KR20200088073A - English speaking evaluation system and method thereof - Google Patents

Abstract

According to an embodiment of the present invention, a system for evaluating English speaking comprises: a user terminal including one or more function modules; a service server receiving and processing information received from the one or more function modules; and a network providing communication between the user terminal and the service server. The service server includes a communication unit providing communication with the user terminal, a storage unit storing the received information and information required for evaluation, and a control unit receiving a voice from the user terminal to generate a text and analyzing a text phrase to perform evaluation based on an analysis result.

Description

ENGLISH SPEAKING EVALUATION SYSTEM AND METHOD THEREOF}

The present invention relates to an English speaking evaluation system and method.

The method of evaluating English speaking can be largely divided into human evaluation and machine evaluation. Evaluation by a person can evaluate the intention of the evaluator and the qualitative part of the sentence, depending on the evaluator's experience and background knowledge, but sometimes it may have unobjective results depending on the evaluator's subject. In particular, the evaluation by a person may be different depending on who evaluates the evaluator and the evaluation time and environment, even if the evaluator is the same evaluator.

Machine-based evaluation can be performed consistently and quickly, without time and environmental constraints, but criteria for more reliable evaluation are needed.

The above description is only to assist in understanding the background of the technical ideas of the present invention, and therefore it cannot be understood as the content corresponding to the prior art known to those skilled in the art of the present invention.

An object of the present invention is to provide a system and method for evaluating English speaking that allows users to evaluate their English speaking skills with high reliability in a short time without limitation of time and environment.

In order to achieve the above object, an English speaking evaluation system according to an embodiment of the present invention includes: a user terminal including one or more function modules, and a service server receiving and processing information received from the one or more function modules; Including a network for providing communication between the user terminal and the service server, the service server includes a communication unit for providing communication with the user terminal, and a storage unit for storing the input information and information necessary for evaluation. And a control unit that receives a voice from the user terminal to generate text, analyzes the generated text syntax, and evaluates the result based on the analyzed result.

The one or more functional modules may include at least one of a display module, an audio module, an input module, a sensor module, a camera module, and a communication module.

The user terminal is a VR (Virtual Reality) device, AR (Augmented Reality) device, mobile phone, PMP (Portable Multimedia Player), MID (Mobile Internet Device), smart phone (Smart Phone), desktop (Desktop), tablet computer (Tablet) PC), a notebook (Note book), a netbook (Net book), a server (Server) and a mobile (Mobile) terminal.

Information received from the one or more function modules may include user identification information and user response voice information.

The control unit includes an artificial intelligence speech recognition module for generating text by receiving speech from the user terminal, a natural language analysis deep learning module for analyzing text phrases generated in the artificial intelligence speech recognition module, and the natural language analysis deep learning module It may include an evaluation module for evaluating based on the results analyzed in the, and a rating calculation module for calculating a user rating based on the results evaluated in the evaluation module.

The natural language analysis deep learning module may perform voice analysis for receiving voice from the user terminal and analyzing at least one of volume, response time, length, persistence, and pronunciation accuracy.

The natural language analysis deep learning module may perform sentence analysis that analyzes at least one of parts of speech, sentence structure, subject, tense, and vocabulary targeting text phrases generated by the artificial intelligence speech recognition module.

The natural language analysis deep learning module compares at least one of sentence originality, expression accuracy, and euphemism expression by comparing the text phrase generated by the AI speech recognition module with another user's answer previously stored. Analysis can be performed.

The evaluation module, based on the results analyzed in the deep learning module of the natural language analysis, the user's response volume, response speed, response persistence, modifier detail, access word diversity, correspondence, topic accuracy, expression accuracy, sentence originality, meaning It can evaluate at least one of transmittance, answer length, tense match, sentence type diversity, vocabulary, vocabulary overlap, problem comprehension, topic solving, sudden response, role performance, and question performance.

The English speaking evaluation system may further include a database that stores information for user authentication and answer information of other users previously stored for each problem.

An English speaking evaluation method according to an embodiment of the present invention comprises the steps of: receiving a user response voice from a user terminal; Evaluating each problem based on the user's answer voice; Comprehensive evaluation based on the problem attributes and the analysis results for each problem; Calculating a user rating based on the problem-specific evaluation and the comprehensive evaluation; And providing the user's evaluation result and the user rating to the user terminal.

The evaluation of each problem based on the user response voice may include: analyzing the user response voice; Generating text by applying artificial intelligence speech recognition technology to the user-answered voice; And parsing the generated text through text through a natural language analysis deep learning model.

The analyzing the user response voice may include analyzing at least one of the volume, response time, length, persistence, and pronunciation accuracy of the user response.

The step of parsing the generated text through the natural language analysis deep learning model may include analyzing at least one of parts of speech, sentence structure, subject, tense, and vocabulary of the generated text.

The evaluation of each problem based on the user response voice may include: analyzing the user response voice; Generating text by applying artificial intelligence speech recognition technology to the user-answered voice; And analyzing the characteristics of the expression by comparing the generated text with answers of other users stored in advance.

Analyzing the characteristics of the expression by comparing the generated text with the answers of other pre-stored users, the sentence originality, expression accuracy, and euphemism expression by comparing the generated text with the answers of other users previously stored It may include the step of analyzing at least one of.

The step of evaluating each problem based on the voice of the user's answer includes the amount of response of the user, the speed of the response, the persistence of the answer, the detail of the modifier, the variety of access words, the ability to respond, the accuracy of the subject, the accuracy of expression, based on the voice of the user's answer And evaluating at least one of sentence originality, meaning conveyance, response length, tense match, sentence type diversity, vocabulary, and vocabulary redundancy.

Comprehensive evaluation based on the problem attribute and the analysis result for each problem may include at least one of the complexity of the problem, the difficulty of the topic, the difficulty of the speed, whether there is a breakthrough, whether to play the role, whether to ask me, and the possible tense Attributes, volume analysis by question, response time analysis, length analysis, tense analysis, sentence structure analysis, part-of-speech analysis, subject analysis, persistence analysis, expression accuracy analysis, euphemism expression analysis, vocabulary analysis, sentence originality analysis, and pronunciation accuracy It may include at least one of the analysis.

Comprehensive evaluation based on the problem attribute and the analysis result for each problem may include at least one of the user's problem understanding, topic solving ability, sudden response ability, role performance, and question performance based on the problem attribute and problem analysis result. And evaluating.

The step of providing the user's evaluation result and the user rating to the user terminal may include providing an average result value of all users corresponding to the user rating to the user terminal.

According to the English speaking evaluation system and method according to an embodiment of the present invention, by evaluating English speaking skills using artificial intelligence and deep learning, the user can quickly and highly reliable English speaking skills without limitation of time and environment. Can be evaluated. In addition, when a user utilizes a VR/AR device, by allowing the user to experience the environment of taking a test in an English speaking test site, the user can achieve an effect of obtaining better results in an actual English speaking test.

1 is a configuration diagram schematically showing an English speaking evaluation system according to an embodiment of the present invention.
2 is a diagram schematically showing the configuration of the user terminal of FIG. 1 according to an embodiment of the present invention.
3 is a diagram schematically showing the configuration of the service server of FIG. 1 according to an embodiment of the present invention.
4 is a flowchart illustrating an English speaking evaluation method according to an embodiment of the present invention.
5 is a flowchart illustrating an evaluation method for each problem in FIG. 4 according to an embodiment of the present invention.
6 is an example of providing an evaluation result and a rating to a user according to an embodiment of the present invention.

In the following description, for the purpose of explanation, many specific details are presented to aid understanding of various embodiments. However, it is apparent that various embodiments may be practiced without these specific details or in one or more equivalent ways. In other instances, well-known structures and devices are shown in block diagram form in order to avoid unnecessarily making various embodiments difficult to understand.

In the drawings, the size or relative size of layers, films, panels, regions, etc. may be exaggerated for clarity. Also, the same reference numerals denote the same components.

Throughout the specification, when a part is "connected" to another part, this includes not only "directly connected" but also "indirectly connected" with another element in between. . However, if a part is described as being "directly connected" to another part, it will mean that there is no other element between the part and the other part. "At least one of X, Y, and Z", and "at least one selected from the group consisting of X, Y, and Z" are one of X, one of Y, one of Z, or two of X, Y, and Z, or Any further combination (eg, XYZ, XYY, YZ, ZZ) will be understood. Here, “and/or” includes all combinations of one or more of the configurations.

Here, terms such as first, second, etc. may be used to describe various elements, elements, regions, layers, and/or sections, but such elements, elements, regions, layers, and/or Or sections are not limited to these terms. These terms are used to distinguish one element, element, region, layer, and/or section from another element, element, region, layer, or section. Thus, the first element, element, region, layer, and/or section in one embodiment may be referred to as the second element, element, region, layer, and/or section in another embodiment.

Spatially relative terms such as "below", "above", etc. may be used for the purpose of description, thereby describing the relationship of one element or feature to another element(s) or feature(s) as shown in the figure do. This is only used to show the relationship of one component to another component in the drawing, and does not mean an absolute position. For example, when the device shown in the figure is turned over, elements depicted as being “below” other elements or features are positioned in a direction “above” the other elements or features. Thus, in one embodiment, the term "below" can include both the top and bottom. In addition, the device may be in other directions (eg, rotated 90 degrees or in other directions), and such spatially relative terms used herein are interpreted accordingly.

The terminology used herein is for the purpose of describing specific embodiments and not for limitation. Throughout the specification, when a part “includes” a certain component, this means that other components may be further included rather than excluding other components unless specifically stated to the contrary. Unless otherwise defined, terms used herein have the same meaning as commonly understood by a person skilled in the art to which the present invention pertains.

1 is a configuration diagram schematically showing an English speaking evaluation system according to an embodiment of the present invention.

1, the English speaking evaluation system includes a user terminal 10, a network 20 and a service server 30, and the user terminal 10 and the service server 30 are connected through a network 20 It transmits and receives all data and information necessary for English speaking evaluation. As an embodiment, the English speaking evaluation system may include a database 40 that stores information for user authentication and answer information of other users pre-stored for each problem.

The user terminal 10 requests the user authentication to the service server 30 through the network 20 when the user starts the English speaking evaluation by running the English speaking evaluation program or application installed on the user terminal 10. When the user terminal 10 starts to speak in English, the user input including the answer according to the user's survey is transmitted to the service server 30, and all data and information required for evaluating the English speaking are provided to the service server 30. It receives and shows it to the user.

The user terminal 10 is described as a typical example of a mobile communication terminal that is connected to the network 20 and can transmit and receive data, but the terminal is not limited to a mobile communication terminal, all information communication devices, multimedia terminals, wired/wireless terminals , Fixed terminals and various terminals such as IP (Internet Protocol) terminals. In addition, the terminal is a VR (Virtual Reality) device, AR (Augmented Reality) device, mobile phone, PMP (Portable Multimedia Player), MID (Mobile Internet Device), smart phone (Smart Phone), desktop (Desktop), tablet computer (Tablet) PC), a notebook (Note book), a net book (Net book), a server (Server) and a mobile terminal having various mobile communication specifications such as an information communication device.

Various types of communication networks may be used as the network 20, for example, WLAN, Wireless LAN, Bluetooth, Wi-Fi, Wibro, Wimax, and high-speed downlink packet access ( Wireless communication methods such as HSDPA, High Speed Downlink Packet Access, or Ethernet, xDSL (ADSL, VDSL), HFC (Hybrid Fiber Coax), FTTC (Fiber to The Curb), FTTH (Fiber To The Home), etc. A wired communication method can be used. Meanwhile, the network 20 is not limited to the above-described communication method, and may include all other well-known or future communication methods in addition to the above-described communication method.

The service server 30 transmits and receives all data and information necessary for the English speaking evaluation through the user terminal 10 and the network 20 to provide an English speaking evaluation service.

In one embodiment of the present invention, the service server 30 receives a user authentication request for starting an English speaking evaluation from the user terminal 10 and performs a user login process. As an embodiment, the service server 30 may provide a launching module required for English speaking evaluation. When the English speaking evaluation starts, the service server 30 receives a user input including an answer according to the user's survey from the user terminal 10 and transmits data and information necessary for the English speaking evaluation of the user.

As an embodiment, the user terminal 10 may include one or more function modules, and the service server 30 may receive and process information received from one or more function modules of the user terminal 10. As an embodiment, the information received from one or more functional modules may include user identification information and user response voice information.

The service server 30 may include a server operated by an English speaking evaluation service provider. As an embodiment, it may include a server operated by an English speaking evaluation service provider such as Pagoda, Hackers, Byungcheol Min, YBM, Siwon School.

The database 40 stores all information necessary for English speaking evaluation, including information for user authentication. As an embodiment, the database 40 may store information for user authentication and answer information of other users previously stored for each problem.

2 is a diagram schematically showing the configuration of the user terminal of FIG. 1 according to an embodiment of the present invention.

Referring to FIG. 2, the user terminal 10 includes one or more functional modules. As an embodiment, the user terminal 10 may include a display module 110, an audio module 120, an input module 130, a sensor module 140, a camera module 150, and a communication module 160. have.

The display module 110 can include any suitable screen or projection system to provide a visible indication to the user. As an embodiment, the display module 110 may include a head mounted display (HMD) included in a VR device.

The audio module 120 may include any suitable audio component for providing audio to the user. As an embodiment, the audio module 120 may include a speaker and/or a microphone.

The input module 130 may be a user interface that gives a user input or command. As an embodiment, the input module 130 may be implemented as a button, keypad, dial, click wheel, touch pad or touch screen.

The sensor module 140 may include a sensor that senses information such as a user's location and environment. As an embodiment, the sensor module 140 may include a magnetic sensor that detects a change in the geomagnetic field.

The camera module 150 may include a device capable of capturing or capturing still images and videos. As an embodiment, the camera module 150 may include at least one of an image sensor, a lens, an image signal processor (ISP), and a flash.

The communication module 160 refers to a module for communicating with the service server 30 through the network 20 and may be embedded in the user terminal 10. As an embodiment, the communication module 160 may include at least one of a mobile communication module, a wireless Internet module, and a short-range communication module.

3 is a diagram schematically showing the configuration of the service server of FIG. 1 according to an embodiment of the present invention.

Referring to FIG. 3, the service server 30 includes a communication unit 310, a storage unit 320, and a control unit 330.

The communication unit 310 exchanges data with the user terminal 10 through the network 20.

The storage unit 320 stores data and programs for the operation of the service server 30. As an embodiment, the storage unit 320 may serve as a database 40 by storing information input from the user terminal 10 and information necessary for English speaking evaluation. That is, the storage unit 320 may store information for user authentication and answer information of other users previously stored for each problem.

The control unit 330 receives the voice from the user terminal 10 to generate text, and analyzes the generated text syntax to evaluate the user's English speaking ability based on the analyzed result. As an embodiment, the control unit 330 may include an artificial intelligence speech recognition module 331, a natural language analysis deep learning module 332, an evaluation module 333, and a grade calculation module 334.

The artificial intelligence speech recognition module 331 receives a voice from the user terminal 10 and generates text by applying artificial intelligence speech recognition technology (STT: Speech-To-Text).

The natural language analysis deep learning module 332 analyzes the text phrase generated in the artificial intelligence speech recognition module 331 through a natural language analysis deep learning model. As an embodiment, the natural language analysis deep learning module 332 may receive a voice from the user terminal 10 and perform voice analysis to analyze at least one of volume, response time, length, persistence, and pronunciation accuracy. As an embodiment, the natural language analysis deep learning module 332 performs sentence analysis that analyzes at least one of parts of speech, sentence structure, subject, tense, and vocabulary targeting text phrases generated by the artificial intelligence speech recognition module 331 can do. As an embodiment, the natural language analysis deep learning module 332 is in the sentence originality, expression accuracy, and euphemism expression by comparing the text phrase generated by the artificial intelligence speech recognition module 331 with the answers of other users previously stored. Comparative analysis may be performed to analyze at least one or more.

The evaluation module 333 evaluates the user's English speaking ability based on the results analyzed by the natural language analysis deep learning module 332. As an embodiment, the evaluation module 333 based on the results analyzed in the natural language analysis deep learning module 332, the user's response volume, response speed, response persistence, modifier detail, access word diversity, correspondence, subject accuracy, At least one of expression accuracy, sentence originality, meaning transferability, response length, tense matching, sentence type diversity, vocabulary, vocabulary redundancy, problem understanding, topic solving ability, sudden response, role performance, and question performance can be evaluated.

The rating calculation module 334 calculates the user rating based on the results evaluated by the rating module 333.

4 is a flowchart illustrating an English speaking evaluation method according to an embodiment of the present invention.

For the English speaking evaluation method according to the present invention, the English speaking evaluation program or application installed in the user terminal is executed by the user. When the English speaking evaluation program or application is executed in the user terminal, the service server may provide a problem for the English speaking evaluation to the user terminal. As an embodiment, the problem may include at least one of the subject matter of the item selected in the survey, the unexpected subject matter, and the role play type problem.

For the method of evaluating English speaking according to the present invention, provision of a problem for evaluating English speaking may include a question banking method. As an embodiment, the question banking method includes question bank registration for pre-registration of types of problems in the database, problem prefix registration for pre-registration of phrases that can be added to the front of the problem, and back of problems. Problems to pre-register phrases that can be added to the situation separately Problems including postfix registration Bank construction stages, and question/selection/progress/role-play type selection of the problem type, topics such as movies, music Selecting a problem Topic selection, problem prefix, problem body, and problem postfix are combined to create a problem type, a problem that automatically composes a sentence according to the problem topic, sentence construction, and outputs at least one of images, images, and sounds to the user terminal. It may include a question submission step that includes an output.

Referring to FIG. 4, the service server receives a user response voice from the user terminal (S10), and evaluates the user response by problem based on the user response voice (S20). As an embodiment, in order to evaluate the user response by problem, the service server provides the user's response volume, response speed, response persistence, modifier detail, access word diversity, correspondence, subject accuracy, expression accuracy, based on the user response voice It is possible to evaluate at least one of sentence originality, meaning conveyance, answer length, tense match, sentence type diversity, vocabulary, and vocabulary overlap.

The service server comprehensively evaluates the user's answers based on the problem attributes and the analysis results for each problem (S30), calculates the user rating based on the evaluation for each problem and the comprehensive evaluation (S40), and provides the evaluation results and rating to the user (S50). As an embodiment, in order to comprehensively evaluate a user response, the service server may include a problem attribute including at least one of a complexity of a problem, a difficulty of a topic, a difficulty of a speed, a suddenness, whether to play a role, whether to ask me, and a possible tense, Analysis of the amount of user response by problem, analysis of response time, length analysis, tense analysis, sentence structure analysis, part-of-speech analysis, topic analysis, persistence analysis, expression accuracy analysis, euphemism expression analysis, vocabulary analysis, sentence originality analysis, and pronunciation accuracy analysis Based on the analysis results for each problem including at least one of the results, the user's answer may be comprehensively evaluated. As an embodiment, in order to comprehensively evaluate user answers, the service server may evaluate at least one of a user's problem understanding, topic solving ability, unexpected response ability, role performance ability, and question performance ability based on problem attributes and analysis results for each problem. .

According to one embodiment of the present invention, in order to calculate the user rating, the service server is an evaluation factor for evaluation by problem, such as response volume, response speed, response persistence, modifier detail, access word diversity, correspondence, subject accuracy, expression accuracy, Scores can be calculated by applying weights to each of sentence originality, meaning conveyance, answer length, tense match, sentence type diversity, vocabulary, and vocabulary overlap. As an embodiment, in order to provide the user's evaluation result and user rating, the service server may provide the user terminal with an average result value of all users corresponding to the user rating together with the user evaluation result value. As an example,

5 is a flowchart illustrating an evaluation method for each problem in FIG. 4 according to an embodiment of the present invention.

Referring to FIG. 5, the service server analyzes the user's answer voice to evaluate the user's answer by problem (S210), generates text by applying artificial intelligence speech recognition technology to the user's answer voice (S220), and generates Parsed text through natural language analysis deep learning model (S230). As an embodiment, the service server may analyze at least one of a user response volume, a response time, a length, a duration, and a pronunciation accuracy for voice analysis. As an embodiment, the service server may analyze at least one of parts of speech, sentence structure, subject, tense, and vocabulary of the generated text in order to analyze text syntax through a deep learning model.

Referring to FIG. 5, the service server analyzes the user's answer voice to evaluate the user's answer by problem (S210), generates text by applying artificial intelligence speech recognition technology to the user's answer voice (S220), and generates The characteristic of the expression is analyzed by comparing the old text with the answers of other users stored in advance (S240). As an embodiment, in order to analyze the characteristics of the user expression through comparison, at least one of sentence originality, expression accuracy, and euphemism expression may be analyzed through comparison of the generated text with the answers of other pre-stored users.

6 is an example of providing an evaluation result and a rating to a user according to an embodiment of the present invention.

Referring to FIG. 6, the user can check the average score of the IH grade, which is the grade to which the user belongs, along with his evaluation result. In addition, the user can check the detailed evaluation result for each evaluation criterion composed of four axes, so that the user can clearly know the part to be supplemented to improve his/her English speaking ability.

According to the embodiments of the present invention as described above, the English speaking evaluation system and method according to an embodiment of the present invention evaluates the English speaking skills using artificial intelligence and deep learning, so that the user is limited in time and environment Without it, it allows you to evaluate your English speaking skills with high confidence in a short time. In addition, when a user utilizes a VR/AR device, by allowing the user to experience the environment of taking a test in an English speaking test site, the user can achieve an effect of obtaining better results in an actual English speaking test.

As described above, the present invention has been described by specific matters such as specific components, etc., and by limited embodiments and drawings, which are provided to help the overall understanding of the present invention, but the present invention is not limited to the above embodiments , Anyone having ordinary knowledge in the field to which the present invention pertains can make various modifications and variations from these descriptions.

Therefore, the spirit of the present invention should not be limited to the described embodiments, and should not be determined, and all claims that are equivalent or equivalent to the scope of the claims as well as the claims described below will be included in the scope of the spirit of the invention. .

10: user terminal 20: network
30: service server 40: database

Claims (20)

A user terminal including one or more function modules;
A service server that receives and processes information received from the one or more function modules; And
Includes a network that provides communication between the user terminal and the service server,
The service server,
The communication unit for providing communication with the user terminal, the storage unit for storing the input information and necessary information for evaluation, and receiving the voice from the user terminal to generate text and analyze the generated text syntax English speaking evaluation system including a control unit to evaluate based on the analyzed results. According to claim 1,
The one or more functional modules are English speaking evaluation system including at least one of a display module, an audio module, an input module, a sensor module, a camera module, and a communication module. According to claim 1,
The user terminal is a VR (Virtual Reality) device, AR (Augmented Reality) device, mobile phone, PMP (Portable Multimedia Player), MID (Mobile Internet Device), smart phone (Smart Phone), desktop (Desktop), tablet computer (Tablet) PC), a notebook (Note book), a net book (Net book), a server (Server) and a mobile (Mobile) terminal including at least one or more of the English speaking evaluation system. According to claim 1,
The information received from the one or more function modules is an English speaking evaluation system including user identification information and user response voice information. According to claim 1,
The control unit,
The artificial intelligence speech recognition module that generates text by receiving speech from the user terminal, the natural language analysis deep learning module that analyzes text phrases generated by the artificial intelligence speech recognition module, and the results analyzed by the natural language analysis deep learning module English evaluation system comprising an evaluation module for evaluating, and a rating calculation module for calculating a user rating based on the results evaluated in the evaluation module. The method of claim 5,
The natural language analysis deep learning module is an English speaking evaluation system that receives a voice from the user terminal and performs voice analysis to analyze at least one of volume, response time, length, persistence, and pronunciation accuracy. The method of claim 5,
The natural language analysis deep learning module is an English speaking evaluation system that performs sentence analysis that analyzes at least one of parts of speech, sentence structure, subject, tense, and vocabulary targeting text phrases generated by the artificial intelligence speech recognition module. The method of claim 5,
The natural language analysis deep learning module compares at least one of sentence originality, expression accuracy, and euphemism expression by comparing the text phrase generated by the AI speech recognition module with another user's answer previously stored. English speaking assessment system to perform analysis. The method of claim 5,
The evaluation module, based on the results analyzed in the deep learning module of the natural language analysis, the user's response volume, response speed, response persistence, modifier detail, access word diversity, correspondence, topic accuracy, expression accuracy, sentence originality, meaning An English speaking assessment system that evaluates at least one of communicativeness, answer length, tense matching, sentence type diversity, vocabulary, vocabulary redundancy, problem comprehension, topic solving, sudden response, role performance, and question performance. The method of claim 1, wherein the English speaking evaluation system,
English speaking evaluation system further comprising a database for storing information for user authentication and answer information of other users pre-stored for each problem. Receiving a user answer voice from the user terminal;
Evaluating each problem based on the user's answer voice;
Comprehensive evaluation based on problem attributes and analysis results by problems;
Calculating a user rating based on the problem-specific evaluation and the comprehensive evaluation; And
And providing an evaluation result of the user and the user rating to a user terminal. The method of claim 11,
Evaluating each problem based on the voice of the user response,
Analyzing the user answer voice;
Generating text by applying artificial intelligence speech recognition technology to the user-answered voice; And
And parsing the generated text through text through a natural language analysis deep learning model. The method of claim 12,
The step of analyzing the user response voice,
And analyzing at least one of the volume, response time, length, persistence, and pronunciation accuracy of the user answer. The method of claim 12,
The step of parsing the generated text through the natural language analysis deep learning model,
And analyzing at least one of parts of speech, sentence structure, subject, tense, and vocabulary of the generated text. The method of claim 11,
Evaluating each problem based on the voice of the user response,
Analyzing the user answer voice;
Generating text by applying artificial intelligence speech recognition technology to the user-answered voice; And
And analyzing the characteristics of expression by comparing the generated text with answers of other users stored in advance. The method of claim 15,
The step of analyzing the characteristics of the expression by comparing the generated text with the answers of other users stored in advance,
And analyzing at least one of sentence originality, expression accuracy, and euphemism by comparing the generated text with answers of other users stored in advance. The method of claim 11,
Evaluating each problem based on the voice of the user response,
Based on the user's response voice, the user's response volume, response speed, response persistence, modifier detail, access word diversity, correspondence, topic accuracy, expression accuracy, sentence originality, meaning transmission, response length, tense match, sentence A method of evaluating English speaking comprising evaluating at least one of formal diversity, vocabulary, and vocabulary overlap. The method of claim 11,
Comprehensive evaluation based on the problem attributes and the analysis results for each problem,
Problem attributes including at least one of the complexity of the problem, Topic difficulty, Speed difficulty, whether there is a breakout, whether to play, whether to ask Me, and possible tenses, and analysis of the amount of each problem, response time analysis, length analysis, and tense Comprehensive evaluation based on analysis results by problem, including at least one of analysis, sentence structure analysis, part-of-speech analysis, subject analysis, persistence analysis, expression accuracy analysis, euphemism expression analysis, vocabulary analysis, sentence originality analysis, and pronunciation accuracy analysis English speaking assessment method comprising the step of. The method of claim 11,
Comprehensive evaluation based on the problem attributes and the analysis results for each problem,
Evaluating at least one of the user's problem understanding, Topic solving ability, sudden response ability, role performance, and question performance based on the problem attribute and the analysis result for each problem. The method of claim 11,
Providing the user's evaluation result and the user rating to the user terminal,
And providing an average result value of all users corresponding to the user grade to the user terminal.

Images