JP7495699B1 - Learning support system and learning support method - Google Patents

Abstract

[Problem] It is possible to provide a lecture by an avatar instructor suitable for each student. [Solution] A learning support system that provides content related to a lecture by an avatar instructor has a storage unit that stores student information, which is information about the student, a setting unit that sets external feature information indicating the external features of the avatar instructor who will lecture to the student and internal feature information indicating the internal features of the avatar instructor based on the student information of the student, a generation unit that generates text data of content to be provided to the student based on the features indicated in the external feature information and/or the features indicated in the internal feature information, an output unit that outputs content in which the avatar instructor, having set the features indicated in the external feature information, speaks out the text data, and a profile presentation unit that presents profile information of the avatar instructor based on the internal feature information to the student. [Selected Figure] FIG.

Description

The present invention relates to technology that supports learning using a computer system.

Patent Document 1 discloses a learning system that provides lectures including video and audio. The learning system provides students with lecture content consisting of video and audio through an avatar lecturer, a virtual person with an appearance that resembles a lecturer. The learning system realizes an interactive pseudo-dialogue between the avatar lecturer and the students by having the avatar lecturer address the students during breaks in the lecture content and detecting the students' reactions.

JP 2021-018316 A

In schools, prep schools, cram schools, and other institutions where real human instructors give lectures to students, it is known that not only the content of the lecture but also the instructor's lecturing style and the type of instructor themselves affect the effectiveness of the students' learning. For example, some students improve their academic ability when they are praised by the instructor, while others improve their academic ability when they are scolded by the instructor. In other words, some students prefer instructors who praise, while others prefer instructors who scold. Also, if students like their instructors, their motivation to study increases, and their academic ability may improve. However, even in individual instruction by real human instructors, it is difficult to assign an appropriate instructor to each and every student.

Even in a learning system that uses a computer to provide lectures by avatar instructors using video and audio, students' learning can be effectively supported if they are provided with a lecture by an avatar instructor that is most suitable for them.

However, in a learning system in which avatar instructors use computers to provide lectures using video and audio, little research has been done on how to assign an appropriate avatar instructor to each student.
One objective of the present disclosure is to provide a technology that enables a lecture to be provided to each student by an avatar instructor that is suitable for that student.

A learning support system according to one embodiment of the present disclosure is a learning support system that provides students with content related to a lecture, consisting of video and audio, by an avatar instructor, which is a virtual person modeled after an instructor, and includes: a storage unit that stores student information, which is information about the student; a setting unit that sets external feature information indicating the external features of the avatar instructor who will lecture the student and internal feature information indicating the internal features of the avatar instructor, based on the student information of the student; a generation unit that generates text data of the content to be provided to the student, based on the features indicated in the external feature information and/or the features indicated in the internal feature information; an output unit that outputs content in which the avatar instructor, having set the features indicated in the external feature information, speaks out the text data; and a profile presentation unit that presents profile information of the avatar instructor based on the internal feature information to the student.

According to one aspect of the present disclosure, it is possible to provide a technology that makes it possible to provide lecture-related content to each student through an avatar instructor that is suitable for each student by adapting the external and internal characteristics of the avatar instructor to each student.

1 is a block diagram showing a functional configuration of a learning support system according to a first embodiment. FIG. 2 is a block diagram showing a hardware configuration of the learning support system. FIG. 13 is a diagram showing a table of student information. 13 is a flowchart of an avatar instructor setting process. FIG. 13 is a diagram showing a table of external feature information. FIG. 13 is a diagram showing a table of internal characteristic information. FIG. 13 is a diagram showing a table of profile information. 13 is a flowchart of a lecture providing process. 13 is a flowchart of an encouraging process. 13 is a flowchart of a preference prompting process. 13 is a flowchart of an additional commentary insertion process. FIG. 11 is a block diagram showing a functional configuration of a learning support system according to a second embodiment. FIG. 11 is a diagram showing an example of concentration determination result information. FIG. 11 is a diagram illustrating an example of evaluation information. 13 is a flowchart of an instructor control process.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
First Embodiment

The learning support system of this embodiment is a computer system that provides students with lecture content consisting of video and audio by a virtual person (hereinafter also referred to as an "avatar lecturer") who resembles a lecturer. The learning support system also improves the effectiveness of the students' learning by having the avatar lecturer speak to the students at specific times during the provision of the lecture content.
FIG. 1 is a block diagram showing the functional configuration of a learning support system according to the first embodiment.

1, the learning support system 10 has a setting unit 11, a generating unit 12, an output unit 13, a profile presenting unit 14, and a storage unit 15. The storage unit 15 stores prepared data required to execute the service and information registered by the execution of the service. The processing of the setting unit 11, the generating unit 12, the output unit 13, and the profile presenting unit 14 is defined by a software program, and the processing is realized by a processor executing the software program. Details of each unit, data, and information will be described later.
FIG. 2 is a block diagram showing the hardware configuration of the learning support system.

Referring to FIG. 2, the learning support system 10 is composed of a server 20 and an information terminal such as a personal computer or a smartphone that can connect to the server 20 via a communication network such as the Internet. A student 90 connects to the server 20 using a browser 33 on an information terminal 30 and uses a service provided by the server 20. The service is a service that provides content related to learning. The service is realized by each unit shown in FIG. 1.

As shown in FIG. 2, the server 20 has, as its hardware, a processing device 21, a main memory 22, a storage device 23, a communication device 24, an input device 25, and a display device 26, which are connected to a bus 27.

The storage device 23 stores data in a writable and readable manner. The storage unit 15 shown in FIG. 1 is realized by the storage device 23. The processing device 21 is a processor that reads data stored in the storage device 23 into the main memory 22 and uses the main memory 22 to execute the processing of the software program. The setting unit 11, the generating unit 12, the output unit 13, and the profile presenting unit 14 shown in FIG. 1 are realized by the processing device 21. The communication device 24 transmits information processed by the processing device 21 via a communication network including wired or wireless or both, and also transmits information received via the communication network to the processing device 21. The received information is used for software processing by the processing device 21. The input device 25 is a device that accepts information input by an operator using a keyboard, mouse, etc., and the input information is used for software processing by the processing device 21. The display device 26 is a device that displays image and text information on a display screen in association with software processing by the processing device 21. The input device 25 and the display device 26 are provided mainly for use by an administrator (not shown) rather than the students 90.

Returning to FIG. 1, the storage unit 15 prestores script data D30, student information D10, and avatar instructor information D20. The avatar instructor information D20 includes inner characteristic information D21, outer characteristic information D22, and profile information D23. As the script data D30, text data that serves as a script of a content that includes a lecture for each unit of content to be learned is recorded. As an example, the text data of the script data D30 is created in standard Japanese. As the student information D10, information about each student 90 is recorded. As the avatar instructor information D20, information about the avatar instructor corresponding to each student 90 is recorded. The inner characteristic information D21 indicates the inner characteristic of the avatar instructor, and the outer characteristic information D22 indicates the outer characteristic of the avatar instructor.
FIG. 3 is a diagram showing a table of student information.

The student information D10 table registers the student number (No.), name, hometown, alma mater, preferences, desired lecturer dialect, and desired lecture style for each student 90. Name indicates the name of the student 90. Hometown indicates the hometown of the student 90. Alma mater indicates the school the student 90 attended. Preferences indicate the preferences of the student 90. Desired lecturer dialect indicates the dialect the student 90 would like the avatar lecturer to use. A dialect is a language system used in a particular region. Desired lecture style indicates the style of lecture the student 90 would like the avatar lecturer to use. For example, student 90, whose student number is 1, has the name "Patent Ichiro," is from "...Nagoya City, Aichi Prefecture," went to "XX High School," likes "cats" and the "Chunichi Dragons," wants the avatar instructor to speak in "Nagoya dialect," and wants a "scolding and encouraging" style of lecture.

Based on student information D10 of each student 90, setting unit 11 sets external feature information D22 and internal feature information D21 of an avatar instructor who will give a lecture to student 90.
FIG. 4 is a flowchart of the instructor avatar setting process.
In step S<b>101 , the setting unit 11 sets external feature information D<b>22 of an avatar lecturer who will give a lecture to students 90 .
FIG. 5 is a diagram showing a table of external feature information.
In the table of external feature information D22, the student number of each student 90 and the external features of the avatar instructor are recorded in association with each other.
Physical characteristics include appearance, voice, language system, and teaching style.

The appearance is information that represents the appearance of the avatar instructor, and as an example, an appearance selected by the student 90 from among appearances prepared in advance is set. The voice is information that represents the voice uttered by the avatar instructor, and as an example, a voice selected by the student 90 from among voices prepared in advance is set. In this way, an avatar instructor with the appearance and voice preferred by the student 90 is set.

The language system is information that indicates the language system used by the avatar instructor for lectures and calls, and is set based on one or more of the origin, preference, and desired instructor dialect in the student information D10 of the student 90. There are no particular limitations on the method of setting the language system, but for example, if the dialect of the student 90's hometown is a widely known dialect such as Nagoya dialect, Hiroshima dialect, Kansai dialect, or Hakata dialect, that dialect may be set as the language system used by the avatar instructor. Also, if there is no widely known dialect in the student 90's hometown, the desired instructor dialect in the student information D10 may be set as the language system used by the avatar instructor. Also, if the student's preferences include animals such as dogs and cats that are widely known for their language systems that use onomatopoeic or mimetic words such as "It's sunny today, woof" and "It's raining today, meow" (hereinafter also referred to as "animal language") in the endings, etc., that animal language may be set as the language system used by the avatar instructor. Furthermore, it is also possible to set a language system that mixes dialects and animal languages.

The lecture style is information that indicates the style of lecture given by the avatar lecturer, and examples include a scolding and encouraging style and a praising and encouraging style. There are no particular limitations on the method for setting the lecture style of the avatar lecturer, but as an example, the same lecture style as the desired lecture style in the student information D10 of student 90 is set. The lecture style desired by the student himself is not necessarily the lecture style that is suitable for that student, but it is sufficient to set the desired lecture style at least as an initial value.

In the example of Figure 5, the avatar instructor associated with student 90, whose student number is 1, is set to have an appearance of "Avatar A," a voice of "Voice a," a language system of "Nagoya dialect + cat language," and a teaching style of "scolding and encouraging."
Returning to FIG. 4, in step S102, the setting unit 11 sets the inner characteristic information D21 of the avatar instructor corresponding to the student 90.
FIG. 6 is a diagram showing a table of internal characteristic information.

In the table of the internal characteristic information D21, the student number of each student 90 and the internal characteristics of the avatar instructor are recorded in association with each other. The internal characteristics include place of origin, school, and preferences. The place of origin is information indicating the virtual place of origin of the avatar instructor, and is set to the same place of origin as the place of origin in the student information D10 of the student 90. The school of origin is information indicating the virtual school of the avatar instructor, and is set to the same school as the school of origin in the student information D10 of the student 90. The preferences are information indicating the virtual preferences of the avatar instructor, and are set to the same preferences as the preferences in the student information D10 of the student 90.
Returning to FIG. 4, in step S103, the setting unit 11 sets the profile information D23 of the avatar instructor corresponding to the student 90.
FIG. 7 is a diagram showing a table of profile information.

In the table of the profile information D23, the student number of each student 90 is recorded in association with the name, birthday, hometown, alma mater, and hobbies of the avatar instructor. The name is the virtual name of the avatar instructor, and as an example, a name that is predetermined in association with the appearance of the avatar instructor is set. The hometown is information indicating the virtual hometown of the avatar instructor, and is set to the same hometown as the hometown in the internal characteristic information D21. The birthday is the virtual birthday of the avatar instructor, and as an example, a birthday that is predetermined in association with the appearance of the avatar instructor is set. The school is information indicating the virtual school of the avatar instructor, and is set to the same school as the school of the avatar instructor in the internal characteristic information D21. The hobbies are information indicating the virtual hobbies of the avatar instructor, and are set to the same content as the preferences in the internal characteristic information D21.

1, the generation unit 12 generates text data of content to be provided to the students 90 based on the features set in the external feature information D22 and/or the features set in the internal feature information D21. The output unit 13 outputs video and audio of the content based on the avatar lecturer having the features set in the external feature information D22 by the setting unit 11 and the text data generated by the generation unit 12. The content includes a lecture and accompanying calls to the student 90.
8 is a flowchart of the lecture providing process. The lecture providing process is executed by the generating unit 12 and the output unit 13.

In step S201, the generation unit 12 acquires language system information from external feature information D22 of the avatar lecturer corresponding to the student 90 to whom the lecture is to be provided. Then, in step S202, the generation unit 12 converts the text data of the script data D30 corresponding to the lecture to be provided into the acquired language system. Specifically, the generation unit 12 includes a natural language generation artificial intelligence, and converts the text data of the script data D30 into a desired language system by the natural language generation artificial intelligence. For example, a script for lecture content using Nagoya dialect and cat language can be generated by providing the natural language generation artificial intelligence with a prompt instructing conversion to a way of speaking using Nagoya dialect and cat language, and the script data D30.

Next, in step S203, the output unit 13 causes the avatar lecturer, whose appearance and voice are set in the external characteristic information D22 corresponding to the student 90, to read aloud the text data of the script converted into the desired language system, and outputs the video and audio of the lecture via the browser 33 of the information terminal 30.

Figure 9 is a flowchart of the encouraging process. The encouraging process is a process of encouraging students 90 before, after, or during a lecture in order to improve their motivation based on their learning and grades. The encouraging process is executed by the generation unit 12 and the output unit 13.

In step S301, the generation unit 12 acquires external feature information D22 of the avatar instructor corresponding to the student 90. Next, in step S302, the generation unit 12 generates encouraging words for the student 90 based on the state of learning and grades of the student 90 and the lecture style of the avatar instructor. For example, if the student 90 has good test results but is slow to progress in learning, and the lecture style of the avatar instructor is to scold and encourage the student, a word of encouragement may be generated that scolds the student for the slow progress and encourages the student to work hard to advance in learning. If the student 90 has good test results but is slow to progress in learning, and the lecture style of the avatar instructor is to praise and encourage the student, a word of encouragement may be generated that praises the student's test results and encourages the student to progress. The method of creating the word of encouragement is not particularly limited, and may be, for example, a method of generating the word of encouragement using natural language generation artificial intelligence, or a method of selecting a word that meets the conditions from among words that have been created and registered in advance.

Next, in step S303, the generation unit 12 converts the generated prompt into the language system of the avatar instructor. Furthermore, in step S304, the output unit 13 causes the avatar instructor, whose appearance and voice are set in the external characteristic information D22, to read out the prompt converted into the desired language system, and outputs video and audio of the lecture via the browser 33 of the information terminal 30.

Figure 10 is a flowchart of the preference calling process. The preference calling process is a process for calling out to the students 90 before, after, or during a lecture in order to motivate the students 90 or to change their mood. The preference calling process is a process executed by the generation unit 12 and the output unit 13.

In step S401, the generation unit 12 acquires preference information registered in the student information D10 of the student 90 or the inner characteristic information D21 of the avatar instructor corresponding to the student 90. The preference registered in the student information D10 of the student 90 matches the preference registered in the inner characteristic information D21 of the avatar instructor corresponding to the student 90.
Next, in step S402 , the generation unit 12 acquires external feature information D22 of the avatar instructor corresponding to the student 90 .

Next, in step S403, the generation unit 12 generates prompts for the student 90 on topics related to the student's preferences based on the student's preferences. For example, if the student 90 and the avatar instructor like cats, a prompt on a topic related to cats is generated. There are no particular limitations on the method of generating the prompts, and the prompts may be generated using a natural language generation artificial intelligence, or may be selected from those that have been created and registered in advance that meet the conditions.

Next, in step S404, the generation unit 12 converts the generated prompt into the language system of the avatar instructor. Furthermore, in step S405, the output unit 13 causes the avatar instructor, whose appearance and voice are set in the external feature information D22, to read out the prompt converted into the desired language system, and outputs video and audio of the lecture via the browser 33 of the information terminal 30.

Figure 11 is a flowchart of the additional commentary insertion process. The additional commentary insertion process is executed by the generation unit 12 and the output unit 13. The additional commentary insertion process is a process for inserting additional commentary such as specific examples and analogies into a lecture. In this additional commentary insertion process, additional commentary such as specific examples and analogies that are associated with interests such as hobbies that are common to the student 90 and the avatar lecturer is inserted into the lecture. The additional commentary insertion process is executed by the generation unit 12 and the output unit 13.

In step S501, the generation unit 12 acquires preference information registered in the student information D10 of the student 90 or the inner characteristic information D21 of the avatar instructor corresponding to the student 90. The preference registered in the student information D10 of the student 90 matches the preference registered in the inner characteristic information D21 of the avatar instructor corresponding to the student 90.
Next, in step S502, the generation unit 12 acquires external feature information D22 of the avatar instructor corresponding to the student 90.

Next, in step S503, the generation unit 12 generates additional commentary that supplements the content of the lecture based on the preferences of the student 90 by linking it to the preferences of the student 90. For example, if the student 90 and the avatar lecturer like cats, an additional commentary associated with cats is generated. The method of creating the call is not particularly limited, and may be, for example, a method of generating the call using natural language generation artificial intelligence, or a method of selecting a call that meets the conditions from among previously created and registered calls.

Next, in step S504, the generation unit 12 converts the generated call-out into the language system of the avatar instructor. Furthermore, in step S505, the output unit 13 causes the avatar instructor, whose appearance and voice are set in the external feature information D22, to read out the call-out converted into the desired language system, and outputs video and audio of the lecture via the browser 33 of the information terminal 30.
Second Embodiment

The learning support system of the second embodiment has the same functions as the first embodiment, and in addition, it acquires the status of students attending a lecture from images and audio of the students, and evaluates and corrects the avatar instructor based on the acquired information on the status of the students, thereby improving the effectiveness of the avatar instructor.

The learning support system of the second embodiment has the same configuration and operation as that of the first embodiment, so the following mainly describes the differences between the first embodiment and the second embodiment.

Figure 12 is a block diagram showing the functional configuration of a learning support system according to the second embodiment. The hardware configuration of the learning support system of the second embodiment is similar to that of the first embodiment shown in Figure 2.

Referring to FIG. 12, in the second embodiment, the learning support system 10 has a concentration determination unit 16 in addition to a setting unit 11, a generation unit 12, an output unit 13, a profile presentation unit 14, and a memory unit 15. The concentration determination unit 16, like the setting unit 11, the generation unit 12, the output unit 13, and the profile presentation unit 14, is realized by a processing device 21, which is a processor that executes the processing of a software program.

In this embodiment, the concentration determination unit 16 determines the degree to which the student 90 is concentrating on the lecture based on an image acquired by the camera 31 of the information terminal 30 to capture the student 90, and the setting unit 11 controls the external characteristics of the avatar instructor corresponding to the student 90 based on the result of the determination. The method of concentration determination is not particularly limited, but for example, the student 90 may be determined to be concentrating if all the following conditions are met: the student 90 is in front of the terminal, has a good posture, has his/her eyes open, and is looking at the screen. Furthermore, the concentration determination unit 16 may use the voice of the student 90 acquired by the microphone 32 of the information terminal 30 in addition to the image to determine the degree to which the student 90 is concentrating on the lecture.

The concentration judgment unit 16 judges the extent to which the student 90 is concentrating on the lecture based on images acquired by the camera 31 of the information terminal 30 to photograph the student 90, and records concentration judgment result information based on the results of that judgment.
FIG. 13 is a diagram showing an example of concentration determination result information.

Based on the images of the students 90 attending the lecture, the concentration determination unit 16 determines the concentration level on a five-point scale from "1" to "5," with the higher the number, the higher the concentration level, calculates the weekly average, and records the weekly concentration level in chronological order in the concentration determination result information D41. There are no particular limitations on the method of calculating the weekly concentration level. For example, the concentration determination unit 16 may calculate and accumulate the concentration level for each lecture, which is the concentration level for that lecture, and calculate the weekly average of the accumulated concentration levels for each lecture to set it as the weekly concentration level. There are also no particular limitations on the method of calculating the concentration level for each lecture. For example, the concentration determination unit 16 may determine whether the students are concentrating based on the images at regular intervals, record a concentration flag indicating whether they are concentrating or not, and perform a five-point evaluation based on the ratio of the number of times they were concentrating and not concentrating during the lecture, and set the evaluation result as the concentration level for each lecture.

In the example of Figure 13, the weekly concentration rate for the first week is "4.1", the weekly concentration rate for the second week is "4.0", the weekly concentration rate for the third week is "4.1", the weekly concentration rate for the fourth week is "3.5", and the weekly concentration rate for the fifth week is "3.2".

The concentration determination unit 16 further calculates the average concentration level for each lecture or each week for the multiple students 90 for each avatar instructor and for each academic level of the students 90, and records this in the evaluation information as an evaluation value.
FIG. 14 is a diagram illustrating an example of evaluation information.

In the evaluation information D42, evaluation values are recorded in association with the number, appearance, voice, and lecture style of each avatar lecturer. The number is a number that identifies the avatar lecturer. The appearance indicates the appearance of the avatar lecturer. The voice indicates the voice of the avatar lecturer. The lecture style indicates the lecture style of the avatar lecturer. For example, the avatar lecturer numbered "001" has the appearance of "Avatar A", the voice "Voice a", and the lecture style of "scolding and encouraging".

The evaluation value of the avatar instructor is calculated for each academic level of the student. The evaluation value for each academic level is the average weekly concentration obtained from students of that academic level. In the evaluation information D42, as an example, the academic levels of students 90 are divided into four levels, from "A" to "D", based on test results. By calculating the evaluation value for each academic level, it is possible to know which avatar instructor is highly rated by students of which academic level.

Figure 15 is a flowchart of the lecturer control process. The lecturer control process is a process for controlling the characteristics of an avatar lecturer who gives a lecture to students 90 so that they can be changed. The lecture control process is a process that is periodically and repeatedly executed by the setting unit 11.

In step S601, the setting unit 11 acquires concentration judgment result information D41. Furthermore, in step S602, the setting unit 11 judges whether or not the avatar instructor needs to be modified based on the concentration judgment result information D41. The method of judgment is not particularly limited, but for example, if the weekly concentration level falls below a predetermined threshold value for a predetermined number of consecutive weeks, it may be judged that the avatar instructor needs to be modified. Also, if the weekly concentration level shows a downward trend for a predetermined number of consecutive weeks, it may be judged that the avatar instructor needs to be modified.

If it is determined that the avatar instructor does not need to be modified, the setting unit 11 ends the series of processes. On the other hand, if it is determined that the avatar instructor needs to be modified, the setting unit 11 obtains from the student 90 in step S603 whether or not the student 90 wishes to change the avatar instructor by asking the student 90, and determines whether or not the student wishes to change the avatar instructor in step S604.

If the student 90 does not wish to modify the avatar instructor, the setting unit 11 ends the series of processes. On the other hand, if the student 90 wishes to modify the avatar instructor, the setting unit 11 modifies at least one of the external characteristics or internal characteristics of the avatar instructor in step S605. There is no particular limitation on the method of modifying the characteristics of the avatar instructor, but for example, the setting unit 11 may modify the characteristics of the avatar instructor based on the evaluation information D42 so that the concentration of the student 90 is increased more than the current characteristics of the avatar instructor.
<Additional Notes>

Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and these embodiments may be used in combination or some of the configurations may be changed within the scope of the technical concept of the present invention.
The above-described embodiment includes the following features. However, the features included in the above-described embodiment are not limited to the following features.
(Item 1)

A learning support system that provides students with content related to a lecture, which is composed of video and audio, by an avatar lecturer, which is a virtual person modeled after a lecturer, includes: a storage unit that stores student information, which is information about the student; a setting unit that sets external feature information indicating the external features of an avatar lecturer who will lecture the student and internal feature information indicating the internal features of the avatar lecturer, based on the student information of the student; a generation unit that generates text data of content to be provided to the student, based on the features indicated in the external feature information and/or the features indicated in the internal feature information; an output unit that outputs content in which the avatar lecturer, having set the features indicated in the external feature information, utters the text data; and a profile presentation unit that presents the student with profile information of the avatar lecturer based on the internal feature information. This makes it possible to provide content related to a lecture by an avatar lecturer suitable for each student by adapting the external and internal features of the avatar lecturer to each student.
(Item 2)

In the learning support system described in item 1, the student information includes student preference information indicating what the student likes, the internal characteristic information includes instructor preference information indicating what the avatar instructor likes, and the setting unit sets the instructor preference information based on the student preference information. This makes it possible to match the preferences of the instructor and the student by setting the instructor's preferences in association with the student's preferences, thereby increasing the student's interest in the instructor and the lecture.
(Item 3)

In the learning support system described in item 2, the setting unit sets the same things as those indicated in the student preference information in the lecturer preference information, the generation unit generates text data of greetings on topics related to the things set in common in the student preference information and the lecturer preference information, and the output unit outputs the greetings before, during, or after a lecture. In this way, by matching the lecturer's preferences with the students' preferences and greetings on the common topic, it is possible to increase the students' liking for the lecturer and their interest in the lecture.
(Item 4)

In the learning support system described in item 2, the setting unit sets the same things as those indicated in the student preference information in the lecturer preference information, the generation unit generates text data of additional commentary associated with each of the things set in common in the student preference information and the lecturer preference information, and the output unit inserts the additional commentary into the lecture and outputs it. In this way, by matching the lecturer's preferences with those of the students and providing specific examples and analogies related to the common preferences, it is possible to increase the students' interest in and understanding of the lecture.
(Item 5)

In the learning support system described in item 1, the student information includes student hometown information indicating the student's hometown, the internal characteristic information includes instructor hometown information indicating the virtual hometown of the avatar instructor, the setting unit sets the same hometown as the hometown indicated in the student hometown information to the instructor hometown information, and the profile presenting unit presents the hometown indicated in the instructor hometown information to the student. In this way, by matching the instructor's hometown with the student's hometown and presenting this to the student, the student's favorable impression of the instructor and interest in the lecture can be increased.
(Item 6)

In the learning support system described in item 5, the internal characteristic information includes language system information indicating a language system spoken by the avatar instructor, the setting unit sets the language system information to a language system used in the birthplace set in the instructor's birthplace information, and the generation unit generates text data in the language system indicated in the language system information. By making the instructor's way of speaking in the language system (dialect) of the birthplace common to the students and the instructor, it is possible to increase the students' liking for the instructor and their interest in the lecture.
(Item 7)

In the learning support system described in item 1, the student information includes student school information indicating the student's alma mater, the inner characteristic information includes instructor school information indicating the virtual alma mater of the avatar instructor, the setting unit sets the same school as the alma mater indicated in the student school information in the instructor school information, and the profile presenting unit presents the school indicated in the instructor school information to the student. According to this, by setting the instructor as an alumni of the same school as the student, it is possible to increase the student's favorable impression of the instructor and interest in the lecture, and to encourage the student to study.
(Item 8)

In the learning support system described in item 1, the external feature information includes instructor appearance information indicating the appearance of the avatar instructor and instructor voice information indicating the voice of the avatar instructor, and the setting unit specifies the appearance and voice preferred by the student based on the student information, sets the appearance to the instructor appearance information, and sets the voice to the instructor voice information. In this way, by matching the instructor's appearance and voice to the student's preferences, the student's interest in the instructor can be increased and understanding of the lecture can be improved.
(Item 9)

In the learning support system described in item 1, the storage unit further stores script data, which is text data created in advance as a script for the lecture, the generation unit generates converted text data by converting the text data of the script data based on the characteristics indicated in the internal characteristic information, and the output unit outputs a read-aloud voice of the converted text data as the voice for the lecture. According to this, a basic script is prepared in advance, and the text of the script is converted into text having the characteristics of an avatar instructor suited to the students and output as voice, so that a lecture can be provided by flexibly adapting the speaking style of the instructor to each of a variety of students. There is no need to prepare a large number of scripts.
(Item 10)

In the learning support system described in item 9, the generation unit includes a natural language generation artificial intelligence that generates text according to instructions. This makes it possible to easily create content tailored to students without the need to prepare a large number of scripts.
(Item 11)

In the learning support system according to item 10, the generation unit provides the natural language generation artificial intelligence with a prompt indicating a predetermined instruction and the script data, thereby causing the natural language generation artificial intelligence to generate the converted text data.
(Item 12)

The learning support system described in item 1 further includes a concentration determination unit that measures the degree to which the student is concentrating on the lecture for each predetermined period based on images acquired to photograph the student attending the lecture and records the degree of concentration for each period in chronological order, and the setting unit modifies at least one of the external characteristics or internal characteristics of the avatar instructor based on the degree of concentration for each period. In this way, the characteristics of the avatar instructor are modified based on the degree of concentration and its time transition, making it possible to maintain a high level of interest in the lecture by the instructor.
(Item 13)

In the learning support system according to item 12, the setting unit determines whether or not the avatar instructor needs to be modified based on a time series progression of the degree of concentration for each period.
(Item 14)

In the learning support system described in item 12, the concentration determination unit calculates an evaluation value for each avatar instructor based on the concentration levels of a plurality of students, and the setting unit modifies the avatar instructor based on the evaluation value so that the students' concentration levels are higher than that of the current avatar instructor. Since the characteristics of the instructor are controlled so as to increase the students' concentration levels, it is possible to control or maintain high interest in the instructor's lecture.
(Item 15)

A learning support system for providing students with lecture-related content consisting of video and audio by an avatar instructor, which is a virtual person modeled after an instructor. The computer stores student information, which is information about the student, and sets external characteristic information indicating the external characteristics of the avatar instructor who will lecture the student and internal characteristic information indicating the internal characteristics of the avatar instructor based on the student information of the student. It generates text data of content to be provided to the student based on the characteristics indicated in the external characteristic information and/or the characteristics indicated in the internal characteristic information. It outputs content in which the avatar instructor, whose characteristics are set in the external characteristic information, speaks out the text data, and presents profile information of the avatar instructor based on the internal characteristic information to the student.

10...Learning support system, 11...Setting unit, 12...Generation unit, 13...Output unit, 14...Profile presentation unit, 15...Memory unit, 16...Concentration judgment unit, 20...Server, 21...Processing device, 22...Main memory, 23...Storage device, 24...Communication device, 25...Input device, 26...Display device, 27...Bus, 30...Information terminal, 31...Camera, 32...Microphone, 33...Browser

Claims (13)

A learning support system that provides students with lecture content consisting of video and audio by an avatar lecturer, which is a virtual person modeled after a lecturer,
A storage unit for storing student information, which is information about the student;
a setting unit that sets external feature information indicating external features of an avatar lecturer who will lecture the students and internal feature information indicating internal features of the avatar lecturer based on student information of the students;
a generation unit that generates text data of content to be provided to the students based on the features indicated in the external feature information and/or the features indicated in the internal feature information;
an output unit that outputs content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
a profile presenting unit that presents profile information of the avatar instructor based on the inner characteristic information to the student;
having
The student information includes student preference information indicating what the student likes,
The internal characteristic information includes instructor preference information indicating things that the avatar instructor likes,
The setting unit sets the instructor preference information based on the student preference information .
Learning support system. The setting unit sets the same things as those indicated in the student preference information in the instructor preference information,
the generating unit generates text data of a prompt on a topic related to a thing commonly set in the student preference information and the lecturer preference information;
The output unit outputs the call before, during, or after a lecture.
The learning support system according to claim 1 . The setting unit sets the same things as those indicated in the student preference information in the instructor preference information,
the generating unit generates text data of additional commentary associated with each of the items commonly set in the student preference information and the lecturer preference information;
The output unit inserts the additional commentary into the lecture and outputs it.
The learning support system according to claim 1 . A learning support system that provides students with lecture content consisting of video and audio by an avatar lecturer, which is a virtual person modeled after a lecturer,
A storage unit for storing student information, which is information about the student;
a setting unit that sets external feature information indicating external features of an avatar lecturer who will lecture the students and internal feature information indicating internal features of the avatar lecturer based on student information of the students;
a generation unit that generates text data of content to be provided to the students based on the features indicated in the external feature information and/or the features indicated in the internal feature information;
an output unit that outputs content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
a profile presenting unit that presents profile information of the avatar instructor based on the inner characteristic information to the student;
having
The student information includes student hometown information indicating the student's hometown,
The internal characteristic information includes teacher hometown information indicating a virtual hometown of the avatar teacher,
The setting unit sets the same hometown as the hometown indicated in the student hometown information to the instructor hometown information,
The profile presenting unit presents the hometown indicated in the instructor's hometown information to the student .
Learning support system. The internal characteristic information includes language system information indicating a language system spoken by the avatar instructor;
The setting unit sets a language system used in the home town set in the teacher's home town information to the language system information,
the generation unit generates text data in a language system indicated in the language system information.
The learning support system according to claim 4 . A learning support system that provides students with lecture content consisting of video and audio by an avatar lecturer, which is a virtual person modeled after a lecturer,
A storage unit for storing student information, which is information about the students;
a setting unit that sets external feature information indicating external features of an avatar lecturer who will lecture the students and internal feature information indicating internal features of the avatar lecturer based on student information of the students;
a generation unit that generates text data of content to be provided to the students based on the features indicated in the external feature information and/or the features indicated in the internal feature information;
an output unit that outputs content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
a profile presenting unit that presents profile information of the avatar instructor based on the inner characteristic information to the student;
having
The student information includes student's alma mater information indicating the student's alma mater,
The internal characteristic information includes teacher alma mater information indicating a virtual alma mater of the avatar teacher,
The setting unit sets the same school as the student's school information in the lecturer's school information,
The profile presentation unit presents the school indicated in the instructor's alma mater information to the student .
Learning support system. A learning support system that provides students with lecture content consisting of video and audio by an avatar lecturer, which is a virtual person modeled after a lecturer,
A storage unit for storing student information, which is information about the student;
a setting unit that sets external feature information indicating external features of an avatar lecturer who will lecture the students and internal feature information indicating internal features of the avatar lecturer based on student information of the students;
a generation unit that generates text data of content to be provided to the students based on the features indicated in the external feature information and/or the features indicated in the internal feature information;
an output unit that outputs content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
a profile presenting unit that presents profile information of the avatar instructor based on the inner characteristic information to the student;
having
a concentration determination unit that measures the degree to which the students are concentrating on the lecture for each predetermined period based on images acquired to photograph the students attending the lecture, and records the degree of concentration for each period in chronological order;
The setting unit modifies at least one of an external feature or an internal feature of the avatar instructor based on the concentration degree for each period .
Learning support system. the setting unit determines whether or not the avatar instructor needs to be modified based on a time series progression of the concentration degree for each period.
The learning support system according to claim 7 . the concentration determination unit calculates an evaluation value for each avatar instructor based on the concentration levels of the multiple students;
the setting unit modifies the avatar instructor based on the evaluation value so that the degree of concentration of the student is higher than that of the current avatar instructor.
The learning support system according to claim 7 . A learning support method for providing a student with a lecture-related content including video and audio by an avatar lecturer, the avatar lecturer being a virtual person modeled after a lecturer, comprising:
The computer
storing attendee information which is information about the attendee;
based on student information of the student, external characteristic information indicating external characteristics of an avatar instructor who will lecture the student and internal characteristic information indicating internal characteristics of the avatar instructor;
generating text data of content to be provided to the students based on the characteristics indicated in the external characteristic information and/or the characteristics indicated in the internal characteristic information;
outputting content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
presenting profile information of the avatar instructor based on the internal characteristic information to the student;
In the method,
The student information includes student preference information indicating what the student likes,
The internal characteristic information includes instructor preference information indicating things that the avatar instructor likes,
The computer,
setting the instructor preference information based on the student preference information;
Learning support system. A learning support method for providing a lecture-related content, which is made up of video and audio, to a student by an avatar teacher, which is a virtual person imitating a teacher, comprising the steps of:
The computer
storing attendee information which is information about the attendee;
based on student information of the student, external characteristic information indicating external characteristics of an avatar instructor who will lecture the student and internal characteristic information indicating internal characteristics of the avatar instructor are set;
generating text data of content to be provided to the students based on the characteristics indicated in the external characteristic information and/or the characteristics indicated in the internal characteristic information;
outputting content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
presenting profile information of the avatar instructor based on the internal characteristic information to the student;
In the method,
The student information includes student hometown information indicating the student's hometown,
The internal characteristic information includes teacher hometown information indicating a virtual hometown of the avatar teacher,
The computer,
The same hometown as that indicated in the student hometown information is set in the instructor hometown information,
presenting the hometown indicated in the instructor's hometown information to the student;
Learning support methods. A learning support method for providing a lecture-related content, which is made up of video and audio, to a student by an avatar teacher, which is a virtual person imitating a teacher, comprising the steps of:
The computer
storing attendee information which is information about the attendee;
based on student information of the student, external characteristic information indicating external characteristics of an avatar instructor who will lecture the student and internal characteristic information indicating internal characteristics of the avatar instructor;
generating text data of content to be provided to the students based on the characteristics indicated in the external characteristic information and/or the characteristics indicated in the internal characteristic information;
outputting content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
presenting profile information of the avatar instructor based on the internal characteristic information to the student;
In the method,
The student information includes student's alma mater information indicating the student's alma mater,
The internal characteristic information includes teacher alma mater information indicating a virtual alma mater of the avatar teacher,
The computer,
The same school as the student's school information is set in the instructor's school information,
presenting the school indicated in the instructor's alma mater information to the student;
Learning support methods. A learning support method for providing a lecture-related content, which is made up of video and audio, to a student by an avatar teacher, which is a virtual person imitating a teacher, comprising the steps of:
The computer
storing attendee information which is information about the attendee;
based on student information of the student, external characteristic information indicating external characteristics of an avatar instructor who will lecture the student and internal characteristic information indicating internal characteristics of the avatar instructor;
generating text data of content to be provided to the students based on the characteristics indicated in the external characteristic information and/or the characteristics indicated in the internal characteristic information;
outputting content in which an avatar instructor having characteristics set in the external characteristic information utters the text data;
presenting profile information of the avatar instructor based on the internal characteristic information to the student;
In the method,
The computer,
Measure the degree to which the students are concentrating on the lecture for each predetermined period based on images acquired to photograph the students attending the lecture, and record the degree of concentration for each period in chronological order;
modifying at least one of the external characteristics and the internal characteristics of the avatar instructor based on the concentration degree for each period;
Learning support methods.

Images