and P_Q > z), increase the teaching speed v and set it to v + Δv, where P represents the proportion of the number of questions answered correctly by the trainee, N is the total number of questions answered, z is the difficulty difference measurement index between questions, g is the passing rate standard, V is the original teaching rhythm, and Δv is the adjustment coefficient.

5. A virtual reality interactive music teaching system based on situational teaching method according to claim 4, characterized in that: The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes: Track the memory effect of specific knowledge points after repeated appearances R Record the memory decay curve of each student; Use deep learning technology to simulate and predict the long-term effects Y of the above records; Develop a review strategy so that the same or similar concepts are re-presented at predetermined intervals; If the historical error frequency Hf corresponds to an important knowledge point K, and when the prediction accuracy Py is less than h, a special intensive training is arranged. Hf represents the historical error rate of a specific knowledge item, r_i determines the review interval, K refers to an important subject point, Py is the prediction accuracy of the subject, and h sets the minimum acceptable accuracy threshold.

6. A virtual reality interactive music teaching system based on situational teaching method according to claim 5, characterized in that: The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes: The attention allocation pattern model AM is introduced to measure the attention distribution of each student; Calculate the attention focus migration path and its frequency PF; Determine the key attraction factor Fk according to statistical laws; If the number of times attention is focused is Fk and it continues to be distracted for more than the time limit τ, once (PF / Td)>δ, the content complexity C of this part is reduced to an appropriate level C_1, where Fk is the attraction factor of the key point, PF is the total frequency of attention flow, Td is the overall viewing time, δ sets the transfer efficiency threshold, and C is the comprehensiveness of the material within the unit.

7. A virtual reality interactive music teaching system based on situational teaching method according to claim 2, characterized in that: The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes: The eye tracking device is used to capture the position POA and trajectory of the visual attention point; Estimate the effective recognition probability Pr for each gaze stop; Focus on those objects with high Pr value areas to provide targeted assistance Ohs; If the point of interest of an object is PA, the stay time is ts, and Pr>pmin when ts / TA≥λ and PA belongs to the interest area IA, increase the teaching interaction intensity SI of this position. Here PA locates the focus seat Pr to evaluate the recognition possibility, IA is the attraction area, TA represents the cumulative visual time, and λ judges the significant stay limit value.

8. A virtual reality interactive music teaching system based on situational teaching method according to claim 7, characterized in that: The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes: Implement a multi-round question-answering system based on natural language processing for real-time communication; Understand the semantic context and analyze the structure and content of students' answers; Extract possible comprehension barriers and mark them out as OBs; If the contextual coherence score CC lags behind the ideal value CL (CCΔCC≤CL), the lecturer is prompted to intervene to explain or change to a more concise example to supplement, CC speech consistency index, ΔCC acceptable fluctuation range, OBs difficult mark.

9. A virtual reality interactive music teaching system based on situational teaching method according to claim 8, characterized in that: The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes: Enhance the sound feature extraction at the audio input, especially the pitch TP change pattern; Observe the emotional fluctuations ET that accompany the tone shift; Form an emotion map ME that matches the current scene; If the emotion score SF reaches the threshold SFT, the current situation CT is switched to a state NT that better matches the emotion expression; among them, TP characteristic pitch information, ET auditory emotion trend, CT current educational link, NT is more suitable for the new situation of emotion reflection, SF thought score, ME psychological map, and SFT drastic change boundary.

10. A virtual reality interactive music teaching system based on situational teaching method according to claim 9, characterized in that: The step of dynamically adjusting the content push order and timing of the teaching scenario based on the real-time feedback data from the students further includes: Optimize the image presentation mode VI to match the best visual experience for different device environments; Test the image rendering quality Q under various resolution settings Rg; Select the screen configuration PCfg that best suits the hardware carrying capacity and does not affect the teaching quality; If the display resolution is set to Rs and the graphics clarity evaluation score is Qd, then if ((Q_ε) / Rs)>=η, this resolution is locked as the recommended configuration Rc; RS display parameters, Q clarity score results, RCfg recommendation plan.

Images