Music teaching training improving method and system

Abstract

The invention discloses a music teaching training improving method and system, and relates to the technical field of combination of artificial intelligence and music teaching, and the method comprises the steps: obtaining the sound data of a teacher and a listener in a music classroom; identifying the sound data of the teacher, extracting a sound fragment of singing demonstration, a sound fragment of musical instrument playing and a sound fragment of knowledge explanation, and evaluating the music basic skill of the teacher; recognizing the sound data of the listener, extracting sound segments of a singing, and sequentially calculating the similarity between each singing sound segment of the listener and the sound segment demonstrated by the listener in a singing manner according to a time sequence; evaluating the music teaching ability of the teacher based on the similarity; and based on the evaluation result of the music basic skill and the music teaching ability of the teacher, determining music teaching training promotion information for the teacher. According to the method, a targeted training plan and a promotion strategy can be formed, and students in the music education major are helped to systematically improve their teaching skills.

Description

Technical Field

[0001] The present application relates to the technical field of the combination of artificial intelligence and music teaching, and in particular to a method and system for improving music teaching training. Background Art

[0002] In the music education major or preschool education major in vocational and technical colleges, the core teaching goal is to cultivate students' ability to teach effectively in various teaching environments, including the ability to prepare lesson plans, implement classroom teaching, evaluate student performance, and adjust teaching strategies to meet the needs of different learners. In order to achieve these goals, students must master music theory knowledge, instrumental performance skills, vocal techniques, teaching methods, psychology and other related knowledge.

[0003] However, in the current education model, students' teaching methods and practical skills are often evaluated through subjective classroom observation or self-analysis, which are often affected by personal bias and subjective judgment and are difficult to provide a comprehensive and objective ability evaluation. In addition, due to the lack of real-time feedback and personalized guidance, it is difficult for students to quickly improve their shortcomings in teaching practice.

[0004] Traditional evaluation methods also face resource and time constraints. In many cases, teachers need to supervise a large number of students at the same time, making it difficult to pay individual attention to and evaluate each student. In addition, due to the lack of systematic evaluation tools, students may not receive immediate feedback on their specific weaknesses while receiving training, making the learning process blind and inefficient. Summary of the invention

[0005] In view of this, the present invention provides a music teaching training improvement method and system, which can form targeted training plans and improvement strategies to help students majoring in music education improve their teaching skills systematically.

[0006] To this end, the present invention provides the following technical solutions:

[0007] On the one hand, the present invention provides a method for improving music teaching and training, the method comprising:

[0008] Acquire the sound data of the lecturer and the listener in the music class;

[0009] Recognize the lecturer's voice data and extract the sound clips of singing demonstration, instrument playing and knowledge explanation;

[0010] Evaluate the instructor's basic music skills based on audio clips of song demonstrations, instrument playing, and knowledge explanations;

[0011] Recognize the voice data of the listener, extract the voice clips of the singing, and sort the singing voice clips of the same music section in the order of the singing time;

[0012] For each music segment, the similarity between each of the listener's singing sound segments and the instructor's singing demonstration sound segments is calculated in chronological order; the instructor's music teaching ability is evaluated based on the similarity;

[0013] Based on the evaluation results of the instructor's basic music skills and music teaching ability, a comprehensive evaluation result of the instructor's music teaching level is obtained;

[0014] Based on the comprehensive evaluation results, determine the music teaching training improvement information for the instructor.

[0015] Furthermore, the sound data of the lecturer and the audience in the music class are obtained, including: real-time sound data is collected through microphones or audio collection devices arranged at the seats of the lecturer and the audience in the music class.

[0016] Furthermore, based on the sound clips of singing demonstration, instrument playing and knowledge explanation, the instructor's basic music skills are evaluated, including:

[0017] An evaluation model based on machine learning is used to evaluate the instructor's basic music skills. The evaluation model based on machine learning includes three evaluation sub-models, which are used to score sound clips of singing demonstration, instrument playing demonstration and knowledge explanation respectively. Each sub-model will score the corresponding audio clip, and the three scores will be combined through weighted average to obtain a comprehensive basic music skills evaluation result; wherein each evaluation sub-model is trained using a sound clip dataset containing manual ratings.

[0018] Furthermore, the similarity between the sound clips of the listener and the lecturer singing the song is calculated, including:

[0019] Feature extraction: Extract key features from the audio of the listener and the lecturer respectively, where the key features include one or more of pitch, rhythm, timbre, Mel frequency cepstrum coefficient, and zero crossing rate;

[0020] Time Alignment: Use Dynamic Time Warping to adjust the timelines of two sound clips so that corresponding notes or beats match correctly;

[0021] Similarity calculation: Using the extracted features, cosine similarity is used to calculate the similarity between two feature vectors.

[0022] Furthermore, the music teaching ability of the instructor is evaluated based on the similarity, including:

[0023] Performing statistical analysis on the calculated multiple similarity values ​​according to the time sequence of the listeners singing the songs, and determining the changing trend of the similarity values;

[0024] According to the changing trend of the similarity value, the music teaching ability of the instructor is evaluated.

[0025] In another aspect, the present invention further provides a music teaching and training improvement system, comprising:

[0026] A sound data acquisition module is used to acquire the sound data of the lecturer and the listener in the music class;

[0027] The first evaluation module recognizes the voice data of the instructor, extracts the sound clips of the singing demonstration, the sound clips of the instrument playing, and the sound clips of the knowledge explanation; and evaluates the instructor's basic music skills based on the sound clips of the singing demonstration, the sound clips of the instrument playing, and the sound clips of the knowledge explanation;

[0028] The second evaluation module recognizes the voice data of the listener, extracts the sound clips of the singing, and sorts the sound clips of the same music segment in the order of singing time; for each music segment, the similarity between each of the listener's singing sound clips and the teacher's singing demonstration sound clips is calculated in chronological order; the teacher's music teaching ability is evaluated based on the similarity;

[0029] A comprehensive evaluation module, which obtains a comprehensive evaluation result of the instructor's music teaching level based on the evaluation result of the first evaluation module on the instructor's music basic skills and the evaluation result of the second evaluation module on the instructor's music teaching ability;

[0030] The personalized training improvement recommendation module determines the music teaching training improvement information for the instructor based on the comprehensive evaluation results obtained by the comprehensive evaluation module.

[0031] Furthermore, the sound data acquisition module is specifically used to collect sound data in real time through microphones or audio acquisition devices set at the seats of lecturers and listeners in the music classroom.

[0032] Furthermore, the first evaluation module is specifically used to:

[0033] An evaluation model based on machine learning is used to evaluate the instructor's basic music skills. The evaluation model based on machine learning includes three evaluation sub-models, which are used to score sound clips of singing demonstration, instrument playing demonstration and knowledge explanation respectively. Each sub-model will score the corresponding audio clip, and the three scores will be combined through weighted average to obtain a comprehensive basic music skills evaluation result; wherein each evaluation sub-model is trained using a sound clip dataset containing manual ratings.

[0034] Furthermore, the second evaluation module is specifically used to:

[0035] Performing statistical analysis on the calculated multiple similarity values ​​according to the time sequence of the listeners singing the songs, and determining the changing trend of the similarity values;

[0036] According to the changing trend of the similarity value, the music teaching ability of the instructor is evaluated.

[0037] Advantages and positive effects of the present invention: In the present invention, the voice data of the lecturer and the listener in the music class are obtained; the voice data of the lecturer is identified to evaluate the lecturer's basic music skills; the voice data of the listener is identified, and the music teaching ability of the lecturer is evaluated based on the similarity between the listener's singing and the lecturer's singing; based on the evaluation results of these two aspects, the music teaching training improvement information for the lecturer is determined. The present invention can provide real-time and objective evaluation data of the teaching ability of students majoring in music education, so that the teacher of the student can accurately identify the strengths and weaknesses of each student in music teaching. Through this continuous monitoring and evaluation, targeted training plans and improvement strategies can be formed to help students majoring in music education systematically improve their teaching skills. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 It is a flowchart of a music teaching training improvement method in an embodiment of the present invention;

[0040] Figure 2 It is a structural block diagram of a music teaching and training improvement system in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 creative work should fall within the scope of protection of the present invention.

[0042] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0043] like Figure 1 As shown, a music teaching training improvement method in an embodiment of the present invention specifically includes the following steps:

[0044] S1. Acquire the sound data of the lecturer and the listener in the music class;

[0045] Among them, the music class includes a lecturer and at least one listener. The lecturer is a student majoring in music education at a vocational and technical college, that is, someone who needs training and improvement in music teaching ability. The listeners can be people who learn to play and sing music, such as inviting kindergarten children, or other interested classmates. In the music class, the lecturer will demonstrate singing, playing, singing while playing, explain knowledge, and interact with the listeners, etc. The listeners will sing along, learn to sing, answer questions, etc.

[0046] Among them, the collection of sound data can be achieved through a microphone or an audio collection device. In the specific implementation, a microphone or an audio collection device is set at the seats of the lecturer and the audience in the classroom to collect the sound data of the people in the seats. The sound data here can be the sound of playing musical instruments, the sound of singing, the sound of the lecturer's explanation, the sound of the audience's interaction, etc.

[0047] In a specific implementation, the sound data may be data collected in real time or stored historical data.

[0048] S2, recognize the voice data of the lecturer, and extract the sound clips of singing demonstration, the sound clips of musical instrument playing, and the sound clips of knowledge explanation;

[0049] In a class, the instructor often demonstrates singing and playing multiple pieces of music. When extracting sound clips, only the singing and playing demonstration of one of the pieces of music can be extracted. The piece of music can be a randomly selected piece or the piece with the most repeated demonstrations. Of course, the singing and playing demonstration of multiple pieces of music can also be extracted, and this application does not limit this. When extracting the singing and playing demonstration of multiple pieces of music, the sound clips need to be classified according to the music, so that the instructor can be evaluated according to the music later.

[0050] In the specific implementation, the machine learning classifier can be used to identify the voice data of the lecturer and extract the sound clips of singing demonstration, instrument playing and knowledge explanation. The specific steps include:

[0051] S21. Preprocessing the original audio signal includes converting the audio signal into a digital signal, then enhancing the high frequency part by pre-emphasis, and cutting the long audio signal into short time segments (usually 20-40 millisecond frames) by framing.

[0052] S22. Extract features that are helpful for classification from the preprocessed audio, such as Mel-frequency cepstral coefficients (MFCCs), zero-crossing rate, energy, spectral centroid, etc.

[0053] S23. Machine learning classifier: Use the extracted feature vectors to train a machine learning model, such as support vector machine (SVM), random forest, neural network, etc., to identify and distinguish different types of audio clips. Use labeled datasets to train the above classifiers and optimize the performance of the model through cross-validation and other techniques.

[0054] S24. Apply the trained model to new audio data to automatically identify and mark the segments of song demonstration, instrument playing and knowledge explanation, and divide the audio stream into three types of audio segments based on the classification results.

[0055] S3. Evaluate the instructor's basic music skills based on the sound clips of singing demonstrations, instrument playing, and knowledge explanations;

[0056] Among them, the basic music skills specifically include singing skills (involving training in pitch, rhythm, timbre and expressiveness, as well as the ability to improve singing level through vocal exercises), expression ability (including musical expressiveness, emotional transmission, etc.), instrument playing ability (playing skills of specific instruments, understanding and application of instrument characteristics), music theory knowledge (in-depth understanding of music theory, such as harmony, form structure, music history, etc., which helps instructors better analyze music and pass it on to listeners), etc.

[0057] In a specific implementation, a machine learning-based evaluation model can be used to evaluate the basic music skills of the instructor. The machine learning-based evaluation model can include three evaluation sub-models, which are used to score the sound clips of singing demonstration, instrument playing demonstration and knowledge explanation. Each sub-model will score the corresponding audio clip, and then these scores can be combined through weighted average or other fusion strategies to obtain a comprehensive evaluation result of basic music skills. During the comprehensive evaluation, the weights can be adjusted according to the teaching objectives and importance to ensure that the final score reflects the capabilities of all key areas. Among them, each evaluation sub-model is trained using a data set of sound clips containing manual scores. The structure of each evaluation sub-model can be the same or different. Specifically, models such as logistic regression, support vector machine (SVM) classifier, random forest classifier, k-nearest neighbor (k-NN) classifier, etc. can be used to grade and score the sound clips.

[0058] The teacher's basic music skills are assessed through the following steps:

[0059] S31. Acoustic feature extraction: For each category of sound clips, detailed acoustic features are further extracted, such as pitch, rhythm, timbre, volume, dynamic range, etc.

[0060] S32. Establish evaluation criteria: According to the standards and requirements of music teaching, establish evaluation indicators for three aspects: singing demonstration, instrument playing demonstration, and knowledge explanation. For example, for singing demonstration, it can be evaluated from the aspects of pitch accuracy (whether it is out of tune), rhythm stability (whether it can accurately follow the rhythm of the music, including the accuracy of the beat and the ability to handle complex rhythms), pronunciation clarity (judging whether the lyrics are clearly expressed, whether the consonants and vowels are accurately pronounced, and whether the language is natural and fluent). For example, for instrument playing demonstration, it can be evaluated from the aspects of pitch accuracy (whether the pitch of the notes is accurate), rhythm stability (including the stability of the beat and the ability to handle complex rhythms), and dynamics (whether the changes in volume are appropriate). For knowledge explanation, it can be evaluated from the aspects of sound clarity (whether the pronunciation is clear and the language is easy to understand), speed and rhythm (whether the speaking speed is moderate and the language rhythm is appropriate), intonation changes (whether there are appropriate intonation fluctuations), continuity and stability (whether the performance is consistent during the explanation, including sound quality, volume and energy level), and individuality and uniqueness (whether it can show personal unique style and characteristics while maintaining professionalism).

[0061] S33. Establishing an evaluation model: Using the labeled training data set, three evaluation sub-models based on machine learning are trained to learn how to score the sound clips of singing demonstration, instrument playing demonstration and knowledge explanation respectively. The annotations in the training data set refer to the scores of the sound clips by professionals according to the above evaluation indicators.

[0062] S34. Apply the trained model to new sound clips to automatically evaluate the instructor's performance and provide a quantitative score. If necessary, an expert system in the field of music teaching can be introduced to proofread and correct the results of the machine evaluation to ensure the professionalism and reliability of the evaluation.

[0063] Through these steps, we can realize the automatic evaluation of the instructor's basic music skills, thus providing strong technical support for music teaching. Such a system can not only improve the efficiency and objectivity of evaluation, but also help teachers better understand their own teaching effects and promote professional growth.

[0064] S4, recognizing the voice data of the listener, extracting the voice clips of the singing, and sorting the singing voice clips of the same music section in the order of the singing time;

[0065] Usually, for a piece of music, after the instructor demonstrates, the listener will practice singing the music several times according to the instructor's demonstration. After each singing practice, the instructor will give specific instructions. Therefore, there will be multiple sound clips in the listener's sound data for the same piece of music. In this application, after identifying the sound clips of singing, the sound clips of singing in the same music paragraph are arranged in order according to the time sequence of singing.

[0066] S5. For each music segment, the similarity between each of the listener's singing sound segments and the instructor's singing demonstration sound segments is calculated in chronological order; and the instructor's music teaching ability is evaluated based on the similarity;

[0067] In a specific implementation, the similarity between the sound clips of the listener and the lecturer singing a song can be calculated by following the steps below:

[0068] S51. Feature extraction: extract key features from the audio of the listener and the lecturer respectively, where the key features may include pitch, rhythm, timbre, MFCC (Mel frequency cepstral coefficient), zero crossing rate, etc. For human voice, MFCC is one of the commonly used features, which can capture the spectral characteristics of the sound.

[0069] S52. Time alignment: Since the lengths of two audio clips may be different, dynamic time warping (DTW) can be used to adjust their time axes so that the corresponding notes or beats can be correctly matched. DTW is a nonlinear warping technology that satisfies the conditions of one-way correspondence, cannot turn back, and aligns from front to back.

[0070] S53, similarity calculation: using the extracted features, calculate the similarity of the two chorus songs. Cosine similarity can be used to calculate the similarity between two feature vectors, where the cosine of a 0 degree angle is 1, indicating complete consistency, and the cosine of a 180 degree angle is -1, indicating complete opposites. In addition, other distance measurement methods can also be used, such as Euclidean distance or Pearson correlation coefficient.

[0071] Among them, music teaching ability specifically refers to tutoring skills (being able to provide guidance and suggestions based on the listeners' learning situation to help them master music knowledge and skills), correction methods (when listeners make mistakes when singing or playing, the instructor needs to have the ability to correct them efficiently so that the listeners can understand and correct the mistakes), etc.

[0072] In the specific implementation, a statistical analysis is performed on the calculated multiple similarity values ​​according to the time sequence of the listeners' singing to determine the changing trend of the similarity values. If the changing trend of the similarity values ​​is getting higher and higher, it means that the listeners' singing level has improved under the guidance of the instructor, and the instructor has strong music teaching ability; if the changing trend of the similarity values ​​is not getting higher and higher, such as remaining unchanged or getting lower and lower, it means that the listeners' singing level has not been improved or progressed under the guidance of the instructor, and the instructor's music teaching ability needs to be improved.

[0073] S6. Based on the evaluation results of the instructor's basic music skills and music teaching ability, a comprehensive evaluation result of the instructor's music teaching level is obtained;

[0074] According to the importance of the two dimensions of basic music skills and music teaching ability in music teaching, corresponding weights are assigned, and weighted scores are calculated based on the weights. The weighted scores are used as comprehensive evaluation results of the instructor's music teaching ability.

[0075] S7. Determine music teaching training improvement information for instructors based on the comprehensive evaluation results.

[0076] Among them, the evaluation results include two aspects: basic music skills and music teaching ability. Correspondingly, the music teaching training improvement information also mainly includes these two aspects. Through these two aspects of music teaching training improvement information, instructors can conduct targeted training and thus improve the level of music teaching.

[0077] Corresponding to the music teaching training improvement method in the above embodiment, Figure 2 As shown, a music teaching and training improvement system in an embodiment of the present invention specifically includes: a sound data acquisition module 100, a first evaluation module 200, a second evaluation module 300, a comprehensive evaluation module 400 and a personalized training improvement recommendation module 500.

[0078] The sound data acquisition module 100 is used to acquire the sound data of the lecturer and the listener in the music class;

[0079] Among them, the music class includes a lecturer and at least one listener. The lecturer is a student majoring in music education at a vocational and technical college, that is, someone who needs training and improvement in music teaching ability. The listeners can be people who learn to play and sing music, such as inviting kindergarten children, or other interested classmates. In the music class, the lecturer will demonstrate singing, playing, singing while playing, explain knowledge, and interact with the listeners, etc. The listeners will sing along, learn to sing, answer questions, etc.

[0080] Among them, the collection of sound data can be achieved through a microphone or an audio collection device. In the specific implementation, a microphone or an audio collection device is set at the seats of the lecturer and the audience in the classroom to collect the sound data of the people in the seats. The sound data here can be the sound of playing musical instruments, the sound of singing, the sound of the lecturer's explanation, the sound of the audience's interaction, etc.

[0081] In a specific implementation, the sound data may be data collected in real time or stored historical data.

[0082] The first evaluation module 200 is used to recognize the voice data of the instructor, extract the voice clips of the song demonstration, the sound clips of the instrument playing and the sound clips of the knowledge explanation; and evaluate the basic music skills of the instructor based on the voice clips of the song demonstration, the sound clips of the instrument playing and the sound clips of the knowledge explanation;

[0083] Among them, the basic music skills specifically include singing skills (involving training in pitch, rhythm, timbre and expressiveness, as well as the ability to improve singing level through vocal exercises), expression ability (including musical expressiveness, emotional transmission, etc.), instrument playing ability (playing skills of specific instruments, understanding and application of instrument characteristics), music theory knowledge (in-depth understanding of music theory, such as harmony, form structure, music history, etc., which helps instructors better analyze music and pass it on to listeners), etc.

[0084] In a class, the instructor often demonstrates singing and playing multiple pieces of music. When extracting sound clips, only the singing and playing demonstration of one of the pieces of music can be extracted. The piece of music can be a randomly selected piece or the piece with the most repeated demonstrations. Of course, the singing and playing demonstration of multiple pieces of music can also be extracted, and this application does not limit this. When extracting the singing and playing demonstration of multiple pieces of music, the sound clips need to be classified according to the music, so that the instructor can be evaluated according to the music later.

[0085] In specific implementations, a machine learning classifier can be used to identify the instructor's voice data and extract sound clips of singing demonstrations, instrument playing, and knowledge explanations.

[0086] In a specific implementation, a machine learning-based evaluation model can be used to evaluate the basic music skills of the instructor. The machine learning-based evaluation model can include three evaluation sub-models, which are used to score the sound clips of singing demonstration, instrument playing demonstration and knowledge explanation. Each sub-model will score the corresponding audio clip, and then these scores can be combined through weighted average or other fusion strategies to obtain a comprehensive evaluation result of basic music skills. During the comprehensive evaluation, the weights can be adjusted according to the teaching objectives and importance to ensure that the final score reflects the capabilities of all key areas. Among them, each evaluation sub-model is trained using a data set of sound clips containing manual scores. The structure of each evaluation sub-model can be the same or different. Specifically, models such as logistic regression, support vector machine (SVM) classifier, random forest classifier, k-nearest neighbor (k-NN) classifier, etc. can be used to grade and score the sound clips.

[0087] The second evaluation module 300 recognizes the voice data of the listener, extracts the voice clips of the singing, and sorts the voice clips of the same music segment in the order of singing time; for each music segment, calculates the similarity between each voice clip of the listener and the voice clip of the teacher's singing demonstration in the order of time; and evaluates the music teaching ability of the teacher based on the similarity;

[0088] Among them, music teaching ability specifically refers to tutoring skills (being able to provide guidance and suggestions based on the listeners' learning situation to help them master music knowledge and skills), correction methods (when listeners make mistakes when singing or playing, the instructor needs to have the ability to correct them efficiently so that the listeners can understand and correct the mistakes), etc.

[0089] Usually, for a piece of music, after the instructor demonstrates, the listener will practice singing the music several times according to the instructor's demonstration. After each singing practice, the instructor will give specific instructions. Therefore, there will be multiple sound clips in the listener's sound data for the same piece of music. In this application, after identifying the sound clips of singing, the sound clips of singing in the same music paragraph are arranged in order according to the time sequence of singing.

[0090] In the specific implementation, a statistical analysis is performed on the calculated multiple similarity values ​​according to the time sequence of the listeners' singing to determine the changing trend of the similarity values. If the changing trend of the similarity values ​​is getting higher and higher, it means that the listeners' singing level has improved under the guidance of the instructor, and the instructor has strong music teaching ability; if the changing trend of the similarity values ​​is not getting higher and higher, such as remaining unchanged or getting lower and lower, it means that the listeners' singing level has not been improved or progressed under the guidance of the instructor, and the instructor's music teaching ability needs to be improved.

[0091] A comprehensive evaluation module 400 obtains a comprehensive evaluation result of the music teaching ability of the instructor based on the evaluation results of the first evaluation module and the second evaluation module;

[0092] In the specific implementation, the comprehensive evaluation module assigns corresponding weights according to the importance of the two dimensions of basic music skills and music teaching ability in music teaching, calculates the weighted score based on the weights, and uses the weighted score as the comprehensive evaluation result of the instructor's music teaching ability.

[0093] The personalized training improvement recommendation module 500 determines the music teaching training improvement information for the instructor based on the comprehensive evaluation results obtained by the comprehensive evaluation module.

[0094] Among them, the evaluation results include two aspects: basic music skills and music teaching ability. Correspondingly, the music teaching training improvement information also mainly includes these two aspects. Through these two aspects of music teaching training improvement information, instructors can conduct targeted training and thus improve the level of music teaching.

[0095] For ease of understanding, here are three specific application scenario examples:

[0096] Application scenario 1:

[0097] Music teaching methods class: In the music teaching methods class of vocational and technical colleges, this system is used as a teaching auxiliary tool. It helps teachers understand the teaching performance of each student and provide targeted improvement plans to ensure that students meet the requirements of teaching ability.

[0098] Application scenario 2:

[0099] Pre-practice training: Before students enter the internship stage, they use this system to conduct training in a simulated teaching environment. The system evaluates students’ teaching performance and gives training suggestions to enhance their practical teaching skills so that they can pass the internship smoothly.

[0100] Application scenario 3:

[0101] Teacher Certification Preparation: When students are preparing for music teacher certification, this system provides personalized tutoring to help students master the teaching skills and knowledge required for the exam and improve their pass rate.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for improving music teaching and training, characterized in that: The method comprises: Acquire the sound data of the lecturer and the listener in the music class; Recognize the lecturer's voice data and extract the sound clips of singing demonstration, instrument playing and knowledge explanation; Evaluate the instructor's basic music skills based on audio clips of song demonstrations, instrument playing, and knowledge explanations; Recognize the voice data of the listener, extract the voice clips of the singing, and sort the singing voice clips of the same music section in the order of the singing time; For each music segment, the similarity between each of the listener's singing sound segments and the instructor's singing demonstration sound segments is calculated in chronological order; the instructor's music teaching ability is evaluated based on the similarity; Based on the evaluation results of the instructor's basic music skills and music teaching ability, a comprehensive evaluation result of the instructor's music teaching level is obtained; Based on the comprehensive evaluation results, determine the music teaching training improvement information for the instructor.

2. A music teaching and training improvement method according to claim 1, characterized in that: Acquiring the sound data of the lecturer and the listener in the music class includes: collecting the sound data in real time through microphones or audio collection devices set at the seats of the lecturer and the listener in the music class.

3. A music teaching and training improvement method according to claim 1, characterized in that: Based on the sound clips of singing demonstration, instrument playing and knowledge explanation, the instructor's basic music skills are evaluated, including: An evaluation model based on machine learning is used to evaluate the instructor's basic music skills. The evaluation model based on machine learning includes three evaluation sub-models, which are used to score sound clips of singing demonstration, instrument playing demonstration and knowledge explanation respectively. Each sub-model will score the corresponding audio clip, and the three scores will be combined through weighted average to obtain a comprehensive basic music skills evaluation result; wherein each evaluation sub-model is trained using a sound clip dataset containing manual ratings.

4. A music teaching and training improvement method according to claim 1, characterized in that: Evaluate the instructor's music teaching ability based on similarity, including: Performing statistical analysis on the calculated multiple similarity values ​​according to the time sequence of the listeners singing the songs, and determining the changing trend of the similarity values; According to the changing trend of the similarity value, the music teaching ability of the instructor is evaluated.

5. A music teaching and training improvement system, characterized in that: include: A sound data acquisition module is used to acquire the sound data of the lecturer and the listener in the music class; The first evaluation module recognizes the voice data of the lecturer and extracts the sound clips of singing demonstration, instrument playing and knowledge explanation; And based on the sound clips of singing demonstration, instrument playing and knowledge explanation, the instructor's basic music skills are evaluated; The second evaluation module recognizes the voice data of the listener, extracts the sound clips of the singing, and sorts the sound clips of the same music segment in the order of singing time; for each music segment, the similarity between each of the listener's singing sound clips and the teacher's singing demonstration sound clips is calculated in chronological order; the teacher's music teaching ability is evaluated based on the similarity; A comprehensive evaluation module, which obtains a comprehensive evaluation result of the instructor's music teaching level based on the evaluation result of the first evaluation module on the instructor's music basic skills and the evaluation result of the second evaluation module on the instructor's music teaching ability; The personalized training improvement recommendation module determines the music teaching training improvement information for the instructor based on the comprehensive evaluation results obtained by the comprehensive evaluation module.

6. A music teaching and training improvement system according to claim 5, characterized in that: The sound data acquisition module is specifically used to collect sound data in real time through microphones or audio acquisition devices set at the seats of lecturers and listeners in the music classroom.

7. A music teaching and training improvement system according to claim 5, characterized in that: The first evaluation module is specifically used for: An evaluation model based on machine learning is used to evaluate the instructor's basic music skills. The evaluation model based on machine learning includes three evaluation sub-models, which are used to score sound clips of singing demonstration, instrument playing demonstration and knowledge explanation respectively. Each sub-model will score the corresponding audio clip, and the three scores will be combined through weighted average to obtain a comprehensive basic music skills evaluation result; wherein each evaluation sub-model is trained using a sound clip dataset containing manual ratings.

8. A music teaching and training improvement system according to claim 5, characterized in that: The second evaluation module is specifically used for: Performing statistical analysis on the calculated multiple similarity values ​​according to the time sequence of the listeners singing the songs, and determining the changing trend of the similarity values; According to the changing trend of the similarity value, the music teaching ability of the instructor is evaluated.

Images