Classroom evaluation method, device and equipment and storage medium

Abstract

The invention discloses a classroom evaluation method, apparatus and device, and a storage medium. The method comprises the steps of obtaining audio data of a to-be-evaluated classroom; performing voice recognition on the audio data to obtain text data; preprocessing the text data to obtain multiple kinds of preprocessed data; the evaluation indexes corresponding to the plurality of pre-processed data are different; inputting the plurality of pre-processed data into a large language model to obtain evaluation results of the evaluation indexes corresponding to the plurality of pre-processed data; and determining the evaluation result of the to-be-evaluated classroom based on the evaluation results corresponding to the plurality of pre-processed data. According to the method, the machine learning model can be used for replacing manual work to carry out classroom observation evaluation, the labor cost is reduced, the feedback speed is increased, and the evaluation effectiveness is high.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a classroom evaluation method, device, equipment and storage medium. Background Art

[0002] Classroom observation is a widely used educational research method, usually conducted by researchers or professional teachers to systematically observe, record, and analyze teaching and learning behaviors in the classroom to gain insights into educational practice. The goal of classroom observation is to provide an in-depth understanding of classroom dynamics and obtain effective classroom evaluations, thereby providing educators with feedback on teaching effectiveness, student engagement, teaching strategies, and classroom environment. Traditional classroom observation and evaluation methods mainly involve observers entering the classroom or observing classroom videos, and using pre-designed observation tools such as scales or checklists to systematically record specific teaching and learning behaviors. Since traditional classroom observation and evaluation methods require in-depth analysis by professionals, and different schools or subjects may use different evaluation criteria, there are problems such as inconsistent evaluation criteria, high costs, and long feedback times. Summary of the invention

[0003] The embodiments of the present application provide a classroom evaluation method, apparatus, device and storage medium, which can use a machine learning model to replace manual classroom observation and evaluation, thereby reducing labor costs, speeding up feedback, and having higher evaluation effectiveness.

[0004] In a first aspect, an embodiment of the present application provides a classroom evaluation method, the method comprising:

[0005] Get the audio data of the class to be evaluated;

[0006] Performing speech recognition on the above audio data to obtain text data;

[0007] Preprocessing the text data to obtain a plurality of preprocessed data; the plurality of preprocessed data respectively correspond to different evaluation indicators;

[0008] Inputting the above-mentioned multiple pre-processed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the above-mentioned multiple pre-processed data;

[0009] The evaluation result of the above-mentioned class to be evaluated is determined based on the evaluation results corresponding to each of the above-mentioned multiple pre-processed data.

[0010] In a possible implementation, the plurality of pre-processed data include classroom teaching data, classroom question-and-answer data, and teacher-student interactive question-and-answer data;

[0011] The above text data is preprocessed to obtain a variety of preprocessed data, including:

[0012] Performing data cleaning on the above text data to obtain the above classroom teaching data;

[0013] Extracting the above classroom question and answer data based on the above classroom teaching data;

[0014] The teacher-student interactive question and answer data is extracted based on the classroom question and answer data.

[0015] In a possible implementation, the extracting the classroom question and answer data based on the classroom teaching data includes:

[0016] Based on preset question and answer recognition rules, all question and answer dialogues are extracted from the above classroom teaching data to obtain the above classroom question and answer data.

[0017] In a possible implementation, performing speech recognition on the audio data to obtain text data includes:

[0018] Obtain voiceprint information of teachers and / or students;

[0019] Based on the voiceprint information, speech recognition is performed on the audio data to obtain text data with identity identifiers; the identity identifiers include teacher identifiers and student identifiers.

[0020] In a possible implementation, the extracting the teacher-student interactive question-and-answer data based on the classroom question-and-answer data includes:

[0021] Based on the above identity identifiers in the above classroom question and answer data, question and answer dialogues containing both the teacher identifier and the student identifier are extracted from the above classroom question and answer data to obtain the above teacher-student interactive question and answer data.

[0022] In a possible implementation, the plurality of pre-processed data are input into the large language model to obtain evaluation results of the evaluation indicators corresponding to the plurality of pre-processed data, including:

[0023] Inputting the classroom teaching data into the large language model to obtain an evaluation result corresponding to a first evaluation indicator, wherein the first evaluation indicator includes at least one of a curriculum standard evaluation and a curriculum standard improvement;

[0024] Inputting the classroom question and answer data into the large language model to obtain an evaluation result corresponding to a second evaluation indicator, wherein the second evaluation indicator includes at least one of effectiveness analysis, knowledge goal, question avoidance, and question optimization;

[0025] The teacher-student interactive question-and-answer data is input into a large language model to obtain an evaluation result corresponding to a third evaluation indicator, where the third evaluation indicator includes at least one of an interactive evaluation and an interactive suggestion.

[0026] In a possible implementation, the large language model is pre-trained and / or fine-tuned based on education field data and the evaluation indicators.

[0027] In a second aspect, an embodiment of the present application provides a classroom evaluation device, the device comprising:

[0028] The acquisition module is used to obtain the audio data of the class to be evaluated;

[0029] A recognition module, used to perform speech recognition on the above audio data to obtain text data;

[0030] A preprocessing module is used to preprocess the text data to obtain a variety of preprocessed data; the evaluation indicators corresponding to the various preprocessed data are different;

[0031] A first evaluation module is used to input the above-mentioned multiple pre-processed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the above-mentioned multiple pre-processed data;

[0032] The second evaluation module is used to determine the evaluation result of the above-mentioned class to be evaluated based on the evaluation results corresponding to the above-mentioned multiple pre-processed data.

[0033] In a possible implementation, the plurality of pre-processed data include classroom teaching data, classroom question-and-answer data, and teacher-student interactive question-and-answer data;

[0034] The above-mentioned preprocessing module includes:

[0035] A data cleaning unit, used for cleaning the text data to obtain the classroom teaching data;

[0036] A first extraction unit, configured to extract the classroom question-and-answer data based on the classroom teaching data;

[0037] The second extraction unit is used to extract the teacher-student interactive question and answer data based on the classroom question and answer data.

[0038] In a possible implementation, the first extraction unit is specifically configured to:

[0039] Based on preset question and answer recognition rules, all question and answer dialogues are extracted from the above classroom teaching data to obtain the above classroom question and answer data.

[0040] In a possible implementation, the identification module is specifically used to:

[0041] Obtain voiceprint information of teachers and / or students;

[0042] Based on the voiceprint information, speech recognition is performed on the audio data to obtain text data with identity identifiers; the identity identifiers include teacher identifiers and student identifiers.

[0043] In a possible implementation, the second extraction unit is specifically configured to:

[0044] Based on the above identity identifiers in the above classroom question and answer data, question and answer dialogues containing both the teacher identifier and the student identifier are extracted from the above classroom question and answer data to obtain the above teacher-student interactive question and answer data.

[0045] In a possible implementation, the first evaluation module is specifically used to:

[0046] Inputting the classroom teaching data into the large language model to obtain an evaluation result corresponding to a first evaluation indicator, wherein the first evaluation indicator includes at least one of a curriculum standard evaluation and a curriculum standard improvement;

[0047] Inputting the classroom question and answer data into the large language model to obtain an evaluation result corresponding to a second evaluation indicator, wherein the second evaluation indicator includes at least one of effectiveness analysis, knowledge goal, question avoidance, and question optimization;

[0048] The teacher-student interactive question-and-answer data is input into a large language model to obtain an evaluation result corresponding to a third evaluation indicator, where the third evaluation indicator includes at least one of an interactive evaluation and an interactive suggestion.

[0049] In a possible implementation, the large language model is pre-trained and / or fine-tuned based on education field data and the evaluation indicators.

[0050] In a third aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above method steps.

[0051] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above method steps.

[0052] In one or more embodiments of the present application, audio data of the classroom to be evaluated is obtained; speech recognition is performed on the audio data to obtain text data; the text data is preprocessed to obtain a variety of preprocessed data; the evaluation indicators corresponding to the various preprocessed data are different; the various preprocessed data are input into a large language model to obtain evaluation results of the evaluation indicators corresponding to the various preprocessed data; the evaluation results of the classroom to be evaluated are determined based on the evaluation results corresponding to the various preprocessed data. This method can use a machine learning model to replace manual classroom observation and evaluation, which reduces labor costs, speeds up feedback, and has high evaluation effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A system architecture diagram of a classroom evaluation system provided in an embodiment of the present application;

[0055] Figure 2 A flow chart of a classroom evaluation method provided in an embodiment of the present application;

[0056] Figure 3 A schematic diagram of a flow chart of text data preprocessing provided in an embodiment of the present application;

[0057] Figure 4 A Venn diagram between text data and various pre-processed data provided in an embodiment of the present application;

[0058] Figure 5 A flowchart of another classroom evaluation method provided in an embodiment of the present application;

[0059] Figure 6 A classroom observation evaluation indicator framework diagram provided for an embodiment of the present application;

[0060] Figure 7 A schematic diagram of the structure of a classroom evaluation device provided in an embodiment of the present application;

[0061] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0062] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0063] In the description of the present application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0064] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the embodiments of this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions. For example, the audio data of the classroom to be evaluated and the education field data involved in this specification are all obtained with full authorization.

[0065] The present application is described in detail below with reference to specific embodiments.

[0066] Please refer to Figure 1 , Figure 1 The following is a schematic diagram of the architecture of a classroom evaluation system provided by an exemplary embodiment of the present application. Figure 1 As shown, the classroom evaluation system may include a terminal 110 and a server 120 .

[0067] The terminal 110 is connected to the server 120 via the network 130. Optionally, the terminal 110 is a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart watch, etc., but is not limited thereto. The terminal 110 installs and runs an application that supports classroom evaluation. The network 130 can be a wireless network or a wired network.

[0068] The server 120 is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, distribution networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The server 120 provides background services for applications running on the terminal 110. For example, the server 120 provides corresponding services for applications running on the terminal 110. In the embodiment of the present application, the server 120 provides background services for applications running on the terminal that support classroom evaluation, such as running or training a large language model for classroom evaluation.

[0069] Those skilled in the art will appreciate that the number of the above-mentioned terminals 110 and servers 120 may be more or less. For example, there may be only one terminal 110 and server 120, or there may be dozens or hundreds of terminals 110 and servers 120, or a greater number. In this case, the above-mentioned system architecture may also include other terminals and servers. The embodiment of the present application does not limit the number of terminals and device types.

[0070] The classroom evaluation method involved in one or more embodiments of the present application can be implemented by a computer program and can be run on a classroom evaluation device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application. The execution subject corresponding to the classroom evaluation method can be the above-mentioned terminal 110, and the server 120 can be responsible for running or training a large language model for classroom evaluation, and sending the model or model call interface to the terminal 110; the execution subject corresponding to the classroom evaluation method can also be the server 120, and the terminal 110 can be responsible for uploading the audio data of the classroom to be evaluated to the server 120. It is determined specifically based on the actual application environment, and the embodiments of the present application do not limit this.

[0071] Next, combine Figure 1 , introduces the classroom evaluation method provided by the embodiment of this specification. For details, please refer to Figure 2 , which is a flow chart of a classroom evaluation method provided by an exemplary embodiment of the present application. Figure 2 As shown, with the execution subject being the server, the classroom evaluation method includes the following steps:

[0072] S201, obtaining audio data of the class to be evaluated.

[0073] Specifically, the classroom to be evaluated in this embodiment is a classroom that needs to be observed and evaluated, and the place where educational or teaching activities are carried out includes but is not limited to classrooms, conference rooms in schools, and online education platforms. The classroom to be evaluated can be an education or teaching unit (such as an online course), which usually includes teacher lectures, student participation, questions and answers, discussions, etc., reflecting the students' learning status and the teacher's teaching methods. By evaluating and analyzing the classroom to be evaluated, educators can understand students' learning situation, improve teaching methods and strategies, and thus improve teaching quality and effectiveness.

[0074] The audio data in this embodiment may be an audio or video recording of the class to be evaluated, that is, after the class to be evaluated is recorded or videotaped, the audio data in the recording or video recording is collected for subsequent processing and analysis to obtain an evaluation result of the class to be evaluated.

[0075] S202, performing speech recognition on the audio data to obtain text data.

[0076] Specifically, this embodiment performs speech recognition and text conversion on the audio data to convert the speech content in the audio data into text content to obtain text data.

[0077] Speech recognition can be achieved using ASR technology (Automatic Speech Recognition). ASR is an automatic speech recognition system based on artificial intelligence and speech signal processing technology, which can convert human speech into computer-readable text or commands to achieve human-computer interaction and natural language processing. In some embodiments, models such as CTC (Connectionist Temporal Classification), AED (Audio Event Detection), and RNN-T (Recurrent Neural Network Transducer) can be used to perform speech recognition on audio data to obtain corresponding text data.

[0078] In some embodiments, the above-mentioned voice recognition of audio data to obtain text data includes: obtaining voiceprint information of teachers and / or students; performing voice recognition on audio data based on the voiceprint information to obtain text data with identity identification; the identity identification includes a teacher identification and a student identification.

[0079] Specifically, voiceprint recognition is a technology that verifies or classifies an individual's identity by analyzing and comparing their voice features. The voiceprint information of teachers / students is feature information with identity identifiers extracted from audio data containing the voice of teachers / students based on voiceprint recognition technology. These feature information may include the timbre, pitch, speaking speed, pronunciation habits, etc. of the voice corresponding to the identity identifier. In this embodiment, the voiceprint information of teachers and / or students can be learned through voiceprint recognition technology to distinguish the voices of teachers and students, and the text data can be annotated by the identity identifier to obtain the corresponding teacher text data and student text data.

[0080] It should be noted that before performing speech recognition on audio data, the audio data needs to be processed to improve the quality of the speech signal and reduce the influence of noise, aliasing and other factors, so as to make the subsequent speech recognition more accurate.

[0081] S203, preprocessing the text data to obtain a variety of preprocessed data.

[0082] Specifically, this embodiment preprocesses the text data converted by speech recognition, which can reduce redundant information and reduce the input length of subsequent models, thereby alleviating model pressure; it can also output preprocessed data of various degrees (containing different information content or information volume) to obtain more standardized, accurate and useful preprocessed data. Each type of preprocessed data corresponds to a corresponding evaluation index, and different preprocessed data correspond to different evaluation indicators. By specifying the evaluation index, the corresponding preprocessed data can be analyzed and processed in a targeted manner to obtain more accurate and effective classroom evaluation results.

[0083] In some embodiments, the above-mentioned multiple pre-processed data include classroom teaching data, classroom question and answer data, and teacher-student interactive question and answer data; for example, Figure 3 As shown in , the information content or amount of each pre-processed data is different. The classroom teaching data is a subset of the text data, the classroom question and answer data is a subset of the classroom teaching data, and the teacher-student interactive question and answer data is a subset of the classroom question and answer data. Figure 4 As shown, the above preprocessing of text data to obtain a variety of preprocessed data specifically includes the following steps:

[0084] S401, cleaning the text data to obtain classroom teaching data.

[0085] Specifically, classroom teaching data is data after data cleaning of text data after speech recognition, which contains information of the entire classroom to be evaluated, such as the content of the teacher's lectures, the content of the teacher's questions, the content of the teacher-student interaction, etc., which can reflect the overall course nature of the classroom to be evaluated. In some embodiments, the above-mentioned data cleaning of text data to obtain classroom teaching data specifically includes: cleaning out redundant, erroneous or incomplete data in the text data; removing stop words in the text data. Exemplarily, the stop words may include some commonly used oral expressions and interjections, such as "um", "ah", etc., and may include some words that are not related to course teaching, such as "stand up in class", "end of get out of class" and other fixed templates. Stop words can be specifically set according to the specific implementation scenario, and are not limited here.

[0086] S402, extracting classroom question and answer data based on classroom teaching data.

[0087] Specifically, the classroom question and answer data includes all dialogue data about questions and answers in the classroom teaching data, such as the content of students' questions and their answers, the content of teachers' self-questions and their answers, the content of teachers' questions and their answers, etc., which can reflect the teaching situation of teachers in the classroom to be evaluated. In some embodiments, the above-mentioned extraction of classroom question and answer data based on classroom teaching data includes: extracting all question and answer dialogues from the classroom teaching data based on preset question and answer recognition rules to obtain classroom question and answer data.

[0088] Exemplarily, the preset question-answer recognition rules may be some rules formulated for identifying question-answer dialogues. For example, rules may be set to identify sentences or phrases containing question words (such as "what", "why", "how", etc.), or sentences or phrases containing answering tones (such as "yes", "no", "right", etc.). The preset question-answer recognition rules may be natural speech processing technology (natural language generation, natural language understanding, etc.) or trained machine learning models (such as recurrent neural networks, long short-term memory networks, etc.). By using the above technologies to identify and extract question-answer dialogues, it can help identify question types, question content, answer content, etc., so as to more accurately extract question-answer dialogues.

[0089] S403, extracting teacher-student interactive question-and-answer data based on the classroom question-and-answer data.

[0090] Specifically, the teacher-student interactive question and answer data includes dialogue data about teacher-student interaction in the classroom question and answer data, such as the content of the teacher's questions and the student's answers, the content of the student's questions and the teacher's answers, etc., which can reflect the status of student learning in the classroom to be evaluated. In some embodiments, the above-mentioned extraction of the teacher-student interactive question and answer data based on the classroom question and answer data includes: based on the identity identifier in the above-mentioned classroom question and answer data, extracting the question and answer dialogue containing both the teacher identifier and the student identifier from the above-mentioned classroom question and answer data to obtain the above-mentioned teacher-student interactive question and answer data.

[0091] Exemplarily, classroom question-and-answer data can be segmented based on the content of the question-and-answer dialogue, with each segment being a question-and-answer dialogue (such as one question and one answer). Voiceprint recognition technology can be used to distinguish between teacher text and student text in text data; when teacher text and student text appear in a question-and-answer dialogue at the same time, it means that the question-and-answer dialogue belongs to a teacher-student interactive dialogue and is classified as teacher-student interactive question-and-answer data.

[0092] S204, inputting the various pre-processed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the various pre-processed data.

[0093] Specifically, each type of preprocessed data corresponds to a specific evaluation indicator. When the preprocessed data is input into the large language model, the model will determine the corresponding evaluation indicator based on the input preprocessed data, and evaluate the input preprocessed data based on the determined evaluation indicator to generate an evaluation result corresponding to the evaluation indicator. For example, if the preprocessed data input into the model is teacher-student interactive question-and-answer data, and the evaluation indicator corresponding to the teacher-student interactive question-and-answer data is interactive evaluation, then the large language model will evaluate the input teacher-student interactive question-and-answer data based on the interactive evaluation indicator, and output the specific evaluation content related to the interactive evaluation, that is, the evaluation result of the interactive evaluation.

[0094] By inputting various preprocessed data into the large language model separately, we can obtain the evaluation results of the corresponding evaluation indicators of various preprocessed data. This can prevent the input of too much irrelevant information, reduce the pressure on the model, and improve the efficiency and accuracy of model evaluation through targeted data analysis.

[0095] Among them, the evaluation indicators are set according to the needs and goals of classroom observation evaluation. For example, they can be set based on the LICC observation framework, that is, according to the four dimensions of student learning (Learning), teacher teaching (Instruction), course nature (Curriculum) and classroom culture (Culture), a series of professional evaluation indicators are determined and matched with the most relevant pre-processed data.

[0096] The large language model (LLM) used in the embodiment of the present application is a natural language processing (Natural Language Processing, NLP) large model, which belongs to an artificial intelligence technology, which can be trained by massive text data, learn the structure, rules and semantics of language, so as to generate text with natural language style or answer natural language questions. Large language models are usually based on neural network technology, containing billions of parameters, and can be trained and optimized by large-scale training data. The large language model used in the embodiment of the present application can use some general large language model architectures, such as LLaMa2, ChatGLM, etc., and can make the large language model clarify specific task requirements and goals by setting prompt words, etc., to achieve the evaluation results of the corresponding evaluation indicators according to the pre-processed data output.

[0097] In some embodiments, the large language model can also be pre-trained and / or fine-tuned based on evaluation indicators corresponding to education field data and pre-processed data to make it more suitable for processing text data and tasks related to classroom observation and evaluation.

[0098] Specifically, in the pre-training stage (Pretrain), a large amount of education field text data can be used to train the general large model. The education field data may include:

[0099] Textbook information: This section includes various versions of educational textbooks, which can cover multiple subjects such as Chinese, mathematics, English, science, etc. These textbook contents can provide basic knowledge points and teaching materials, and are important references for teaching activities.

[0100] Courseware information: This part includes the text content on the courseware, as well as the standards or explanations corresponding to these text contents, which are usually used to assist teaching or learning understanding.

[0101] Classroom video: This part uses speech-to-text technology (ASR) to convert the voice of teachers and students during class into text, which can provide the interaction and communication content between teachers and students during class.

[0102] Lesson plan: This part includes the teacher's pre-design and planning of classroom content, including teaching objectives, content, methods, steps, etc. It can provide teachers with detailed information on how to arrange teaching content and organize teaching activities. It is the core guiding document for teaching activities.

[0103] Knowledge point system information: This part includes the hierarchical relationship of knowledge points in each subject, that is, the mutual relationship and logical structure between each knowledge point.

[0104] Exercise Bank: This part includes homework and exam questions of various subjects.

[0105] Classroom evaluation information: This information comes from the evaluation indicators and evaluation reports of previous classroom observations.

[0106] Other information related to teaching: such as teaching software, teaching strategies and other types of teaching resources or information.

[0107] After the pre-training step, the large language model can fully learn the content in the education field and obtain a pre-trained model in the education field.

[0108] In the fine-tuning (Supervised Fine-Tuning, SFT) stage, the model can be supervised fine-tuned to make it more suitable for classroom observation evaluation according to the task requirements of the embodiment of the present application. For example, a large amount of pre-processed data of classroom observations and known evaluation results of corresponding evaluation indicators can be used to fine-tune the general large language model or the pre-trained model in the above-mentioned education field to improve the accuracy and effectiveness of the evaluation results output by it.

[0109] S205, determining the evaluation result of the class to be evaluated based on the evaluation results corresponding to the various preprocessed data.

[0110] Specifically, the evaluation results of various preprocessed data are integrated and summarized. Different preprocessed data use different evaluation indicators and evaluation methods. After summarizing the evaluation results of various preprocessed data, a comprehensive and fair evaluation result of the class to be evaluated can be obtained. The final evaluation result of the class to be evaluated can be an evaluation report containing the specific evaluation content of each evaluation indicator.

[0111] In one or more embodiments of the present application, audio data of the class to be evaluated is obtained; speech recognition is performed on the audio data to obtain text data; the text data is preprocessed to obtain a variety of preprocessed data; the evaluation indicators corresponding to the various preprocessed data are different; the various preprocessed data are input into a large language model to obtain evaluation results of the evaluation indicators corresponding to the various preprocessed data; the evaluation results of the class to be evaluated are determined based on the evaluation results corresponding to the various preprocessed data. This method can use a machine learning model to replace manual classroom observation and evaluation, which reduces labor costs, speeds up feedback, and has high evaluation effectiveness, and can provide better support and guidance for teachers and students.

[0112] Please refer to the following Figure 5 , which is a flow chart of another classroom evaluation method provided by an exemplary embodiment of the present application. Figure 5 As shown in the figure, the classroom evaluation method includes the following steps:

[0113] S501, obtaining audio data of the class to be evaluated.

[0114] Specifically, S501 is consistent with S201 and will not be described in detail here.

[0115] S502: Perform speech recognition on the audio data to obtain text data.

[0116] Specifically, S502 is consistent with S202 and will not be described in detail here.

[0117] S503, performing data cleaning on the text data to obtain classroom teaching data.

[0118] Specifically, redundant, erroneous or incomplete data in the text data is cleaned, and stop words in the text data are removed to obtain classroom teaching data.

[0119] S504, extracting classroom question and answer data based on classroom teaching data.

[0120] Specifically, all question-and-answer dialogues are extracted from classroom teaching data based on preset question-and-answer recognition rules to obtain classroom question-and-answer data.

[0121] S505, extracting teacher-student interactive question-and-answer data based on the classroom question-and-answer data.

[0122] Specifically, based on the identity identifiers in the classroom question and answer data, question and answer dialogues containing both the teacher identifier and the student identifier are extracted from the classroom question and answer data to obtain teacher-student interactive question and answer data.

[0123] S506, inputting the classroom teaching data into the large language model to obtain an evaluation result corresponding to the first evaluation indicator.

[0124] Specifically, Figure 6 The figure shows a framework diagram of a classroom observation evaluation index proposed in the embodiment of the present application; the framework divides classroom observation into three dimensions and eight evaluation indicators. Based on the framework and the large language model, the classroom observation evaluation task can be abstracted into a natural language generation problem, which inputs the relevant data of classroom observation and outputs the classroom observation evaluation (corresponding to the evaluation indicators of eight dimensions).

[0125] In this embodiment, under the dimension of course nature, in order to examine whether the teaching objectives have been achieved, it is necessary to input the information of the entire class, that is, the classroom teaching data. The first evaluation indicator under this dimension includes course standard evaluation and course standard improvement. In the field of education, course standards refer to the standards of the course, that is, the basic norms and requirements of the course. Both course standard evaluation and course standard improvement are aimed at better understanding and improving the teaching objectives of the course.

[0126] Curriculum standard evaluation refers to the evaluation of the relationship between the current curriculum and the new curriculum standard. The new curriculum standard is the basic programmatic document of the national curriculum, which reflects the basic norms and quality requirements of the country for basic education curriculum. By comparing the current curriculum with the new curriculum standard and evaluating their conformity and adaptability, we can better understand whether the curriculum meets the basic norms and quality requirements of the country, and whether it can effectively help students achieve all-round development.

[0127] Curriculum improvement refers to the adjustment and optimization of the requirements of the new curriculum standards to better meet the requirements of the new curriculum standards, to cultivate students' innovative ability, practical ability, humanistic qualities and interdisciplinary integration, and to comprehensively improve students' comprehensive quality.

[0128] It can be understood that this embodiment inputs classroom teaching data into a large language model, and after reasoning and analysis by the large model, an evaluation report containing specific contents of curriculum evaluation and curriculum improvement can be output.

[0129] S507, input the classroom question and answer data into the large language model to obtain the evaluation result corresponding to the second evaluation indicator.

[0130] Specifically, Figure 6 As shown, in this embodiment, in the teacher teaching dimension, in order to analyze the teaching dialogue content, it is necessary to input all the question-and-answer dialogues in the entire class, including the teacher's own self-questioning and self-answering form, that is, the classroom question-and-answer data. The second evaluation index under this dimension includes effectiveness analysis, knowledge goals, question avoidance, and question optimization.

[0131] Effectiveness analysis refers to analyzing whether the teacher's current questions are effective. So-called effective questions mean that the questions should focus on teaching objectives, implement teaching priorities, have thinking value, and help promote student learning. The questions should be concise and clear, of appropriate difficulty, and closed questions and open questions should be combined. Effectiveness analysis can be considered comprehensively based on the questions themselves and the students' answers, so it is necessary to analyze all the question-answering dialogues in the classroom question-answering data.

[0132] Knowledge objectives refer to the knowledge points that students are required to master or understand during the course or teaching process. These objectives can include facts, concepts, principles, etc., as well as the relationship and application between them.

[0133] Question avoidance refers to avoiding questions that may cause students to feel uncomfortable or have a negative effect during teaching. These questions may involve personal privacy, embarrassing or controversial topics. Good questioning skills can help students develop their thinking and creativity while avoiding unnecessary social conflicts.

[0134] Question optimization refers to asking high-quality questions during the teaching process that can stimulate students' thinking ability, help students understand knowledge points, and encourage them to actively participate in discussions.

[0135] It can be understood that this embodiment inputs classroom question and answer data into a large language model. After reasoning and analysis by the large model, an evaluation report containing specific contents of effectiveness analysis, knowledge objectives, question avoidance and question optimization can be output.

[0136] S508, inputting the teacher-student interactive question-and-answer data into the large language model to obtain an evaluation result corresponding to the third evaluation indicator.

[0137] Specifically, Figure 6 As shown, in this embodiment, in the student learning dimension, in order to examine the interaction between teachers and students, it is necessary to input the interactive question-and-answer dialogue between teachers and students, that is, the teacher-student interactive question-and-answer data. The third evaluation index under this dimension includes interactive evaluation and interactive suggestions.

[0138] Interactive evaluation refers to the evaluation of the interactive question-and-answer data between teachers and students. It focuses on the quality of communication and interaction between teachers and students, including the quality of questions, the accuracy of answers, the depth and breadth of discussions, etc. By analyzing these data, we can evaluate the learning effect of students and the guidance effect of teachers to find out the areas for improvement and the methods that need to be improved.

[0139] Interaction suggestions refer to improvement suggestions based on the results of interactive evaluation. These suggestions may include how to improve the quality of questions, how to guide students to think more deeply, how to increase opportunities for interaction between teachers and students, etc. These suggestions can help teachers better design teaching links and students' learning paths to promote more effective learning outcomes. At the same time, students can also improve their learning strategies and thinking methods based on these suggestions to improve learning outcomes.

[0140] It can be understood that in this embodiment, the teacher-student interactive question and answer data is input into the large language model. After the reasoning analysis of the large model, an evaluation report containing the specific content of the interactive evaluation and interactive suggestions can be output.

[0141] S509: Determine the evaluation result of the class to be evaluated based on the evaluation results corresponding to each evaluation indicator.

[0142] Specifically, the evaluation results corresponding to each evaluation indicator are integrated and summarized to generate the evaluation results of the class to be evaluated.

[0143] Please refer to the following Figure 7 , Figure 7 The following is a schematic diagram of a classroom evaluation device provided by an exemplary embodiment of the present application. Figure 7As shown, the classroom evaluation device 700 includes:

[0144] An acquisition module 710 is used to acquire audio data of the class to be evaluated;

[0145] The recognition module 720 is used to perform speech recognition on the above audio data to obtain text data;

[0146] A preprocessing module 730 is used to preprocess the text data to obtain a plurality of preprocessed data; each of the plurality of preprocessed data corresponds to a different evaluation index;

[0147] A first evaluation module 740 is used to input the above-mentioned multiple pre-processed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the above-mentioned multiple pre-processed data;

[0148] The second evaluation module 750 is used to determine the evaluation result of the above-mentioned class to be evaluated based on the evaluation results corresponding to the above-mentioned multiple pre-processed data.

[0149] In some possible embodiments, the above-mentioned multiple pre-processed data include classroom teaching data, classroom question-and-answer data, and teacher-student interactive question-and-answer data;

[0150] The pre-processing module 730 includes:

[0151] A data cleaning unit, used for cleaning the text data to obtain the classroom teaching data;

[0152] A first extraction unit, configured to extract the classroom question-and-answer data based on the classroom teaching data;

[0153] The second extraction unit is used to extract the teacher-student interactive question and answer data based on the classroom question and answer data.

[0154] In some possible embodiments, the first extraction unit is specifically used to:

[0155] Based on preset question and answer recognition rules, all question and answer dialogues are extracted from the above classroom teaching data to obtain the above classroom question and answer data.

[0156] In some possible embodiments, the identification module 720 is specifically used to:

[0157] Obtain voiceprint information of teachers and / or students;

[0158] Based on the voiceprint information, speech recognition is performed on the audio data to obtain text data with identity identifiers; the identity identifiers include teacher identifiers and student identifiers.

[0159] In some possible embodiments, the second extraction unit is specifically used to:

[0160] Based on the above identity identifiers in the above classroom question and answer data, question and answer dialogues containing both the teacher identifier and the student identifier are extracted from the above classroom question and answer data to obtain the above teacher-student interactive question and answer data.

[0161] In some possible embodiments, the first evaluation module 740 is specifically used to:

[0162] Inputting the classroom teaching data into the large language model to obtain an evaluation result corresponding to a first evaluation indicator, wherein the first evaluation indicator includes at least one of a curriculum standard evaluation and a curriculum standard improvement;

[0163] Inputting the classroom question and answer data into the large language model to obtain an evaluation result corresponding to a second evaluation indicator, wherein the second evaluation indicator includes at least one of effectiveness analysis, knowledge goal, question avoidance, and question optimization;

[0164] The teacher-student interactive question-and-answer data is input into a large language model to obtain an evaluation result corresponding to a third evaluation indicator, where the third evaluation indicator includes at least one of an interactive evaluation and an interactive suggestion.

[0165] In some possible embodiments, the large language model is pre-trained and / or fine-tuned based on education field data and the evaluation indicators.

[0166] The division of each module in the above-mentioned classroom evaluation device is only for illustration. In other embodiments, the classroom evaluation device can be divided into different modules as needed to complete all or part of the functions of the above-mentioned classroom evaluation device. The implementation of each module in the classroom evaluation device provided in the embodiment of this specification can be in the form of a computer program. The computer program can be run on a terminal or a server. The program module composed of the computer program can be stored in the memory of the terminal or the server. When the computer program is executed by the processor, all or part of the steps of the classroom evaluation method described in the embodiment of this specification are implemented.

[0167] See also Figure 8 , Figure 8 This is a schematic diagram of the structure of an electronic device provided by an exemplary embodiment of this specification. Figure 8 As shown, the electronic device 800 may include: at least one processor 810, at least one communication bus 820, a user interface 830, at least one network interface 840, and a memory 850. The communication bus 820 may be used to realize the connection and communication of the above components.

[0168] The user interface 830 may include a display screen (Display) and a camera (Camera), and the optional user interface may also include a standard wired interface and a wireless interface.

[0169] The network interface 840 may optionally include a Bluetooth module, a Near Field Communication (NFC) module, a Wireless Fidelity (Wi-Fi) module, and the like.

[0170] Among them, the processor 810 may include one or more processing cores. The processor 810 uses various interfaces and lines to connect various parts within the entire electronic device 800, and executes various functions and processes data of the routing electronic device 800 by running or executing instructions, programs, code sets or instruction sets stored in the memory 850, and calling data stored in the memory 850. Optionally, the processor 810 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 810 can integrate one or more combinations of a processor (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 810, but may be implemented separately through a chip.

[0171] The memory 850 may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory 850 includes a non-transitory computer-readable medium. The memory 850 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 850 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as an acquisition function, an evaluation function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 850 may optionally be at least one storage device located away from the aforementioned processor 810. As Figure 8 As shown, the memory 850 as a computer storage medium may include an operating system, a network communication module, a user interface module, and program instructions.

[0172] Specifically, the processor 810 may be used to call the program instructions stored in the memory 850, and specifically perform the following operations:

[0173] Get the audio data of the class to be evaluated;

[0174] Performing speech recognition on the above audio data to obtain text data;

[0175] Preprocessing the text data to obtain a plurality of preprocessed data; the plurality of preprocessed data respectively correspond to different evaluation indicators;

[0176] Inputting the above-mentioned multiple pre-processed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the above-mentioned multiple pre-processed data;

[0177] The evaluation result of the above-mentioned class to be evaluated is determined based on the evaluation results corresponding to each of the above-mentioned multiple pre-processed data.

[0178] In some possible embodiments, the above-mentioned multiple pre-processed data include classroom teaching data, classroom question-and-answer data, and teacher-student interactive question-and-answer data;

[0179] When the processor 810 performs preprocessing on the text data to obtain a variety of preprocessed data, it is specifically used to perform:

[0180] Performing data cleaning on the above text data to obtain the above classroom teaching data;

[0181] Extracting the above classroom question and answer data based on the above classroom teaching data;

[0182] The teacher-student interactive question and answer data is extracted based on the classroom question and answer data.

[0183] In some possible embodiments, when the processor 810 extracts the classroom question and answer data based on the classroom teaching data, it is specifically used to execute:

[0184] Based on preset question and answer recognition rules, all question and answer dialogues are extracted from the above classroom teaching data to obtain the above classroom question and answer data.

[0185] In some possible embodiments, when the processor 810 performs speech recognition on the audio data to obtain text data, it is specifically configured to perform:

[0186] Obtain voiceprint information of teachers and / or students;

[0187] Based on the voiceprint information, speech recognition is performed on the audio data to obtain text data with identity identifiers; the identity identifiers include teacher identifiers and student identifiers.

[0188] In some possible embodiments, when the processor 810 extracts the teacher-student interactive question and answer data based on the classroom question and answer data, it is specifically used to execute:

[0189] Based on the above identity identifiers in the above classroom question and answer data, question and answer dialogues containing both the teacher identifier and the student identifier are extracted from the above classroom question and answer data to obtain the above teacher-student interactive question and answer data.

[0190] In some possible embodiments, when the processor 810 inputs the plurality of preprocessed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the plurality of preprocessed data, the processor 810 is specifically configured to execute:

[0191] Inputting the classroom teaching data into the large language model to obtain an evaluation result corresponding to a first evaluation indicator, wherein the first evaluation indicator includes at least one of a curriculum standard evaluation and a curriculum standard improvement;

[0192] Inputting the classroom question and answer data into the large language model to obtain an evaluation result corresponding to a second evaluation indicator, wherein the second evaluation indicator includes at least one of effectiveness analysis, knowledge goal, question avoidance, and question optimization;

[0193] The teacher-student interactive question-and-answer data is input into a large language model to obtain an evaluation result corresponding to a third evaluation indicator, where the third evaluation indicator includes at least one of an interactive evaluation and an interactive suggestion.

[0194] In some possible embodiments, the large language model is pre-trained and / or fine-tuned based on education field data and the evaluation indicators.

[0195] The present application also provides a computer-readable storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes one or more steps in the above embodiment. If the components of the above classroom evaluation device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium.

[0196] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The above-mentioned computer program product includes one or more computer instructions. When the above-mentioned computer program instructions are loaded and executed on a computer, the above-mentioned process or function according to the embodiment of the present application is generated in whole or in part. The above-mentioned computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The above-mentioned computer instructions can be stored in a computer-readable storage medium or transmitted by the above-mentioned computer-readable storage medium. The above-mentioned computer instructions can be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (such as coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The above-mentioned computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, a data center, etc. that contains one or more available media integrated. The above-mentioned available media can be magnetic media (for example, floppy disks, hard disks, tapes), optical media (for example, digital versatile discs (DVD)), or semiconductor media (for example, solid state disks (SSD)), etc.

[0197] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes. In the absence of conflict, the technical features in this embodiment and the implementation scheme can be combined arbitrarily.

[0198] The embodiments described above are merely preferred embodiments of the present application and are not intended to limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements made to the technical solutions of the present application by ordinary technicians in this field should fall within the protection scope determined by the claims.

[0199] The above describes a specific embodiment of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims and the specification can be performed in an order different from the order in the embodiments recorded in the specification and still achieve the desired results. In addition, the process depicted in the drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A classroom evaluation method, characterized in that: The method comprises: Get the audio data of the class to be evaluated; Performing speech recognition on the audio data to obtain text data; Preprocessing the text data to obtain a plurality of preprocessed data; the plurality of preprocessed data respectively correspond to different evaluation indicators; Inputting the plurality of preprocessed data into a large language model to obtain evaluation results of evaluation indicators corresponding to the plurality of preprocessed data; The evaluation result of the class to be evaluated is determined based on the evaluation results corresponding to each of the multiple preprocessed data.

2. The method according to claim 1, characterized in that: The multiple pre-processed data include classroom teaching data, classroom question-and-answer data, and teacher-student interactive question-and-answer data; The text data is preprocessed to obtain a variety of preprocessed data, including: Performing data cleaning on the text data to obtain the classroom teaching data; Extracting the classroom question and answer data based on the classroom teaching data; The teacher-student interactive question and answer data is extracted based on the classroom question and answer data.

3. The method according to claim 2, characterized in that The extracting the classroom question and answer data based on the classroom teaching data comprises: All question-and-answer dialogues are extracted from the classroom teaching data based on preset question-and-answer recognition rules to obtain the classroom question-and-answer data.

4. The method according to claim 2, characterized in that: The performing speech recognition on the audio data to obtain text data includes: Obtain voiceprint information of teachers and / or students; The audio data is subjected to speech recognition based on the voiceprint information to obtain text data with identity identifiers; the identity identifiers include a teacher identifier and a student identifier.

5. The method according to claim 4, characterized in that The extracting the teacher-student interactive question-and-answer data based on the classroom question-and-answer data comprises: Based on the identity identifier in the classroom question and answer data, the question and answer dialogue containing both the teacher identifier and the student identifier is extracted from the classroom question and answer data to obtain the teacher-student interactive question and answer data.

6. The method according to claim 2, characterized in that The inputting the plurality of pre-processed data into the large language model to obtain evaluation results of the evaluation indicators corresponding to the plurality of pre-processed data respectively includes: Inputting the classroom teaching data into a large language model to obtain an evaluation result corresponding to a first evaluation indicator, wherein the first evaluation indicator includes at least one of a curriculum standard evaluation and a curriculum standard improvement; Inputting the classroom question and answer data into a large language model to obtain an evaluation result corresponding to a second evaluation indicator, wherein the second evaluation indicator includes at least one of effectiveness analysis, knowledge objectives, question avoidance, and question optimization; The teacher-student interactive question-and-answer data is input into a large language model to obtain an evaluation result corresponding to a third evaluation indicator, where the third evaluation indicator includes at least one of an interactive evaluation and an interactive suggestion.

7. The method according to claim 1 or 6, characterized in that: The large language model is obtained by pre-training and / or fine-tuning based on education field data and the evaluation index.

8. A classroom evaluation device, characterized in that: The device comprises: The acquisition module is used to obtain the audio data of the class to be evaluated; A recognition module, used for performing speech recognition on the audio data to obtain text data; A preprocessing module, used for preprocessing the text data to obtain a plurality of preprocessed data; the plurality of preprocessed data respectively correspond to different evaluation indicators; A first evaluation module, used for inputting the plurality of pre-processed data into a large language model to obtain evaluation results of evaluation indicators corresponding to the plurality of pre-processed data; The second evaluation module is used to determine the evaluation result of the class to be evaluated based on the evaluation results corresponding to the multiple pre-processed data.

9. A computer storage medium, characterized in that: The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 7.

Images