JP6419924B1 - Learning support system and learning support method - Google Patents

Abstract

[PROBLEMS] To present an optimal teacher suitable for a student. In a learning support system that supports learning by a computer, a storage unit (student DB 611) for storing student attribute information, a determination unit (teaching processing unit 605) for determining the progress of each student's learning, and When registering a new student, a search unit (teaching processing unit 605) that searches the storage unit for a student having attribute information similar to the new student, and an attribute similar to the new student acquired by the search unit Identification means (teaching processing unit 605) for identifying a teacher determined to have a relatively high evaluation among teachers in charge of students having information, and presentation means (object) for presenting the teacher identified by the identification means And a processing unit 608). [Selection] Figure 4

Description

  The present invention relates to a learning support system and a learning support method.

  Patent Document 1 discloses a student terminal having a question transmitting means for issuing a question to a teacher terminal and sending an e-mail via a LAN, and an answer receiving means for receiving an answer from the teacher terminal; A technique is disclosed that includes a teacher terminal having a question receiving means for receiving a question from a terminal and an answer transmitting means for sending an answer to the student terminal via a LAN.

  According to such a technique, it is possible to realize a learning support system capable of dialogue between a teacher and students in CAI on a LAN.

JP 09-185598 A

  However, the technique disclosed in Patent Document 1 has a problem that learning does not progress as expected when the compatibility between the teacher and the student is not good.

  An object of the present invention is to provide a learning support system and a learning support method capable of presenting an optimal teacher suitable for a student.

In order to solve the above-described problems, the present invention provides a learning support system that supports learning by a computer, and includes surface attribute information indicating measurable surface attributes of a student and deep attributes other than the surface attribute information. Presenting a virtual teacher having first storage means for storing deep attribute information to be displayed, an avatar drawn by the computer, and a voice synthesized by the computer, and displaying the presented virtual teacher Teaching means for teaching the student the learning content, second storage means for storing attribute information indicating the type of the avatar of the virtual teacher, the characteristics of the synthesized speech, and the teaching method ; The determination means for determining the progress of learning of each student, and when registering a new student, the surface attribute information of the new student, The attribute information similar to the new student acquired by the search means, the search means for searching the student having the attribute information similar to the first storage means with reference to layer attribute information Identifying means for identifying, from the second storage means, the attribute information related to the virtual teacher determined to have a relatively high evaluation among the virtual teachers presented to the students; based on said attribute information specified by the specifying means, and a control means for performing control to be presented to the teaching means the virtual teacher for the new the student, the teaching according to the teaching means has changing means for modifying to identify the effect is high the teaching method of the method, the teaching means, among the teaching methods that have been modified by the modification means are learning effect is higher For, also used for other of the student, and wherein the.
According to such a configuration, it is possible to present an optimal teacher suitable for the student.

In the learning support method for supporting learning by a computer, the present invention provides surface attribute information indicating a measurable surface attribute of a student and deep attribute information indicating a deep attribute other than the surface attribute information. A virtual teacher having a first storage step of storing in the storage means, an avatar drawn by the computer, and a voice synthesized by the computer is presented, and learning is performed via the presented virtual teacher. A teaching step of teaching content to the student, a second storing step of storing attribute information indicating the type of the avatar of the virtual teacher, the characteristics of the synthesized speech, and a teaching method in the storage means; A determination step of determining the progress of learning of each of the students, and when registering a new student, the surface layer of the new student A search step of searching the student having the attribute information similar to the storage means with reference to sex information and the deep attribute information, and the similar to the new student acquired in the search step A specifying step of specifying, from the storage means, the attribute information related to the virtual teacher determined to have a relatively high evaluation among the virtual teachers presented to the students having attribute information; , on the basis of the said attribute information specified in the specifying step, and a control step of performing control to present the virtual teacher to the teaching steps for the new the student, the by the teaching step to identify the effect is high the teaching method of teaching methods have altered step of modifying the teaching methods, the teaching step, the modified scan Of the modified the teaching methods in-up, for those learning effect is high, also be used for other of the student, and wherein the.
According to such a method, it is possible to present an optimal teacher suitable for the student.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the learning assistance system and the learning assistance method which can show the optimal teacher suitable for the student.

It is a figure which shows the structural example of the learning assistance system which concerns on embodiment of this invention. It is a figure which shows the detailed structural example of the client apparatus shown in FIG. It is a figure which shows the detailed structural example of the server apparatus shown in FIG. 4 is a functional block realized by cooperation of hardware and software of the server apparatus shown in FIG. 3. It is a figure which shows an example of the attribute information of a student. It is a figure which shows the detail of the attribute information of the student shown in FIG. It is a figure which shows an example of the attribute information of a teacher. It is a figure which shows the detail of the attribute information of the teacher shown in FIG. It is a flowchart which shows an example of the process performed when registering a new student. It is a figure which shows an example of a teacher's avatar. It is a flowchart which shows an example of the process performed when teaching from a teacher to a student. It is a flowchart which shows the detail of the modification process of the teacher of FIG. 11, and a method. It is a flowchart which shows an example of the process performed when performing coaching from a teacher to a student.

  Next, an embodiment of the present invention will be described.

(A) Description of Configuration of Embodiment of the Present Invention FIG. 1 is a diagram illustrating a configuration example of a learning support system according to an embodiment of the present invention. As shown in FIG. 1, the learning support system 1 includes a position information input device 10, an HMD (Head Mount Display) 20, a voice input / output device 30, a client device 40, and a server device 60. The server device 60 is connected through a network 50 in a communicable state.

  The position information input device 10 is configured as, for example, a wristband type device that a student wears on a wrist or the like. For example, the position information input device 10 detects wrist movement by an acceleration sensor, acquires information drawn on a desk or space by a student, and supplies the information to the client device 40. As the position information input device 10, for example, a joystick, a pointing stick, a touch pad, a data glove, a trackball, a pen tablet, a mouse, a light pen, and a joypad can be used.

  The HMD 20 is a display that a student wears on the head and displays the video supplied from the client device 40 as different images for both eyes. Further, the HMD 20 has an acceleration sensor, and functions as a so-called virtual reality (VR) display by detecting the direction of the head and supplying an image corresponding to the detection result from the client device 40. Note that the HMD 20 is a “non-transparent” type that cannot completely cover the eyes and see the outside. Alternatively, the HMD 20 can be seen from the outside with a half-mirror display, or a small video that captures the outside. It may be a “transmission type” that allows a camera image to be seen. Moreover, an image sensor may be provided in the client device 40, and the direction of the HMD 20 may be detected by the image sensor.

  The audio input / output device 30 is a device for inputting and outputting audio, and is configured by, for example, a microphone and a speaker or headphones. The voice input / output device 30 converts the voice uttered by the student into an electrical signal and supplies it to the client device 40. The voice input / output device 30 converts the electrical signal supplied from the client device 40 into voice and outputs the voice. Note that three-dimensional sound may be realized by changing the volume and phase of the sound output from the headphones between the right ear and the left ear in accordance with the head movement detected by the HMD 20.

  The network 50 is configured by, for example, the Internet or a LAN (Local Area Network), and can exchange information between the client device 40 and the server device 60 by exchanging data in a packet format.

  The server device 60 is configured by a computer such as a workstation, for example, and supplies image, sound, and text data in response to a request from the client device 40 and processes information supplied from the client device 40.

  FIG. 2 is a diagram illustrating a detailed configuration example of the client device 40 illustrated in FIG. 1. As shown in FIG. 2, the client device 40 includes a CPU (Central Processing Unit) 41, a ROM (Read Only Memory) 42, a RAM (Read Access Memory) 43, an HDD (Hard Disk Drive) 44, and a GPU (Graphics Processing Unit). 45, an I / F (Interface) 46, and a bus 47.

  Here, the CPU 41 executes arithmetic processing based on programs stored in the ROM 42, the RAM 43, and the HDD 44, and controls each unit of the apparatus.

  The ROM 42 is a semiconductor storage device that stores basic programs and data executed by the CPU 41. The RAM 43 is a semiconductor memory device that temporarily stores a program executed by the CPU 41 and temporarily stores data generated when the program is executed.

  The HDD 44 is a magnetic storage device that stores basic programs executed by the CPU 41 and data.

  The GPU 45 executes a drawing process based on the drawing data supplied from the CPU 41, and supplies the obtained image to the HMD 20 for display.

  The I / F 46 inputs data supplied from the position information input device 10, the HMD 20, and the voice input / output device 30 and supplies voice data to the voice input / output device 30. Further, the I / F 46 transmits and receives data to and from the server device 60 via the network 50.

  The bus 47 is a signal line group for electrically connecting the CPU 41, ROM 42, RAM 43, HDD 44, GPU 45, and I / F 46 to each other and enabling data exchange between them.

  FIG. 3 is a diagram illustrating a detailed configuration example of the server device 60 illustrated in FIG. 1. As illustrated in FIG. 3, the server device 60 includes a CPU 61, a ROM 62, a RAM 63, an HDD 64, an I / F 65, and a bus 66.

  Here, the CPU 61 executes arithmetic processing based on programs stored in the ROM 62, the RAM 63, and the HDD 64, and controls each unit of the apparatus.

  The ROM 62 is a semiconductor memory device that stores basic programs and data executed by the CPU 61. The RAM 63 is a semiconductor storage device that temporarily stores a program executed by the CPU 61 and temporarily stores data generated when the program is executed.

  The HDD 64 is a magnetic storage device that stores basic programs executed by the CPU 61 and data.

  The I / F 65 transmits and receives data to and from the client device 40 via the network 50.

  The bus 66 is a signal line group for electrically connecting the CPU 61, the ROM 62, the RAM 63, the HDD 64, and the I / F 65 to each other and enabling data exchange between them.

  FIG. 4 is a functional block realized by the CPU 61 executing the programs stored in the ROM 62 and the HDD 64 in the server device 60 shown in FIG. The example shown in FIG. 4 includes a position information processing unit 601, a speech recognition processing unit 602, a speech synthesis processing unit 603, an image generation processing unit 604, a teaching processing unit 605, a coaching processing unit 606, a text processing unit 607, and an object processing unit 608. , A teaching DB (Data Base) 609, a coaching DB 610, a student DB 611, and a teacher DB 612.

  Here, the position information processing unit 601 receives position information input from the position information input device 10 of the client device 40 and transmitted via the client device 40 and the network 50, and detects movement of the student's wrist. Execute.

  The speech recognition processing unit 602 inputs speech data input from the speech input / output device 30 of the client device 40 and transmitted via the client device 40 and the network 50, and executes processing for converting into text data by speech recognition processing. To do.

  The speech synthesis processing unit 603 converts the text data supplied from the text processing unit 607 and speech attribute information into speech data (for example, WAV data), and supplies the speech data to the client device 40 via the network 50. In the client device 40, the audio data is supplied to the audio input / output device 30 to be output as audio. The attribute information supplied from the text processing unit 607 is also supplied with information on the fundamental frequency of the voice, gender, inflection, speech rate, formant component placement, and the magnitude of the speech. By synthesizing speech, it is possible to reproduce speech that includes speaker characteristics and emotions.

  The image generation processing unit 604 executes processing for generating polygon data based on data (for example, data related to an object) supplied from the object processing unit 608, and the obtained polygon data is transmitted to the client device via the network 50. 40. In the client device 40, polygon data is received, an image is generated by the GPU 45 through drawing processing, layout processing, rendering processing, and the like, and is supplied to the HMD 20 for display.

  The teaching processing unit 605 inputs information on the position of the student's wrist supplied from the position information processing unit 601 and information on the student's utterance supplied from the text processing unit 607, and based on teaching data stored in the teaching DB 609. Information for drawing an object (for example, a teacher's avatar) is supplied to the object processing unit 608, and text data for causing the object to speak is supplied to the text processing unit 607. Accordingly, it is possible to execute an operation of teaching a student a desired content, obtaining a response from the student, and presenting information corresponding to the response to the student.

  The coaching processing unit 606 inputs information regarding the position of the student's wrist supplied from the position information processing unit 601 and information about the student's utterance supplied from the text processing unit 607, and the student's utterance supplied from the text processing unit 607. Is input to the object processing unit 608 based on the coaching data stored in the coaching DB 6410, and information for drawing an object (for example, a teacher's avatar) is supplied to the text processing unit 607. Supply text data for uttering avatar. In this way, students can hear deeper psychological opinions and impressions and spread the opinions and impressions they have heard.

  The text processing unit 607 executes the morphological analysis processing, the syntax analysis processing, the semantic analysis processing, and the context analysis processing as the natural language analysis processing on the text data supplied from the speech recognition processing unit 602, and is obtained. The processed result is supplied to the teaching processing unit 605 and the coaching processing unit 606. The text processing unit 607 performs morphological analysis processing, syntax analysis processing, semantic analysis processing, and context analysis processing on the text data supplied from the teaching processing unit 605 and the coaching processing unit 606 as necessary. The text data is converted and supplied to the speech synthesis processing unit 603.

  The object processing unit 608 generates a teacher avatar based on information supplied from the supply to the teaching processing unit 605 and the coaching processing unit 606, and supplies the teacher avatar to the image generation processing unit 604. The object processing unit 608 also displays an object (for example, a question sentence, an image, a moving image, and a stereoscopic image) indicating the content to be taught based on information supplied from the supply to the teaching processing unit 605 and the coaching processing unit 606. To the image generation processing unit 604.

  The teaching DB 609 is a database that stores learning contents taught to students. The coaching DB 610 is a database that stores information related to coaching performed on students.

  The student DB 611 is a database that stores attribute information related to students. The teacher DB 612 is a database that stores attribute information related to teachers.

  FIG. 5 is a diagram showing an example of student attribute information stored in the student DB 611 shown in FIG. As shown in FIG. 5, student attribute information includes personal attribute information and collective attribute information. Here, the personal attribute information is attribute information of each student, and includes surface layer attribute information and deep layer attribute information. The surface layer attribute information is information regarding superficial attributes that can be observed from the appearance, and the deep layer attribute information is information regarding deep attributes that cannot be observed from the appearance. Further, collective deep attribute information is included as an attribute including a plurality of deep attribute information. The collective deep attribute information is attribute information that includes a plurality of deep attribute information and whose contents change over time.

  FIG. 6 is a diagram showing an example of student attribute information shown in FIG. As shown on the left side of FIG. 6, the surface layer attributes constituting the student attribute information include, for example, age, gender, height, weight, deviation value, and IQ (Intelligence Quotient). Of course, these are merely examples, and attribute information other than these may be included.

  Further, as shown on the right side of FIG. 6, the deep attribute information includes interest, affiliation, problem, mission, skepticism, fear, and the like. Interest refers to the object of interest, affiliation refers to the region or group that the student is considered to belong to, problem refers to the student's awareness of the problem, and mission refers to the student's own mission Skepticism indicates the object of skepticism that the student has, and awe indicates the object of awe that the student envisions. Of course, these are merely examples, and attribute information other than these may be included.

  FIG. 7 is a diagram illustrating an example of attribute information stored in the teacher DB 612 illustrated in FIG. As shown in FIG. 7, the teacher attribute information includes personal attribute information and collective deep attribute information as in the case of students. Here, the personal attribute information is attribute information of each teacher, and includes surface layer attribute information and deep layer attribute information. The surface layer attribute information is information indicating surface attributes that can be observed from the appearance, and the deep layer attribute information is information indicating deep attributes that cannot be observed from the appearance. Further, collective deep attribute information is included as an attribute including a plurality of deep attribute information. The collective deep attribute information is attribute information that includes a plurality of deep attribute information and whose contents change over time.

  FIG. 8 is a diagram illustrating an example of teacher attribute information illustrated in FIG. 7. As shown in FIG. 8, the surface layer attribute information constituting the teacher attribute information includes, for example, age, sex, height, weight, fundamental frequency of voice, speaking speed, and avatar. Of course, these are merely examples, and attribute information other than these may be included.

  Moreover, as deep layer attribute information, it has 1st deep layer attribute information-3rd deep layer attribute information. Here, the first deep attribute information has interest, affiliation, problem, mission, skepticism, fear, and the like. Interest indicates the interest of the teacher, affiliation indicates the region or group that the teacher thinks that the teacher belongs to, the problem indicates the teacher's awareness of the problem, and the mission is the mission that the teacher envisions The skepticism indicates the object of the skepticism that the teacher has, and the awe indicates the object of the awe that the teacher envisions. Of course, these are merely examples, and attribute information other than these may be included.

  As the second deep attribute information, priority order at the time of teaching, etc. is shown, and emphasis on understanding, basic emphasis, application emphasis, progress emphasis, memory emphasis, and examination emphasis. Here, emphasis on understanding emphasizes student understanding, emphasis on basic emphasis on students' basic skills, emphasis on application emphasizes student's application skills, emphasis on progress emphasizes the progress of learning, and emphasis on memory means the importance of students. Emphasis is placed on memory consolidation, and emphasis on exams indicates emphasis on student test results. In addition, you may make it show these as a numerical value between 1-10 which makes 10 the largest, for example. Of course, these are merely examples, and attribute information other than these may be included.

  As the third deep attribute information, there are a first method to a sixth method. These methods show how to teach. As a first method, for example, in the case of language, a method for teaching a phonetic symbol of a word first is shown. As a second method, for example, a method for teaching a word and teaching a phonetic symbol later is shown. Note that the method may be a very detailed method of teaching, or a slightly rougher method. Of course, these are merely examples, and attribute information other than these may be included.

(B) Description of Operation of the Embodiment of the Present Invention Next, the operation of the embodiment of the present invention will be described. FIG. 9 is a flowchart illustrating an example of processing executed in the server device 60 when a new student is registered. For example, when a new student registers in the learning support system 1 shown in FIG. 1 shown in FIG. 1, the process shown in FIG. 9 is started. When the process shown in FIG. 9 is started, the following steps are executed.

  In step S10, surface attribute information such as age, sex, and height of a newly registered student is input. More specifically, the teaching processing unit 605 shown in FIG. 4 reads text data for registering a new student from the teaching DB 609 (for example, text data that prompts each item of surface attribute information to be input), and performs text processing. The data is supplied to the speech synthesis processing unit 603 via the unit 607. As a result, the text data is converted into voice data by the voice synthesis processing unit 603 and supplied to the client device 40 via the network 50. The client device 40 supplies the received audio data to the audio input / output device 30. The voice input / output device 30 generates and outputs voice corresponding to new student registration text data. At this time, the object processing unit 608 generates a registration avatar (for example, a guidance avatar), supplies the avatar to the client device 40 via the image generation processing unit 604 and the network 50, and sends a VR (Virtual) to the HMD 20. Reality) The image may be displayed in synchronization with the sound. The student operates the position information input device 10 according to the voice guidance by the avatar image, and inputs, for example, age, sex, height, weight, and the like as position information. The position information indicating the surface layer attribute information input in this way is supplied from the client device 40 to the server device 60 via the network 50.

  In step S <b> 11, the position information processing unit 601 of the server device 60 processes the position information supplied from the client device 40 to acquire attribute information, and supplies the attribute information to the teaching processing unit 605. The teaching processing unit 605 registers the supplied surface layer attribute information in the student DB 611 as new student attribute information.

  In step S12, the object processing unit 608 generates a student avatar with reference to the surface layer attribute information registered in the student DB 611 in step S11, and registers the student avatar in the student DB 611. The student avatars generated in this way are supplied from the object processing unit 608 to the image generation processing unit 604, displayed on the HMD 20 via the network 50 and the client device 40, and corrected as necessary. You may do it. For example, hair color, facial expression, height, weight, and the like may be corrected based on information input from the position information input device 10.

In step S <b> 13, the teaching processing unit 605 reads data (image data, text data, etc.) related to the achievement test registered in the teaching DB 609 and supplies the data to the text processing unit 607 and the object processing unit 608. The object processing unit 608 generates object (for example, object related to the problem) data for performing the achievement test and an avatar explaining the data, and supplies the data to the image generation processing unit 604. The image generation processing unit 604 generates a polygon based on the object data supplied from the object processing unit 608, executes a rendering process and the like, and supplies it to the client device 40 via the network 50. The client device 40 displays the supplied information on the HMD 20. Thereby, the academic ability test problem and the teacher's avatar are displayed on the HMD 20. In addition, the text processing unit 607 supplies text data for explaining the problem to the speech synthesis processing unit 603. The voice synthesis processing unit 603 executes voice synthesis processing based on the text data supplied from the text processing unit 607 and supplies the obtained voice data to the client device 40 via the network 50. The client device 40 outputs the supplied audio information via the audio input / output device 30. As a result, the academic ability test problem and the voice of the teacher's avatar are output from the voice input / output device 30. In more detail, for example, a teacher's avatar utters “Please write a graph of y = 1/4 × x ² ” and “Please write a graph of y = 1/4 × x ² ”. Is displayed as a character. The HMD 20 displays a virtual graph paper in front of the student. The student can draw a curve obtained by moving his / her hand on a virtual graph paper. In the case of science, the teacher's avatar utters “Please perform an experiment to generate oxygen.” And “Please perform an experiment to generate oxygen.” Is displayed as text. In addition, the HMD 20 displays laboratory instruments such as test tubes and flasks, and chemicals such as hydrogen peroxide, hydrochloric acid, sulfuric acid, ammonia water, manganese dioxide, platinum, and boric acid on the front of the student. The student can perform the operation of generating oxygen by moving hands and placing laboratory instruments and selecting and feeding chemicals. Such academic ability tests can be conducted for each department.

  In step S14, the teaching processing unit 605 obtains and scores the results of the achievement test performed in step S13. Then, based on the scored result and the result of the achievement test for all the students stored in the teaching DB 609, the deviation value of the newly registered student is calculated. For example, in the example of the mathematical test described above, if the drawn curve is completely correct, the score is 10 points. If only the shape (curved convex curve) is correct, the score is 5 points. Will be scored as 0 points. In addition, in the case of science experiments, if all the experimental procedures are correct, the score is 10 points, and even if part of the procedure is wrong, if the drug selection is correct, the score is 5 points. If the selection is wrong, it can be scored as 0 points. The deviation value can be obtained by summing up the results of scoring as described above.

  In step S15, the teaching processing unit 605 registers the deviation value calculated in step S14 as the “deviation value” of the surface layer attribute information in the student DB 611. For example, “mathematics = 65”, “English = 55”, “national language = 67”, “society = 71”, and “science = 72” are registered as deviation values of newly registered students. Not only the academic ability test but also the IQ test may be performed in the same manner, and the result of the IQ test may be registered in the student DB 611 as surface layer attribute information.

  In step S <b> 16, the teaching processing unit 605 reads information related to the psychological test registered in the teaching DB 609 and supplies the information to the text processing unit 607 and the object processing unit 608. The object processing unit 608 generates an object for executing a psychological test and an avatar explaining the object, and supplies the generated object to the image generation processing unit 604. The image generation processing unit 604 generates a polygon based on the object information supplied from the object processing unit 608, executes a rendering process and the like, and supplies it to the client device 40 via the network 50. The client device 40 displays the supplied information on the HMD 20. Thereby, the psychological test problem and the teacher's avatar are displayed on the HMD 20. Further, the text processing unit 607 supplies text data for explaining the problem and voice attribute information to the voice synthesis processing unit 603. The voice synthesis processing unit 603 executes voice synthesis processing based on the text data supplied from the text processing unit 607 and the attribute information, and supplies the obtained voice data to the client device 40 via the network 50. . The client device 40 outputs the supplied audio information via the audio input / output device 30. As a result, the psychological test problem and the voice of the teacher's avatar are output from the voice input / output device 30. More specifically, for example, a plurality of broken castles are displayed in front of the students of the HMD 20, and for example, scientists, mathematicians, scholars, sociologists, etc. live in each of the castles. , Today there is a narration that the castle owner is absent. There is a narration that students are allowed to select one of these castles and enter. As a result, it is determined that the old castle selected by the student shows “interest” in the deep attribute. For example, when an old castle where “scientists” live is selected, it is determined that the object of interest is “science”. Note that the above is an example, and various other psychological tests can be executed.

  In step S17, the teaching processing unit 605 registers the result of the psychological test performed in step S16 as deep attribute information in the student DB 611. For example, “Science” is registered as the “interest” of the newly registered student, “School of Science” is registered as “Affiliation”, “I am not good at memorizing subjects” is registered as “Problems”, and “Mission” "Making medicines for incurable diseases in the future" is registered, "skepticism" is registered "whether the current study method is correct", and "grandfather of a former university professor" is registered as "awe". In addition, the above is an example and you may make it register other than these. In addition, instead of registering as “words”, for example, numerical values may be registered. For example, when “extroversion” and “introversion” are registered as the deep attributes, a large value may be registered when the outward tendency is strong, and a small value may be registered when the inward tendency is strong. Of course, in addition to extroversion or introversion, indices such as sociability, nervousness, adhesiveness, manifestation quality, mitotic quality, and circulation quality may be registered as numerical values. Note that the deep attribute information is not fixed, and new deep attribute information may be introduced according to changes in society, for example. As an example, if it is a social situation where the group's belonging consciousness is evaluated, the belonging consciousness is added as deep attribute information, and conversely, if it is a social situation where it is not evaluated, the independent consciousness is added as deep attribute information. It may be. In other words, the collective deep attribute is a deep psychological attribute common to society, and is considered to be an attribute that changes in accordance with changes in society and fashion.

  In step S18, the teaching processing unit 605 obtains attributes similar to the surface attribute information registered in step S11, the achievement test result registered in step S15, and the deep attribute information that is the result of the psychological test registered in step S17. An existing student is identified from the student DB 611. For example, a weight value is set for each of the various attribute information described above, and a value obtained by adding the products of the weight value and each attribute information is calculated as the similarity. Then, the existing students are sorted in descending order of similarity, and the sorted upper students are identified as similar students. For example, when the degree of similarity is set to a value of 0 to 100, students can be sorted in descending order of degree of similarity and the top 100 names can be selected. Note that the weight values described above may be changed in weighting by a learning process based on, for example, regression analysis using the teacher's evaluation in the process described later as an index to improve the accuracy of similarity determination.

  In step S19, the teaching processing unit 605 identifies a teacher who is determined to be compatible with the student determined to be similar in step S18. For example, in the above-described example, the top 100 students determined to have a high degree of similarity are sorted in the order of increasing deviation values. Then, it is possible to determine that the student located at the top is a student having good compatibility with the teacher. Then, it is possible to identify a teacher in charge of a student with a high degree of similarity as a compatible teacher. For example, four teachers estimated to be compatible are identified. The compatibility is not judged from the upward trend of the deviation value, but the quality of the compatibility is judged by other index values, for example, indicators that students can relax (relief), easy to understand, and easy to talk to. You may make it do.

  In step S <b> 20, the teaching processing unit 605 acquires data related to the teacher specified in step S <b> 19 from the teacher DB 612 and supplies the data to the text processing unit 607 and the object processing unit 608. For example, data related to the four teachers identified in step S19 is selected from the teacher DB 612 and supplied to the text processing unit 607 and the object processing unit 608. As a result, the avatar images and attribute information of the four teachers as shown in FIG. 10 are displayed on the HMD 20. The student can select a desired teacher by operating the position information input device 10 and listen to the teacher's story. Of course, you may be able to talk. By displaying such an avatar image and listening to the desired teacher's way of speaking, the student can identify the ideal teacher. For example, the student selects “xx teacher” as an ideal teacher in FIG.

  In step S21, the teacher selected in step S20 is set as a teacher in charge. More specifically, the teaching processing unit 605 registers the teacher selected in step S20 in the student DB 611 as the teacher in charge of the student.

  As described above, by registering the surface attribute information and the deep attribute information of the student by the process of the flowchart shown in FIG. 9, the existing student having the surface attribute information and the deep attribute information having high similarity is specified, Teachers estimated to be compatible with the identified students can be identified.

  Next, the teaching process will be described with reference to FIG. When the processing of the flowchart shown in FIG. 11 is started, the following steps are executed. Note that the following flowchart is executed when, for example, the student wears the HMD 20 to receive a lesson (teaching) and the position information input device 10 and the voice input / output device 30.

  In step S30, the teaching processing unit 605 determines whether or not it is the first teaching for the student, and when it is determined that it is the first time (step S30: Y), the process proceeds to step S31, and otherwise (step S30). In S30: N), the process proceeds to step S32. For example, when a student takes a lesson for the first time, it is determined as Y and the process proceeds to step S31.

  In step S31, the teaching processing unit 605 selects the subject selected by the student (eg, Japanese language, society, science, mathematics, English, etc.), the deviation value of the selected subject at that time, and the deep attribute information of the student. A curriculum is generated based on the deep attribute information of the teacher. For example, the subject selected by the student is “English”, the deviation value at that time is “50”, the student's interest in the deep attribute information is “science”, and the teacher attaches importance to the second deep attribute information. And the second method (for example, grammar-oriented) has a high score. In that case, the student's English grade is normal, the student has a high level of logical thinking ability, and the teacher emphasizes basics and grammar, so a curriculum that teaches grammar logically is generated. To do. The above is an example, and the curriculum may be generated by other methods.

  In step S <b> 32, the teaching processing unit 605 identifies a location that was previously ended from the student curriculum stored in the teaching DB 609, for example. For example, when the student is taking a class on English grammar, if the student progresses to the basic problem of the second sentence type out of five sentence types, the application problem starts. In this way, since the class from the previous time is automatically resumed by the student wearing the HMD 20 or the like in this way, it is not necessary to memorize the contents of the previous class, and the class is started quickly. be able to.

  In step S <b> 33, the teacher's avatar is displayed on the HMD 20. More specifically, the teaching processing unit 605 searches the student DB 611 based on information for identifying a student (for example, ID information) to identify the student, and information on the teacher selected by the identified student. Obtained from the teacher DB 612. For example, in FIG. 10, when xx teacher is selected, information corresponding to xx teacher is acquired from the teacher DB 612. At this time, avatar and voice attribute information (voice fundamental frequency, speaking speed, intonation, etc.) is also acquired.

  In step S34, the teaching processing unit 605 executes processing for displaying the learning content. More specifically, the teaching processing unit 605 acquires information regarding the learning content of the student from the teaching DB 609 and supplies the information to the text processing unit 607 and the object processing unit 608. As a result, the object processing unit 608 generates information for displaying (operating) the teacher's avatar and information related to the learning content (for example, an example sentence in the case of English), and the image generation processing unit 604 generates the information. Supply. The image generation processing unit 604 executes polygon processing and rendering processing based on the information supplied from the object processing unit 608 and supplies the obtained image information to the client device 40 via the network 50. The client device 40 executes a drawing process based on the received polygon information, and supplies it to the HMD 20 for display. Also, the text processing unit 607 generates text data for uttering the teacher's avatar and attribute information of the teacher's voice, and supplies the text data to the speech synthesis processing unit 603. The voice synthesis processing unit 603 synthesizes voice data based on the text data supplied from the text processing unit 607 and supplies the synthesized voice data to the client device 40 via the network 50. The client device 40 outputs the received audio data from the audio input / output device 30. Thus, for example, the avatar of xx teacher shown in FIG. 10 is synthesized as a VR image and displayed on the HMD 20, and the xx teacher displays an English example sentence (for example, “She is painting the wall blue” . ”) Can be displayed. In order to encourage students to stay in memory, an avatar (she is painting the wall blue.), An example sentence (She is painting the wall blue.), A sentence pattern (fifth sentence pattern) (SOVC)) may be displayed.

  In step S35, the teaching processing unit 605 acquires a student's reaction. For example, the teaching processing unit 605 obtains the student's reaction (emotion etc.) from the change in the frequency characteristics of the speech uttered by the student by the voice recognition processing unit 602. For example, when the frequency is lower than usual, it can be determined that the user is bored. Also, the teaching processing unit 605 acquires the student's reaction by detecting the student's movement by the position information processing unit 601. The student's facial expression may be captured by a camera or the like, and the student's reaction may be acquired from the student's facial expression reflected in the captured image. Note that the student's reaction can be classified into three categories of “active”, “normal”, and “dull” based on the information acquired at this time. Or you may make it evaluate by 10-step evaluation which makes the most active state 10.

  In step S36, the teaching processing unit 605 performs an understanding level test. For example, in the present example, a comprehension test for the fifth sentence pattern in English is executed. More specifically, the teaching processing unit 605 acquires information related to the understanding level test from the teaching DB 609 and supplies the information to the text processing unit 607 and the object processing unit 608. As a result, the problem of understanding test and the teacher's avatar are displayed on the HMD 20, and voice guidance is output from the voice input / output device 30. The student can answer the comprehension test by operating the position information input device 10 or speaking to the voice input / output device 30.

  In step S <b> 37, the teaching processing unit 605 receives a teacher evaluation input. More specifically, the teaching processing unit 605 acquires information related to teacher evaluation from the teaching DB 609 and supplies the information to the text processing unit 607 and the object processing unit 608. As a result, information for evaluating the teacher (for example, an input box for five-step evaluation) is displayed on the HMD 20, and voice guidance is output from the voice input / output device 30. The student can evaluate the teacher by operating the position information input device 10 or speaking to the voice input / output device 30.

  In step S <b> 38, the teaching processing unit 605 registers the student reaction, the understanding test result, and the teacher evaluation in the student DB 611. For example, “dull” is registered as the student's response, “deviation value 50” is registered as the result of the comprehension test, and “3 in five-level evaluation” is registered as the evaluation of the current teacher of the student. . The comprehension test, teacher evaluation, and student response do not need to be acquired every time, but may be acquired and registered at a frequency as necessary.

  In step S39, the teaching processing unit 605 executes processing for modifying the teacher and the method based on the student's reaction registered in step S38, the result of the understanding level test, and the teacher's evaluation. Details of this processing will be described later with reference to FIG.

  According to the above process, the teaching process from the teacher to the student can be executed.

  Next, details of the teacher and method modification processing shown in step S39 of FIG. 11 will be described with reference to FIG. When the process of FIG. 12 is started, the following steps are executed.

  In step S50, the teaching processing unit 605 creates an A / B model in which the surface layer attribute information and the deep layer attribute information are changed. For example, as an example, the fundamental frequency of the voice of the surface layer attribute information can be changed, the speaking speed can be changed, or the avatar can be changed. At this time, instead of changing all at once, one set of teachers who have changed part is created (A / B model is created). In addition, the first deep attribute information that affects the subject matter of parables and popular stories that are spoken during the class may be changed according to the A / B test in the same manner as described above.

  In step S51, the teaching processing unit 605 performs setting so that the A / B model created in step S50 is alternately presented. As a result, for example, the teacher A is set so that the teacher A whose fundamental frequency is several hertz and the teacher B whose fundamental frequency is several hertz are alternately presented. In addition, a set of teachers that change part of the first deep attribute information that affects the subject matter of parables and popular stories that are spoken during class may be created and set as an A / B model. Good.

  In step S52, the teaching processing unit 605 executes a process of selecting the A / B model set last time based on the student's reaction, the result of the understanding test, and the teacher's evaluation stored in step S38 of FIG. To do. For example, a value obtained by multiplying each of the student's reaction, the result of the comprehension test, and the teacher's evaluation by multiplying the weight function is calculated for each A / B model, and the highest value A model can be selected. Needless to say, the evaluation may be performed using three or more models.

  In step S53, the teaching processing unit 605 identifies a method having a high effect in the third depth attribute illustrated in FIG. More specifically, a method that is relatively effective when applied to the same student (having a relatively high understanding test result) is identified. As an example, in the case of English, when comparing the first method that teaches pronunciation after teaching phonetic symbols, and the second method that teaches pronunciation without teaching phonetic symbols, the first method is the first method The first method is selected when the comprehension test result is better than the two methods.

  In step S54, the teaching processing unit 605 executes processing for modifying the method. For example, in the previous English example, the first method that teaches the phonetic symbols and then teaches the actual pronunciation is modified, for example, as a method that teaches a plurality of similar consonants or vowels as an opposing structure of differences. As a modification method, there are a method of changing the teaching order, a method of changing the teaching viewpoint, a method of changing the teaching priority, and the like. Of course, other changing methods may be used. In short, since the purpose is to obtain a more effective method by changing the teaching contents rather than fixing them, small changes may be possible.

  In step S55, the teaching processing unit 605 copies the modified method. More specifically, the method modified in step S54 is copied as another method. For example, in the above example, the first method of teaching the phonetic symbol and then teaching the actual pronunciation is modified as, for example, a method of teaching a plurality of similar consonants or vowels as an opposing structure of differences, Copy as 7 methods (new method).

  In step S56, the teaching processing unit 605 identifies a method having a low effect from the methods before and after the modification. For example, in the above-described example, the first method before modification is compared with the seventh method after modification, and the first method is specified as a method having a relatively low effect. Note that if there is a method that is clearly ineffective among the methods before and after modification, select the method, and if the effect is almost the same, for example, either randomly, based on random numbers, etc. May be selected.

  In step S57, the teaching processing unit 605 deletes the method having the low effect identified in step S56.

  In step S58, the teaching processing unit 605 identifies a method having a high effect from the methods before and after the modification. For example, in the above-described example, the first method before modification and the second method after modification are compared, and the seventh method is identified as a method having a relatively high effect. Note that if the method before and after the modification has substantially the same effect, no identification is performed in the process of step S58. The process in step S58 is a process for specifying a method that has a clearly high effect, and is not specified when the effect is not clear.

  In step S59, the teaching processing unit 605 executes a process for diffusing the highly effective method specified in step S58. More specifically, as shown in FIG. 8, the method belongs to the third deep attribute information, and as shown in FIG. 7, the deep attribute information belongs to the personal attribute information of each teacher. For this reason, if a method belonging to a certain teacher (caused by modification) is found to be effective, it is effective by diffusing (copying) that method to the deep attribute information of other teachers. Methods can be shared as collective deep attribute information. When the process of step S59 ends, the process returns (returns) to the original process.

  According to the above processing, the surface layer attribute information and the deep layer attribute information of the teacher can be evolved to be optimal for each student by the A / B test. Also, methods can be evolved into more effective methods by modification, copying (copying), and selection (淘汰).

  Next, with reference to FIG. 13, processing executed when the teacher performs coaching on the student will be described. Coaching is a method of deriving students' self-realization and achievement of goals through dialogue. Teachers can encourage voluntary action by listening to the student's story, communicating and approving what they feel, and asking questions.

  In step S70, the coaching processing unit 606 determines a coaching theme. As a method for determining the theme of coaching, for example, a list of themes can be presented via the HMD 20 or the voice input / output device 30 and selected from the presented lists. Alternatively, when a predetermined keyword appears in a conversation between a student and a teacher, the theme of coaching can be determined as a trigger. For example, if a student says such as “I'm aiming for a profession in the future”, “the future course” can be set as the theme of coaching.

  In step S71, the coaching processing unit 606 acquires a question corresponding to the theme identified in step S70 from the coaching DB 610. For example, in the case of the “future career” mentioned above, “What kind of occupation do you have a longing for?” “What is the reason for having a longing for that occupation?” “Is there anyone who has a job?”, “What is the purpose and reason for working for you?”, Etc. “Interest”, “(Future) affiliation”, “Problem (consciousness)”, The search identifies questions that lead to “mission,” “skepticism,” and “awe”.

  In step S72, the voice recognition processing unit 602 acquires a student's remarks. More specifically, voice data uttered by a student, converted into voice data by the voice input / output device 30, and transmitted via the client device 40 and the network 50 is acquired. As a result, the student speaks “I want to find a job that uses English in the future.”

  In step S73, the speech recognition processing unit 602 performs speech recognition processing on the speech data acquired via the network 50 in step S72, and the text processing unit 607 performs text conversion. In the example described above, the spoken voice data “I want to work in English in the future.” Is converted into text.

  In step S <b> 74, the coaching processing unit 606 acquires the student's statement made into text in step S <b> 73 and acquires a question corresponding to the statement from the coaching DB 610. For example, if the above-mentioned “I want to work in English in the future” is a student's statement, “Why is it English?” Is selected. Such conversation contents can be realized by presuming the conversation contents in advance and storing appropriate conversations based on the assumed conversation and based on the coaching technique. Coaching is a conversation to bring out opinions and emotions buried deep in the students. Compared with free conversation (daily conversation, etc.), teachers are passive, so there are fewer questions to prepare. That's it.

  In step S75, the coaching processing unit 606 supplies the question acquired in step S74 to the text processing unit 607, which is voiced by the speech synthesis processing unit 603 and output. As a result, the question acquired in step S74 is output from the voice input / output device 30 of the client device 40.

  In step S76, the coaching processing unit 606 determines whether or not a conclusion has been reached by coaching. If it is determined that the conclusion has been reached (step S76: Y), the process proceeds to step S77, and otherwise (step S76: In N), the process returns to step S72, and the same process as described above is repeated. For example, when a student utters a specific keyword (for example, “Yes?”, “Sure.”, “I see.”, Etc.), it is estimated from the voice that the student's emotion is high (for example, When the fundamental frequency of the voice is higher than normal), when the student makes an intention to end (for example, when a predetermined command is input from the position information input device 10), the conclusion is reached. Determination can be made and the process can proceed to step S77.

  In step S77, the coaching processing unit 606 stores the conclusion reached in step S76. For example, “I wanted to be a profession that uses English in the future, but I want to become a doctor or a nurse because there is a desire to help people.” As a conclusion, the result is stored in the student DB 611. .

  In step S78, the coaching processing unit 606 searches the student DB 611 for a student having an attribute similar to that of the student who has been coached and coached in the past with the same theme, and obtains a conclusion thereof. More specifically, for example, a student who has deep attribute information similar to the student and has coached on the theme of “future course” in the above example is identified and the conclusion of the coaching is obtained. . As for the similarity of the deep attribute information, for example, the similarity can be determined by comparing the values obtained by the product-sum operation of each of the deep attribute information shown in FIG. 6 and the weight value. Of course, other methods may be used.

  In step S79, the coaching processing unit 606 feeds back the validity of the conclusion. More specifically, first, a plurality of conclusions acquired in step S78 are compared with the current conclusion. As a result, if the conclusion is the same as or similar to the conclusion included in a plurality of conclusions, the fact that it is appropriate is fed back. Note that a model that estimates a conclusion similar to a certain conclusion may be constructed using a technique such as Skip-gram, and the validity of the result may be determined based on such an estimation result. In the case of a theme such as a future course, actual information of a student (senior) who has advanced to the course may be registered and compared with the registration result. For example, when the theme is a future course, deep attribute information at the present time of students who have actually become nurses or doctors (for example, satisfaction with work, ease of work, decent work, salary level, work-life balance, etc. May be investigated and registered, and compared with such deep attribute information. With such a comparison, for example, regarding the career path, it is possible to accurately determine the talent and aptitude of the student by comparing the deep attribute information between the person who is successful in the field and the student.

  According to the above flowchart, by coaching the student, for example, the future course can be further clarified and the motivation can be increased, thereby increasing the awareness of learning. Further, by registering the result obtained by coaching as deep attribute information, it is possible to more accurately determine student attributes. Also, by using the avatar synthesized by the server device 60 as coaching, a more probable conclusion can be obtained by making thoughts and judgments without prejudice than when confronting a live human having a stake. be able to.

  In addition, the question used by the teacher during coaching may be modified and optimized by the same process as the method modification process related to teaching described with reference to FIG.

(C) Description of Modified Embodiment Each of the above embodiments is an example, and it is needless to say that the present invention is not limited to the case described above. For example, in the above example, the position information processing unit 601, the speech recognition processing unit 602, the speech synthesis processing unit 603, the image generation processing unit 604, the teaching processing unit 605, the coaching processing unit 606, the text processing unit 607, and the object processing unit 608. The teaching DB 609, the coaching DB 610, the student DB, and the teacher DB 612 are all arranged in the server device 60, but these are arranged in the client device 40, or they are distributed to the client device 40 and the server device 60. Or may be arranged.

  In the above embodiment, the teacher uses the avatar and the synthesized speech generated by the computer processing. However, for example, for a student who does not like the teaching by the computer, a teacher by a living person is used. You may make it allocate. In addition, at the time of initial registration and at the time of coaching, a live human teacher may be assigned. That is, the teacher attributes shown in FIGS. 7 and 8 may include real teacher attribute data.

  In the above embodiment, the devices arranged on the student side are the position information input device 10, the HMD 20, the voice input / output device 30, and the client device 40, but instead of these or one of them. The unit may be a mobile terminal (for example, a smartphone). For example, the HMD 20, the voice input / output device 30, and the client device 40 can be replaced by attaching a smartphone to head mounted goggles. Of course, a two-dimensional image or video may be used instead of the VR image.

  In the above embodiment, teaching is taught in teaching, but other contents may be taught. For example, it may be possible to teach work contents, correspondence to customers, etc. as employee training. In the present embodiment, since teaching is performed using VR video, for example, employee training can provide training with higher reality by using VR video of actual customers.

  In the above embodiment, the teacher's evaluation is determined based on the progress of learning and the like, but may be determined by other methods. For example, the student's guardian, not the student himself / herself, may make the determination based on a change in the student's approach to learning.

  In addition, the students themselves can freely modify the attributes of the teacher (surface attribute information and deep attribute information), and the teacher thus modified may be exchanged with other students, for example, for a fee / free of charge. . As a result, the teacher can create a teacher who is better at teaching, and a plurality of students can share the teacher created in this way.

  The method modification may also allow the student to be involved. For example, students may be able to find a method that suits them. Note that the method obtained by such modification may be exchanged with other students for a fee / free of charge.

  Further, in the above embodiment, the case where both the teacher and the student are described as an example, but a plurality of teachers and / or students may be provided. For example, there may be one teacher, and a plurality of students who are compatible with the teacher may gather to advance the class. Further, a plurality of teachers may be provided, and a plurality of students having good compatibility with those teachers may be gathered to advance the class. That is, the ratio of the number of teachers and students may be singular or plural.

  As an academic ability test, it may be possible to take a VR examination at home. Specifically, it may be possible to take the examination at a place convenient for the student (for example, home, school, library, etc.) without visiting the venue. In addition, immediately after the test, an answer by the teacher in charge may be made to correct the mistake early and to fix the memory.

  In the above embodiment, reference books are not described in detail, but for example, a plurality of types of reference books, dictionaries, teaching materials, etc. are registered, and the registered reference books etc. Reference books that match the level or that match the textbooks used at the school may be selectable.

  Moreover, the flowcharts shown in FIGS. 9 and 11 to 19 are examples, and the present invention is not limited to these flowcharts.

1: Learning support system 10: Position information input device 20: HMD
30: Audio input / output device 40: Client device 41: CPU
42: ROM
43: RAM
44: HDD
46: I / F
47: Bus 50: Network 60: Server device 61: CPU
62: ROM
63: RAM
64: HDD
65: I / F
66: Bus 601: Position information processing unit 602: Speech recognition processing unit 603: Speech synthesis processing unit 604: Image generation processing unit 605: Teaching processing unit 606: Coaching processing unit 607: Text processing unit 608: Object processing unit 609: Teaching DB
610: Coaching DB
611: Student DB
612: Teacher DB

Claims (2)

In a learning support system that supports learning with a computer,
First storage means for storing surface attribute information indicating measurable surface attributes of the student and deep attribute information indicating deep attributes other than the surface attribute information;
Teaching means for presenting a virtual teacher having an avatar drawn by the computer and a voice synthesized by the computer, and teaching the student the learning content via the presented virtual teacher;
Second storage means for storing attribute information indicating the type of the avatar of the virtual teacher, the characteristics of the synthesized speech, and the teaching method ;
Determining means for determining a progress state of each of the students;
When registering a new student, a search for searching for the student having similar attribute information from the first storage means with reference to the surface attribute information and the deep attribute information of the new student Means,
The virtual teacher determined to have a relatively high evaluation among the virtual teachers presented to the student having the attribute information similar to the new student acquired by the search means Specifying means for specifying from the second storage means the attribute information related to a teacher,
Control means for performing control to present the virtual teacher to the teaching means for the new student based on the attribute information specified by the specifying means,
A modification means for identifying and modifying the teaching method that is highly effective among the teaching methods by the teaching means,
Among the teaching methods modified by the modifying means, the teaching means that has a high learning effect is also used for the other students.
A learning support system characterized by that. In a learning support method for supporting learning by a computer,
A first storage step of storing in the storage means surface attribute information indicating a measurable surface attribute of the student and deep attribute information indicating a deep attribute other than the surface attribute information;
A teaching step of presenting a virtual teacher having an avatar drawn by the computer and a voice synthesized by the computer, and teaching the student the learning content through the presented virtual teacher;
A second storing step of storing in the storage means attribute information indicating the type of the avatar of the virtual teacher, the characteristics of the synthesized voice, and the teaching method ;
A determination step of determining a progress state of each of the students;
When registering a new student, a search step of searching for the student having the similar attribute information from the storage means with reference to the surface attribute information and the deep attribute information of the new student. ,
The virtual teacher determined to have a relatively high evaluation among the virtual teachers presented to the students having the attribute information similar to the new student acquired in the search step A specific step of identifying the attribute information related to a teacher from the storage means;
A control step of performing control to present the virtual teacher to the teaching step for the new student based on the attribute information specified in the specifying step;
A modification step of identifying the teaching method that is highly effective among the teaching methods by the teaching step and modifying the teaching method;
In the teaching step, the teaching method modified in the modification step is used for the other students for those having a high learning effect.
A learning support method characterized by that.

Images