Artificial intelligence system and program of artificial intelligence system

Abstract

To provide an artificial intelligence system and a program which allow a user to smoothly communicate with pre-trained artificial intelligence means without requiring the user to input information on a keyboard or reading information with their eyes.SOLUTION: An artificial intelligence system 100 includes a user terminal 110, a first server 120, and a second server 130. When voice is input on a talk screen 112 of a browser 111 of the user terminal, the browser 111 converts the voice into text data, inputs the text data to an input field in the talk screen 112, and transmits the text data, as instruction command information PT, to a bot 121 of the first server. The bot 121 generates text data of a content based on the instruction command information, using pre-trained artificial intelligence means 131 of the second server, and transmits the generated text data GT obtained by the execution to the user terminal 110. The browser 111 of the user terminal 110 outputs the content of the generated text data, with voice, using a speaker.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an artificial intelligence system comprising a user terminal, a first server having a bot, and a second server having a pre-trained artificial intelligence means, in which the user terminal can freely communicate with the pre-trained artificial intelligence means via the bot, and a program for the system. [Background technology]

[0002] Conventionally, there is known a program for a chat system having a content inquiring step in which a chatbot inquires on a chat screen of the chat user's terminal about what the chat user would like to be reminded of, a content registering step in which the chatbot registers chat data received from the chat user's terminal after the content inquiring step as reminder content in a database, a date and time inquiring step in which the chatbot inquires on a chat screen of the chat user's terminal about the date and time at which the chat user would like to be notified of the content, a date and time registering step in which the chatbot registers chat data received from the chat user's terminal after the date and time inquiring step as reminder date and time in a database, a reminder date and time determination step in which the chatbot accesses a date and time control unit of a chat server to determine whether the current date and time has reached the registered reminder date and time, and a transmission notification step in which the chatbot sends the reminder content registered in the database to the chat user's terminal when it has determined that the current date and time has reached the registered reminder date and time (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-159469 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the programs of the conventional chat systems described above were configured so that the user could send text data by typing on a keyboard, and notifications from the chatbot were displayed as text data, which required the user to type text on a keyboard, which was cumbersome. Furthermore, notifications were only displayed as text, which required the user to read them visually, which could be annoying.

[0005] Therefore, the present invention solves the problems of the prior art as described above. In other words, the object of the present invention is to provide an artificial intelligence system and a program thereof that can smoothly interact with pre-trained artificial intelligence means without the user needing to input data using a keyboard or read with their eyes. [Means for solving the problem]

[0006] The invention of claim 1 is an artificial intelligence system comprising a user terminal, a first server having a bot, and a second server having a pre-trained artificial intelligence means, in which the user terminal can freely communicate with the pre-trained artificial intelligence means via the bot, wherein the browser of the user terminal communicates with the first server to display a talk screen, and when a predetermined operation is performed on the talk screen, the microphone of the user terminal becomes active and when voice input is performed, the browser converts the input voice into text and enters the text data into an input field on the talk screen and sends it to the bot of the first server as instruction command information, and the bot of the first server uses the pre-trained artificial intelligence means of the second server to generate text data based on the instruction command information and sends the generated text data obtained by the execution to the user terminal, and the user terminal displays the received generated text data on the talk screen of the browser, and the browser outputs the content of the generated text data aloud using a speaker, thereby solving the above-mentioned problem.

[0007] The invention of claim 2 further solves the above-mentioned problem by configuring the artificial intelligence system as described in claim 1, in addition to the configuration, the talk screen is provided with input and output language selection, and when voice input is received, the browser recognizes the input voice as the selected language, converts it to text, enters the text data into an input field on the talk screen, and transmits it to the bot of the first server as instruction command information, and when the user terminal receives the generated text data, displays the received generated text data on the talk screen of the browser, and the browser recognizes the content of the generated text data as the selected language and outputs it as voice using a speaker.

[0008] The invention of claim 3 further solves the above-mentioned problem by, in addition to the configuration of the artificial intelligence system described in claim 2, providing a voice selection option for audio output on the talk screen, and configuring the browser to output the contents of the generated text data using a speaker in the voice selected by the user.

[0009] The invention of claim 4 further solves the above-mentioned problem by having, in addition to the configuration of the artificial intelligence system described in any one of claims 1 to 3, the talk screen has a talk display area that displays instruction and command information and generated text data which are the interaction between the user and the pre-trained artificial intelligence means, and a prompt list display area that displays pre-registered standard instruction and command information items, and when an instruction and command information item is selected, the content of the instruction and command information of the selected instruction and command information item is entered into an input field on the talk screen and transmitted to the pre-trained artificial intelligence means of the second server via the bot of the first server.

[0010] The invention of claim 5 further solves the above-mentioned problem by, in addition to the configuration of the artificial intelligence system described in claim 4, allowing command information items to be freely added and set in the prompt list display area by user operation and administrator operation, command information items set by user operation are displayed in a user-defined command information item group arranged above the common command information item group set by administrator operation, and the command information item that was most recently selected is rearranged and displayed at the top in the user-defined command information item group.

[0011] The invention of claim 6 is a program for an artificial intelligence system in which a user terminal can freely communicate with a pre-trained artificial intelligence means of a second server via a bot of a first server, comprising: a predetermined operation presence / absence determination step in which a browser of the user terminal communicates with the first server to display a talk screen and determines whether a predetermined operation has been performed on the talk screen; a microphone activation step in which a microphone of the user terminal is activated when the predetermined operation has been performed; a voice input presence / absence determination step in which a voice input presence / absence determination step in which, when the voice input has been performed, the browser converts the input voice into text, inputs the text data into an input field on the talk screen, and transmits the text data to the bot of the first server as command information; a text data generation step in which the bot of the first server uses the pre-trained artificial intelligence means of the second server to generate text data based on the command information; a text data transmission step in which the bot of the first server transmits the generated text data obtained by execution to the user terminal; and a display / audio output step in which the user terminal displays the received generated text data on the talk screen of the browser and the browser outputs the content of the generated text data by using a speaker. This solves the above-mentioned problem. Effect of the Invention

[0012] The artificial intelligence system of the present invention comprises a user terminal, a first server having a bot, and a second server having a pre-trained artificial intelligence means, not only enabling the user terminal to communicate with the pre-trained artificial intelligence means via the bot, but also achieving the following unique effects.

[0013] According to the artificial intelligence system of the invention of claim 1, the user simply performs a specified operation on the talk screen to input voice, and the answer from the pre-trained artificial intelligence means is output as voice, eliminating the need for the user to input text using a keyboard, thereby avoiding the inconvenience. Furthermore, the answers are outputted aloud, eliminating the need for the user to read them visually, which can be annoying. In other words, voice input is followed by voice responses, allowing smooth communication with pre-trained artificial intelligence using the mouth and ears.

[0014] According to the artificial intelligence system of the invention of claim 2, in addition to the effect achieved by the invention of claim 1, communication can be mainly carried out via voice in the selected language, thereby enabling the user to speak multiple languages.

[0015] According to the artificial intelligence system of the invention of claim 3, in addition to the effect achieved by the invention of claim 2, the voice and speaking style can be freely set to the user's preferred voice and speaking style, and the user can audibly hear the contents of the generated text data as an answer from the pre-trained artificial intelligence means in audio output in the user's preferred voice and speaking style.

[0016] According to the artificial intelligence system of the invention of claim 4, in addition to the effect achieved by any one of the inventions of claims 1 to 3, the contents of pre-registered instruction and command information regarding the content before and after the exchange are transmitted simply by a selection operation, so that the user can give, for example, detailed instructions and commands that have been pre-registered, without having to input text from a keyboard. Furthermore, you can get answers to your detailed instructions and commands.

[0017] According to the artificial intelligence system of the invention of claim 5, in addition to the effect of the invention of claim 4, the most recently used instruction command information item is displayed at the top, and frequently used instruction command information items are displayed at the top, making it easier for each user to use.

[0018] According to the program of the artificial intelligence system of the invention of claim 6, similar to the effect achieved by the invention of claim 1, the answer from the pre-trained artificial intelligence means is output by voice simply by performing a specified operation on the talk screen and inputting voice, so that the user does not need to input text by keyboard, thereby avoiding the inconvenience. Furthermore, the answers are outputted aloud, eliminating the need for the user to read them visually, which can be annoying. In other words, voice input is followed by voice responses, allowing smooth communication with pre-trained artificial intelligence using the mouth and ears. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a diagram showing the concept of an artificial intelligence system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a chart showing an example of the operation of the artificial intelligence system according to the embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram showing an example of a talk screen of the artificial intelligence system according to an embodiment of the present invention. [Figure 4] 1A and 1B are diagrams showing examples of a voice input screen of an artificial intelligence system according to an embodiment of the present invention. [Diagram 5] A figure showing generated text data being displayed on the talk screen of an artificial intelligence system according to an embodiment of the present invention, and audio being output by a browser. [Figure 6] 1A and 1B are diagrams showing how to register a prompt on the talk screen of an artificial intelligence system according to an embodiment of the present invention. [Figure 7] FIG. 13 is a diagram showing a state in which a prompt is selected on the talk screen of the artificial intelligence system according to an embodiment of the present invention. [Figure 8]1A and 1B are diagrams showing examples of the chat screen of a smartphone user terminal of an artificial intelligence system according to an embodiment of the present invention. [Figure 9] 1A and 1B are diagrams showing the state when voice input is made on the chat screen of a smartphone user terminal of an artificial intelligence system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] The artificial intelligence system of the present invention comprises a user terminal, a first server having a bot, and a second server having a pre-trained artificial intelligence means, and the browser of the user terminal communicates with the first server to display a talk screen, and when a specified operation is performed on the talk screen, the microphone of the user terminal becomes active, and when voice input is performed, the browser converts the input voice into text and enters the text data into an input field on the talk screen and sends it to the bot of the first server as instruction command information, and the bot of the first server uses the pre-trained artificial intelligence means of the second server to generate text data based on the instruction command information and sends the generated text data obtained by execution to the user terminal, and the user terminal displays the received generated text data on the talk screen of the browser, and the browser outputs the content of the generated text data as voice using a speaker. Therefore, the user does not need to input text on a keyboard, the user does not need to read with his or her eyes, and the specific embodiment of the system may be any one as long as the user can smoothly communicate with the pre-trained artificial intelligence means using his or her mouth and ears. The program for the artificial intelligence system of the present invention includes a predetermined operation presence / absence determination step in which a browser of a user terminal communicates with a first server to display a talk screen and determines whether or not a predetermined operation has been performed on the talk screen; a microphone activation step in which a microphone of the user terminal is activated when the predetermined operation has been performed; a voice input presence / absence determination step in which a voice input presence / absence determination step in which, when the voice input has been performed, the browser converts the input voice into text, inputs the text data into an input field on the talk screen, and transmits the text data as instruction command information to a bot of the first server; and a command transmission step in which the bot of the first server uses the pre-trained artificial intelligence means of the second server to transmit the instruction command. The specific implementation of the method may be any as long as it has a text data generating step of generating text data based on command information, a text data transmitting step of a bot of the first server transmitting the generated text data obtained by the execution to a user terminal, and a display / audio output step of the user terminal displaying the received generated text data on the talk screen of the browser and the browser outputting the contents of the generated text data as audio using a speaker, thereby eliminating the need for the user to input text using a keyboard, eliminating the need for the user to read with their eyes, and allowing smooth communication with the pre-trained artificial intelligence means using their mouths and ears.

[0021] For example, the user terminal may be a desktop personal computer terminal, a notebook personal computer terminal, a smartphone terminal, a tablet terminal, or the like that sends and receives information, and may be any terminal that can be freely connected to a server via a communication network including a wide area network, such as the Internet, a local network, or a telephone line. Furthermore, the first server and the second server may each be a single server or multiple servers on a cloud. The pre-trained artificial intelligence means may be composed of an interactive model, also known as a large-scale language model, such as ChatGPT (Generative Pre-trained Transformer) (hereinafter referred to as ChatGPT), and may be configured to be installed on one server or multiple servers on the cloud. EXAMPLES

[0022] An artificial intelligence system 100 according to an embodiment of the present invention will now be described with reference to FIGS. Here, FIG. 1 is a diagram showing the concept of an artificial intelligence system 100 which is an embodiment of the present invention, FIG. 2 is a chart diagram showing an example of the operation of the artificial intelligence system 100 which is an embodiment of the present invention, FIG. 3 is a diagram showing an example of a talk screen 112 of the artificial intelligence system 100 which is an embodiment of the present invention, FIG. 4(A) is a diagram showing an example of a voice input screen 117 of the artificial intelligence system 100 which is an embodiment of the present invention, FIG. 4(B) is a diagram showing the appearance of the voice input screen 117 of the artificial intelligence system 100 which is an embodiment of the present invention, where the voice input is converted into text and the text data TX is displayed, FIG. 5 is a diagram showing the appearance of the generated text data GT being displayed on the talk screen 112 of the artificial intelligence system 100 which is an embodiment of the present invention, and the browser 111 outputting voice, and FIG. 6(A) is a diagram showing an example of a prompt registration window 118 displayed by the browser 111 when registering a prompt on the talk screen 112 of the artificial intelligence system 100 which is an embodiment of the present invention. FIG. 6(B) is a diagram showing an example of a prompt entered in the prompt registration window 118, FIG. 7 is a diagram showing a state in which a prompt is selected on the talk screen 112 of the artificial intelligence system 100 which is an embodiment of the present invention, FIG. 8(A) is a diagram showing an example of the talk screen 112 of the smartphone user terminal 110 of the artificial intelligence system 100 which is an embodiment of the present invention, FIG. 8(B) is a diagram showing a state in which the microphone becomes active when the microphone icon button 114b on the talk screen 112 shown in FIG. 8(A) is operated, FIG. 9(A) is a diagram showing a state in which voice is input on the talk screen 112 of the smartphone user terminal 110 of the artificial intelligence system 100 which is an embodiment of the present invention and the input voice is converted into text data TX, and FIG. 9(B) is a diagram showing a state in which generated text data GT is output as a voice in response to the instruction command information PT input on the talk screen 112 shown in FIG. 9(A).

[0023] As shown in Figure 1, an artificial intelligence system 100 embodying the present invention includes a user terminal 110, a first server 120 having a bot 121, and a second server 130 having a chat GPT, which is an example of a pre-trained artificial intelligence means 131. The user terminal 110 is configured to be able to communicate freely with chat GPT, which is an example of a pre-trained artificial intelligence means 131, via the bot 121. Specifically, the browser 111 of the user terminal 110 communicates with the first server 120 and displays a talk screen 112 . When a predetermined operation is performed on the talk screen 112, the microphone of the user terminal 110 becomes active.

[0024] Next, when voice input is made, the browser 111 converts the input voice into text, inputs the text data TX into the input field 114a on the talk screen 112, and transmits it to the bot 121 of the first server 120 as instruction command information PT. Then, the bot 121 of the first server 120 uses the chat GPT, which is an example of the pre-trained artificial intelligence means 131 of the second server 130, to generate text data based on the instruction command information PT.

[0025] Then, the bot 121 of the first server 120 transmits the generated text data GT obtained through the execution to the user terminal 110. Then, the user terminal 110 displays the received generated text data GT on the talk screen 112 of the browser 111. In addition, the browser 111 is configured to output the contents of the generated text data GT by voice using a speaker.

[0026] This allows a response from the chat GPT, which is an example of the pre-trained artificial intelligence means 131, to be output as voice simply by performing a predetermined operation on the talk screen 112 and inputting voice. As a result, users do not need to input text using a keyboard, which can be a hassle. In addition, the answer is output as voice. As a result, users do not need to read the text visually, which can be a hassle-free experience. That is, a voice response is given by inputting a voice. As a result, you can use your mouth and ears to smoothly communicate with Chat GPT, an example of a pre-trained artificial intelligence means 131. The user terminal 110 may be a smartphone.

[0027] Next, the operation of the artificial intelligence system 100 will be described in more detail. As shown in FIG. 2, in step S1, as a talk screen display step, the browser 111 of the user terminal 110 communicates with the first server 120 and displays the talk screen 112. More specifically, when the user terminal 110 accesses the first server 120 and makes an HTTP request, the first server 120 transmits data of the talk screen 112 to the user terminal 110 as an HTTP response.

[0028] Then, as shown in FIG. 3, the browser 111 of the user terminal 110 displays a talk screen 112. When user terminal 110 accesses first server 120, a user ID and a password may be input as a login operation.

[0029] Here, the talk screen 112 has, for example, a document item display area 113, a talk display area 114, and a prompt list display area 115. Of these, the document item display area 113 is provided to display a document item 113a. In addition, the talk display area 114 is configured to display interactions between the user and the chat GPT, which is an example of the pre-trained artificial intelligence means 131 of the second server 130, via the bot 121 of the first server 120.

[0030] The talk display area 114 is configured like a chat screen. Here, the chat talk screen refers to a screen that displays exchanges of text and file data with one or more partners (including the bot 121) in chronological order on a single screen. Furthermore, the prompt list display area 115 is provided to display prompt items 115b.

[0031] In step S2, as a predetermined operation presence / absence determination step, it is determined whether or not a predetermined operation has been performed on the talk screen 112. More specifically, as an example of a predetermined operation, the user terminal 110 determines whether or not the microphone icon button 114b arranged in the input field 114a provided in the talk display area 114 of the talk screen 112 has been operated. If it is determined that an operation has been performed, the process proceeds to step S3, whereas if it is determined that an operation has not been performed yet, step S2 is repeated.

[0032] Although the operation of the microphone icon button 114b has been described as an example of the predetermined operation, the present invention is not limited to this. For example, the user may right-click the mouse to display a menu and select "Turn on microphone" from the menu items, or may press a microphone button displayed on the display unit of user terminal 110.

[0033] In step S3, which is a microphone activation step, user terminal 110 activates the microphone of user terminal 110. As shown in FIG. 4(A), in this embodiment, as an example, the browser 111 also switches the display from the talk screen 112 to a voice input screen 117. The voice input screen 117 displays the message "Please speak." Here, as an example, it is assumed that the user inputs "What is the mileage?" by voice. Then, as shown in FIG. 4(B), the display of the microphone icon changes to indicate that the browser 111 is detecting the voice input through the microphone.

[0034] In step S4, as a voice input presence / absence determination step, browser 111 of user terminal 110 determines whether or not a voice input has been made. If it is determined that there has been a voice input, the process proceeds to step S5, and if it is determined that there has not been a voice input, step S4 is repeated.

[0035] In step S5, as an instruction command transmission step, the browser 111 converts the input voice into text, inputs the text data TX into the input field 114a on the talk screen 112, and transmits it to the bot 121 of the first server 120 as instruction command information PT. More specifically, as shown in FIG. 4B, for example, the input voice is converted into text and the text data TX, "What is the mileage?" is displayed on the voice input screen 117. Then, as shown in FIG. 5, the browser 111 switches the display from the voice input screen 117 to the talk screen 112, and inputs text data TX, such as “What is the mileage?”, into the input field 114a of the talk screen 112 and transmits it to the bot 121 of the first server 120. Although the browser 111 switches the display from the talk screen 112 to the voice input screen 117 and from the voice input screen 117 to the talk screen 112, the talk screen 112 may also function as the voice input screen 117 without switching.

[0036] In step S6, as a text data generation step, the bot 121 of the first server 120 generates text data based on the instruction command information PT using the chat GPT, which is an example of the pre-trained artificial intelligence means 131 of the second server 130. In order to prevent the main topic from shifting, the bot 121 may send the content specified as the main topic by user operation, the content of the previous exchange, or the content of the exchange two or three exchanges ago, together with the instruction command information PT to the chat GPT, which is an example of the pre-trained artificial intelligence means 131 of the second server 130, to generate text data. In step S7, as a text data transmission step, the bot 121 of the first server 120 transmits the generated text data GT obtained by the execution to the user terminal 110.

[0037] In step S8, as shown in FIG. 5, as a display / audio output step, the user terminal 110 displays the received generated text data GT in the talk display area 114 of the talk screen 112 of the browser 111. At the same time, the browser 111 outputs the contents of the generated text data GT as voice using the speaker of the user terminal 110.

[0038] As a result, as described above, a response from the chat GPT, which is an example of the pre-trained artificial intelligence means 131, is output as voice simply by performing a predetermined operation on the talk screen 112 and inputting voice. As a result, users do not need to input text using a keyboard, which can be a hassle. In addition, the answer is output as voice. As a result, users do not need to read the text visually, which can be a hassle-free experience. That is, a voice response is given by inputting a voice. As a result, you can use your mouth and ears to smoothly communicate with Chat GPT, an example of a pre-trained artificial intelligence means 131.

[0039] At each exchange in the talk display area 114 of the talk screen 112, a clipboard icon 114c and a speaker icon 114d are provided. Of these, when the clipboard icon 114c is operated, the contents of the exchanged instruction command information PT and generated text data GT are temporarily recorded in the storage unit of the user terminal 110110 as a so-called clipboard.

[0040] Furthermore, when the speaker icon 114d is operated, the contents of the generated text data GT, which is the exchange, are converted into voice data and output as voice using the speaker of the user terminal 110. This causes the audio to be output again. As a result, the user can once again listen to the contents of the generated text data GT with his / her ears, and can relisten to the parts he / she missed.

[0041] Furthermore, a configuration may be adopted in which options are provided for the voice of the audio output, and the audio output is performed in the voice selected by the user. For example, options such as male / female, ages in their teens and twenties, 30s and 40s, and 50s and 60s, accents (standard and dialects), and speaking speed may be provided. This allows users to freely set the voice and speaking style to their preference. As a result, the user can hear the contents of the generated text data GT as a response from chat GPT, an example of a pre-trained artificial intelligence means 131, in a voice output using the user's preferred voice or speaking style.

[0042] Furthermore, in this embodiment, as shown in FIG. 3 and FIG. 5, the talk screen 112 is provided with an option to select an input / output language. More specifically, talk screen 112 is provided with a language selection button 116 . When the language selection button 116 is operated, one of a plurality of languages ​​such as Japanese and English can be freely selected. Then, when voice input is made, the browser 111 recognizes the input voice as the selected language, converts it to text, inputs the text data TX into the input field 114a on the talk screen 112, and transmits it to the bot 121 of the first server 120 as instruction command information PT.

[0043] Furthermore, when the user terminal 110 receives the generated text data GT, the user terminal 110 displays the received generated text data GT on the talk screen 112 of the browser 111 . In addition, the browser 111 is configured to recognize the contents of the generated text data GT as the selected language and output the same by voice using a speaker. This allows communication to be mainly carried out through voice in the selected language. As a result, it is possible to accommodate users in multiple languages.

[0044] In this embodiment, as described above, the talk screen 112 has a talk display area 114 that displays instruction and command information PT and generated text data GT, which are the exchange between the user and the chat GPT, which is an example of a pre-trained artificial intelligence means 131, and a prompt list display area 115 that displays prompt items 115b, which are standard instruction and command information items that have been set and registered in advance.

[0045] As shown in FIG. 5, the prompt list display area 115 is provided with a "New Prompt" button 115a. When the "new prompt" button 115a is operated, the browser 111 pops up and displays a prompt registration window 118, as shown in FIG. 6(A). The prompt registration window 118 is provided with a title input field 118a and a prompt setting input field 118b.

[0046] As an example, assume that "Summary of previous text (300 characters)" is entered in the title entry field 118a, as shown in FIG. 6(B). Furthermore, as an example, assume that "#Command Please summarize the following text in 300 characters. #Environment Please write in a sales-like format. #Content [$$content$$]" is entered into the prompt setting input field 118b, and the save button is operated. Then, the input information is registered in the first server 120. Furthermore, as shown in FIG. 7, a new prompt item 115b "Summary of previous text (300 characters)" is added and displayed in the prompt list display area 115.

[0047] When a selection operation is performed on prompt item 115b, which is an instruction command information item, the contents of the instruction command information PT of prompt item 115b, which is the selected instruction command information item, are entered into input field 114a of talk screen 112 and transmitted to chat GPT, which is an example of pre-trained artificial intelligence means 131 of second server 130, via bot 121 of first server 120.

[0048] As a result, the contents of instruction command information PT that have been set and registered beforehand regarding the contents before and after the exchange can be transmitted simply by performing a selection operation. As a result, the user can issue detailed instructions, for example, that have been pre-registered, without having to input text from a keyboard. Furthermore, you can get answers to your detailed instructions and commands.

[0049] Furthermore, in this embodiment, in the prompt list display area 115, prompt items 115b, which are instruction command information items, can be freely added and set by user operations and administrator operations. Furthermore, the prompt item 115b, which is an instruction command information item set by a user operation, is displayed in a user-defined instruction command information item group 115d arranged above the common instruction command information item group 115c set by an administrator operation.

[0050] In the user-defined command information item group 115d, the prompt item 115b, which is the command information item most recently selected, is rearranged and displayed at the top. As a result, the prompt item 115b which is the most recently used instruction / command information item is displayed at the top, and the prompt item 115b which is the frequently used instruction / command information item is displayed at the top. As a result, it is possible to improve usability for each user.

[0051] As shown in Figures 6(A) and 6(B), the prompt setting input field 118b of the prompt registration window 118 can be configured to add not only command information about what to do, but also environmental information about the intended use and condition information such as character restrictions for the command.

[0052] This results in detailed instruction and command information PT. As a result, by selecting the prompt item 115b for the generated text data GT as an answer from the chat GPT, which is an example of the pre-trained artificial intelligence means 131, new generated text data GT can be obtained based on the previous generated text data GT. In addition, for generated text data GT as an answer from chat GPT, which is an example of a pre-trained artificial intelligence means 131, a prompt may be added to the instruction command information PT entered by the user through voice input or character input via a keyboard, and new generated text data GT may be obtained based on the previous generated text data GT.

[0053] In addition, prompt items 115b include a prompt to summarize the previous text in about 300 characters, a prompt to format the previous text for an email report, a prompt to format the previous text in Markdown format, a prompt to summarize the previous text in table format, a prompt to format the previous text in chat format, and a prompt to format the previous text in report format.

[0054] Among these, the prompt to summarize the previous sentence in about 300 characters is an instruction to summarize the previous sentence in about 300 characters. In addition, the prompt to format the previous text as an e-mail report instructs the user to prepare the previous text by dividing it into items such as a title, recipient's name, purpose, summary, detailed reasons, and issues. The prompt to convert the previous text to Markdown format is an instruction to rewrite the previous text according to the rules of Markdown format.

[0055] The prompt to summarize the previous sentence in a table format is an instruction to summarize the previous sentence in a table format. The prompt to format the previous sentence in chat format instructs the user to rephrase the previous sentence as a multi-party exchange of short one-liners. The prompt to write the previous text in report format is an instruction or command to write the previous text in a report format, dividing it into categories such as a title (large font size), date, purpose, results, comments, and future issues, in order to report the text to a superior.

[0056] Further, prompt items 115b include a prompt to do research (point out), a prompt to do research (gather information / put it in a table), a prompt to do research (gather information / summarize), a prompt for a contract (summary), a prompt for a contract (main points), a prompt for a contract (point out risks), a prompt for a contract (revisions), and a prompt for a contract (translation).

[0057] Among these, the prompt to research (point out) is an instruction or command to pick out specific points from a previously entered sentence, the contents of a URL, or the like. In addition, the prompt for research (collecting information / compiling in table format) is an instruction to compile predetermined content, such as previously entered text or content listed in a URL, in table format. A research (information gathering / summarization) prompt is an instruction or command to summarize a given piece of information, such as a previously entered sentence or the contents of a URL.

[0058] The prompt for contract (summary) is an instruction to summarize the specified contract contents, such as the contract (text) previously entered or the contract contents listed in the URL. The prompt for "contract (main points)" is an instruction to summarize the main points of a specified contract, such as a previously entered contract (text) or the contract contents listed in a URL. The prompt "Contract (risk indication)" is an instruction to indicate possible risk areas in the specified contract content, such as a previously entered contract (document) or the contract content listed in a URL. The prompt for the contract (revision proposal) is an instruction or command to display, for example, in table format, the pre-revision and proposed revisions in order to show the state before and after revisions to a specified contract content, such as a previously entered contract (text) or the contract content listed in a URL. The contract (translation) prompt is an instruction to translate a specific contract content, such as a previously entered contract (text) or the contract content listed in a URL.

[0059] When the arrow on the left side of each prompt item 115b is operated, the content entered in the prompt setting input field 118b is displayed below that item. In addition, when the "+" button on the right side of each prompt item 115b is operated, the content entered in the prompt setting input field 118b for that item is entered into the input field 114a of the talk display area 114 and is also sent to the bot 121 of the first server 120.

[0060] Also, assume that the "+New Document" displayed in the document item display area 113 is operated. Then, a new document item 113a is added and displayed, and the display of the exchange in the talk display area 114 is switched to the new one. Further, the document item 113 a is provided with a freely changeable item name, and is associated with the exchange in the talk display area 114 . When the selected document item is changed, the display of the exchange in the talk display area 114 is switched.

[0061] As shown in FIGS. 8(A) to 9(B), the user terminal 110 may be a smartphone. As shown in FIG. 8(A), a browser 111 of a user terminal 110, which is a smartphone, displays a talk screen 112. The smartphone version of the talk screen 112 is simpler than that of the PC version, and includes an input field 114a, a microphone icon button 114b, a send button, an icon of a bot 121, and a language selection button 116.

[0062] When the user taps on microphone icon button 114b provided in input field 114a of talk screen 112 as an example of a predetermined operation, the microphone of user terminal 110 becomes active. In addition, as shown in FIG. 8(B), the smartphone version of the talk screen 112 also serves as the voice input screen 117 of the PC version described above, and as an example, a microphone icon is displayed along with the message "Please speak." Here, as shown in FIG. 9(A), it is assumed that the user inputs, by way of example, "What is the mileage?" by voice.

[0063] Then, as one example, the display of the microphone icon changes to indicate that the browser 111 is detecting voice input through the microphone. The browser 111 converts the input voice into text, inputs the text data TX into the input field 114a on the talk screen 112, and transmits it to the bot 121 of the first server 120 as instruction command information PT.

[0064] More specifically, as an example, the input voice is converted into text and the text data TX, ie, "What is the mileage?" is displayed on the talk screen 112. Then, the browser 111 inputs the text data TX, “What is the mileage?” into the input field 114 a of the talk screen 112 , and transmits it to the bot 121 of the first server 120 .

[0065] Then, the bot 121 of the first server 120 uses the chat GPT, which is an example of the pre-trained artificial intelligence means 131 of the second server 130, to generate text data based on the instruction command information PT. Then, the bot 121 of the first server 120 transmits the generated text data GT obtained through the execution to the user terminal 110. Then, the user terminal 110 displays the received generated text data GT on the talk screen 112 of the browser 111. At the same time, the browser 111 outputs the contents of the generated text data GT as voice using the speaker of the user terminal 110.

[0066] The artificial intelligence system 100 thus obtained, which is an embodiment of the present invention, comprises a user terminal 110, a first server 120 having a bot 121, and a second server 130 having a chat GPT which is an example of a pre-trained artificial intelligence means 131, and the browser 111 of the user terminal 110 communicates with the first server 120 to display a talk screen 112, and when a predetermined operation is performed on the talk screen 112, the microphone of the user terminal 110 becomes active, and when voice input is performed, the browser 111 converts the input voice into text, inputs the text data TX into an input field 114a on the talk screen 112, and transmits it to the bot 121 of the first server 120 as instruction command information PT, and The client 121 generates text data based on the instruction command information PT using chat GPT, an example of pre-trained artificial intelligence means 131 of the second server 130, and transmits the generated text data GT obtained by the execution to the user terminal 110. The user terminal 110 displays the received generated text data GT on the talk screen 112 of the browser 111, and the browser 111 outputs the content of the generated text data GT as voice using a speaker. This configuration eliminates the need for the user to input text using a keyboard, and also eliminates the need for the user to read with their eyes, avoiding the inconvenience, and allows the user to smoothly communicate with the chat GPT, an example of pre-trained artificial intelligence means 131, using their mouth and ears.

[0067] In addition, the talk screen 112 is provided with a language selection for input and output, and when voice input is made, the browser 111 recognizes the input voice as the selected language, converts it to text, and inputs the text data TX into the input field 114a on the talk screen 112, and transmits it to the bot 121 of the first server 120 as instruction command information PT. When the user terminal 110 receives the generated text data GT, the received generated text data GT is displayed on the talk screen 112 of the browser 111, and the browser 111 recognizes the contents of the generated text data GT as the selected language and outputs the content as voice using the speaker, thereby enabling the user to speak multiple languages.

[0068] Furthermore, the talk screen 112 is provided with a voice output selection option, and the browser 111 is configured to output the contents of the generated text data GT using a speaker in a voice selected by the user, so that the user can hear the contents of the generated text data GT as a response from the chat GPT, which is an example of the pre-trained artificial intelligence means 131, in a voice output in the user's preferred voice or speaking style.

[0069] In addition, the talk screen 112 has a talk display area 114 that displays instruction and command information PT and generated text data GT, which are the exchange between the user and the chat GPT, which is an example of the pre-trained artificial intelligence means 131, and a prompt list display area 115 that displays a prompt item 115b, which is a standard instruction and command information item that has been set up and registered in advance.When a selection operation is performed on the prompt item 115b, which is an instruction and command information item, the contents of the instruction and command information PT of the prompt item 115b, which is the selected instruction and command information item, are entered into the input field 114a of the talk screen 112 and transmitted to the chat GPT, which is an example of the pre-trained artificial intelligence means 131 of the second server 130, via the bot 121 of the first server 120.This means that the user does not need to input text from a keyboard, and can give, for example, detailed instructions and commands that have been set up and registered in advance, and obtain responses to those detailed instructions and commands.

[0070] Furthermore, in the prompt list display area 115, prompt items 115b, which are instruction command information items, can be freely added and set by user operation and administrator operation, and prompt items 115b, which are instruction command information items set by user operation, are displayed in a user-configured instruction command information item group 115d arranged above the common instruction command information item group 115c set by administrator operation, and in the user-configured instruction command information item group 115d, the prompt item 115b, which is the instruction command information item that was most recently selected, is rearranged and displayed at the top, thereby making it easy for each user to use.

[0071] Furthermore, the program of the artificial intelligence system 100 which is an embodiment of the present invention includes steps S1 and S2 in which the browser 111 of the user terminal 110 communicates with the first server 120 to display the talk screen 112 and determines whether or not a predetermined operation has been performed on the talk screen 112, a microphone activation step S3 in which the microphone of the user terminal 110 is activated when the predetermined operation has been performed, a voice input presence / absence determination step S4 in which it is determined whether or not voice input has been performed, an instruction command transmission step S5 in which, when voice input has been performed, the browser 111 converts the input voice into text, inputs the text data TX into the input field 114a on the talk screen 112, and transmits it as instruction command information PT to the bot 121 of the first server 120, and The system includes a text data generation step S6 in which text data is generated based on the instruction command information PT using a chat GPT as an example, a text data transmission step S7 in which the bot 121 of the first server 120 transmits the generated text data GT obtained by the execution to the user terminal 110, and a display / audio output step S8 in which the user terminal 110 displays the received generated text data GT on the talk screen 112 of the browser 111 and the browser 111 outputs the contents of the generated text data GT as audio using a speaker. This eliminates the need for the user to input text using a keyboard, and eliminates the need for the user to read with their eyes, which is an annoyance, and has a significant effect in that the user can smoothly communicate with the chat GPT, an example of the pre-trained artificial intelligence means 131, using their mouths and ears. [Explanation of symbols]

[0072] 100 ··· Artificial Intelligence Systems 110 User terminal 111 ··· Browser 112 ··· Chat screen 113 Document item display area 113a··· Document item 114 ··· Talk display area 114a Input field 114b Microphone icon button 114c··· Clipboard icon 114d··· Speaker icon 115 Prompt list display area 115a "New Prompt" button 115b Prompt item (standard instruction information item) 115c... Common instruction command information item group 115d User-defined instruction and command information items 116 Language selection button 117 ··· Voice input screen 118 ··· Prompt registration window 118a··· Title input field 118b Prompt setting input field 120 1st Server 121 ··· Bot 130 Second Server 131 ··· Pre-trained artificial intelligence means TX: Text data PT... Directive command information GT: Generated text data

Claims

1. An artificial intelligence system comprising a user terminal, a first server having a bot, and a second server having a pre-trained artificial intelligence means, the user terminal being capable of freely communicating with the pre-trained artificial intelligence means via the bot, a browser of the user terminal communicates with the first server to display a talk screen, and when a predetermined operation is performed on the talk screen, a microphone of the user terminal becomes active and voice input is performed, the browser converts the input voice into text, inputs the text data into an input field on the talk screen, and transmits it to a bot of the first server as instruction command information; The bot of the first server uses the pre-trained artificial intelligence means of the second server to generate text data based on the instruction command information, and transmits the generated text data to the user terminal; An artificial intelligence system characterized in that the user terminal displays the received generated text data on a browser's talk screen, and the browser outputs the contents of the generated text data aloud using a speaker.

2. The talk screen is provided with a language selection for input and output, and when voice input is received, the browser recognizes the input voice as the selected language, converts it into text, inputs the text data into an input field on the talk screen, and transmits it to the bot of the first server as instruction command information; The artificial intelligence system of claim 1, characterized in that, when the user terminal receives the generated text data, it displays the received generated text data on the talk screen of the browser, and the browser recognizes the content of the generated text data as the selected language and outputs it aloud using a speaker.

3. The artificial intelligence system described in claim 2, characterized in that the talk screen is provided with a voice selection option for audio output, and the browser is configured to output the contents of the generated text data using a speaker in the voice selected by the user.

4. The talk screen has a talk display area for displaying instruction command information and generated text data which are an exchange between a user and a pre-trained artificial intelligence means, and a prompt list display area for displaying a preset and registered standard instruction command information item, The artificial intelligence system described in any one of claims 1 to 3, characterized in that when a selection operation of the instruction command information item is performed, the contents of the instruction command information of the selected instruction command information item are entered into an input field on the talk screen and transmitted to the pre-trained artificial intelligence means of the second server via a bot of the first server.

5. In the prompt list display area, command information items can be added and set by a user operation and an administrator operation, and the command information items set by the user operation are displayed in a user-defined command information item group arranged above a common command information item group set by the administrator operation, 5. The artificial intelligence system according to claim 4, characterized in that, in the group of user-defined instruction command information items, the instruction command information item that has been most recently selected is rearranged and displayed at the top.

6. A program for an artificial intelligence system in which a user terminal can communicate with a pre-trained artificial intelligence means of a second server via a bot of a first server, a predetermined operation presence / absence determination step in which a browser of the user terminal communicates with a first server to display a talk screen and determines whether or not a predetermined operation has been performed on the talk screen; a microphone activation step of activating a microphone of the user terminal when the predetermined operation is performed; a voice input presence / absence determination step for determining whether or not a voice input has been made; When the voice input is received, the browser converts the input voice into text, inputs the text data into an input field on the talk screen, and transmits the text data as instruction command information to a bot of a first server; A text data generating step in which the bot of the first server generates text data based on instruction command information using the pre-trained artificial intelligence means of the second server; A text data transmission step in which the bot of the first server transmits generated text data obtained by execution to a user terminal; A program for an artificial intelligence system, characterized in that the user terminal displays the received generated text data on a talk screen of a browser, and the browser outputs the contents of the generated text data as voice using a speaker.

Images