JP2002259376A - Natural language generating system for electronic pet - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a natural language generating system for electronic pet capable of performing natural language generation and automatically performing word registration as well by especially setting a parameter to a word. SOLUTION: In this natural language generating system for electronic pet using an electronic pet such as 'Seaman', in terminal equipment for storing the program of the electronic pet, a conversation between a human-face fish 'Seaman' and an operator is performed by generating a natural language. Besides, a word is registered corresponding to a category, for example, and semantic information corresponding to the relevant category is registered for each category. When using such a word, a conversation sentence is prepared by referring to the registered semantic information, the conversation is performed in the natural language and the conversation is automatically branched or continued.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a natural language generation system for an electronic pet.

[0002]

2. Description of the Related Art With the recent advancement in electronics technology, there is an electronic pet program that accumulates information regarding the personality of an operator as knowledge and realizes a response suitable for the operator based on the accumulated various data. For example, there is a system that talks with an electronic pet swimming in a water tank and communicates with an operator.

[0003]

In such a communication system with an electronic pet, communication proceeds by conversation between the electronic pet and a user. At this time, voice recognition and voice synthesis of the electronic pet and the user are performed using a voice recognition engine and a voice synthesis engine.

However, in conventional speech synthesis, for example, a preset conversation is sequentially generated, and the electronic pet asks a question according to the preset text data, and the personality information database is created based on the answer from the user. Generate. Therefore, the conversation cannot be changed according to the content of the conversation, and the conversation cannot be progressed by natural language generation. Also, words that can be recognized by the electronic pet are set in advance, and it is difficult to automatically register the words as the conversation progresses.

The present invention has been made in view of the above problems, and provides a natural language generation system for an electronic pet that can generate a natural language by setting parameters for words and can also automatically register new words. Aim.

[0006]

According to the first aspect of the present invention, there is provided an electronic apparatus comprising: a plurality of terminal devices; and a server computer connected via a network and connected to enable data communication. A natural language generation system for a pet, the storage unit storing words corresponding to each category and semantic information corresponding to the category, and a storage unit configured to generate a language of the electronic pet when storing the language of the electronic pet. This can be achieved by providing a natural language generation system for an electronic pet having a language generation unit that searches for a category corresponding to the word and generates a natural language based on semantic information corresponding to the category.

Here, the category is, for example, the action described in claim 4, the place described in claim 5, or the occupation described in claim 6. Then, for example, a corresponding category is searched according to words used in generating a language of the electronic pet, a related language is generated from semantic information on the category, and natural language generation and natural conversation are performed.

According to a second aspect of the present invention, in the first aspect, the language generating means selects a language following the word based on the semantic information stored in the storage means.

With this configuration, the language spoken by the electronic pet can be converted into a natural conversation, and the conversation can be branched and advanced. According to a third aspect of the present invention, there is provided an electronic pet natural language generation system including a plurality of terminal devices and a server computer connected to the terminal devices via a network and capable of data communication. Storage means for storing words corresponding to each category, and semantic information corresponding to the categories, and confirmation means for confirming whether a new word is included in the response of the operator to the question from the electronic pet. When,
When a new word is included by the confirmation means, a natural language generation system for an electronic pet is provided which has a registration means for asking a question based on semantic information from the storage means and registering the new word in a corresponding category. Can be achieved by:

Here, the category is, for example, the action described in claim 4, the place described in claim 5, or the occupation described in claim 6, similarly to the above. Then, for example, a corresponding category is searched according to the word used when generating the language of the electronic pet, and a related language is generated from the semantic information regarding the category.

With this configuration, it is possible to ask a question based on the semantic information included in the category, know the category of the new word, and automatically register the new word.

[0012] The above object is achieved according to the seventh aspect of the present invention.
A method for generating a natural language of an electronic pet, comprising: a plurality of terminal devices; and a server computer connected via a network and connected to enable data communication. And a category corresponding to the word stored in the storage means storing semantic information corresponding to the category,
This can be achieved by providing a natural language generation method for an electronic pet that performs a language generation process for generating the natural language based on the semantic information corresponding to the category.

This embodiment is an invention of a method for generating a natural language of an electronic pet. With such a configuration, a related language is generated from the semantic information on the category, and a natural language is generated. Have a conversation.

[0014] The above object is attained by the invention according to claim 8.
A method for generating a natural language of an electronic pet, comprising: a plurality of terminal devices; and a server computer connected to the terminal devices via a network and capable of performing data communication, comprising: a word corresponding to each category; A process of storing corresponding semantic information; a process of checking whether a new word is included in an operator's response to the question from the electronic pet; and a process in which a new word is included by the process. This can be achieved by providing a method for generating a natural language of an electronic pet that performs a process of asking a question based on semantic information and registering the new word in a corresponding category.

This embodiment is also an invention of a method for generating a natural language of an electronic pet. With this configuration, a new word can be registered in an appropriate category from the semantic information included in the category.

The above object is attained according to the ninth aspect of the present invention.
A program used for an electronic pet natural language generation system including a plurality of terminal devices and a server computer connected via a network circuit and connected for data communication, wherein the program generates the language of the electronic pet. At this time, a language corresponding to each category and a category corresponding to the language stored in the storage unit storing the semantic information corresponding to the category are searched, and the natural language is determined based on the semantic information corresponding to the category. This can be achieved by providing a natural language generation program for an electronic pet that performs language generation processing for generating a natural language.

In this embodiment, the natural language generation of the electronic pet is realized by a program recorded on, for example, a floppy (registered trademark) disk or a CD-ROM. Generate relevant language from information, natural language generation,
You can have a natural conversation that leads first.

According to the tenth aspect of the present invention, a natural language generation of an electronic pet including a plurality of terminal devices and a server computer connected to the terminal devices via a network and capable of data communication. A program for storing a word corresponding to each category and semantic information corresponding to the category, and confirming whether a new word is included in a response of the operator to a question from the electronic pet. By providing a natural language generation program for an electronic pet that performs a process and a process of asking a question based on the semantic information when a new word is included in the process and registering the new word in a corresponding category. Can be achieved.

This embodiment also has a configuration in which the natural language generation of the electronic pet is realized by, for example, a program recorded on a floppy disk or a CD-ROM.
For example, it is possible to sequentially retrieve personal information and the like of an operator, write personal information in a personality information database, and create a database.

[0020]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a system configuration diagram of a natural language generation system using an electronic pet according to the present invention. As shown in FIG. 1, the natural language generation system for an electronic pet according to the present embodiment includes personal computers 1a and 1b, which are a plurality of platforms owned by a user, a computer game machine 6, a mobile phone 7, an electronic pet, and the like. And a server computer 3 owned by a service provider that provides a service for providing a mermaid "Seaman".

The personal computers 1a, 1b,
The game machine 6 is connected to the Internet 21 via each of the communication devices 2a, 2b, and 5, and the mobile phone 7 is connected to the Internet 21 via a not-shown relay station or central control station.

The personal computers connected to the Internet 21 are indicated by 1a and 1b. Many computers other than the above two computers are connected to the Internet 21. The personal computer 1a uses, for example, Windows as the OS,
The personal computer 1b uses, for example, a Macintosh as the OS.

First, the configuration of the personal computer 1 used by the receiver and the sender in this embodiment is shown in FIG. In the description of the configuration of the personal computer 1, the personal computer 1a will be representatively described.

As shown in FIG. 2, a data bus 10 for transmitting and receiving data is provided inside the personal computer 1a.
The data bus 10 is connected to a central processing unit (hereinafter, referred to as a CPU) 11, a RAM 12, a real-time clock (RTC) 17, and the like.

The CPU 11 outputs a question to a user and accepts an answer, which will be described later, as speech of an electronic pet displayed on a display device by speech synthesis. A response receiving process for selecting and registering a corresponding item from a plurality of answer candidates set as above and storing the selected item as personality information of the user is performed. During this process,
The RAM 12 is used as a work area.

A real-time clock (RTC) 1
Reference numeral 7 outputs, for example, current time information used for date and time information of data update, and calendar information such as an arbitrary day of the week. The input device 16 is a keyboard, a mouse, and the like.
Notify.

The display device 14 is a display such as a CRT or an LCD, and displays a Seaman described later. The communication interface 13 is connected to the communication device 2 (2
a) and transmits / receives data to / from the server computer 3 via the communication device 2 (2a) and the Internet 21.

The audio input / output device 18 is connected to a speaker 19 and a microphone 20 as external devices, and A / D / D
/ A converter. The voice input / output device 18 converts (D / A converts) voice data generated by a voice synthesis program to be described later into analog voice, outputs the voice data to the speaker 19, and converts the voice input from the microphone 20 into digital voice. The data is converted (A / D converted) and output.

The storage device 15 is made up of a magnetic disk or a magneto-optical disk, and writes and reads data and programs under the control of the CPU 11. The storage device 15 stores an electronic pet program in which an image and an operation process of a mermaid fish "Seaman", which is an electronic pet displayed on the display device 14, and a voice of the mermaid fish "Seaman" are stored. A speech synthesis program for outputting speech based on text data, a speech recognition program for converting input speech of a sender or a recipient into text data, and the like are stored.

In the storage device 15, a group of words used and understood by a mermaid "Seaman" in an application is registered as a "dictionary". FIG. 3 is a diagram showing the structure of this dictionary, which is registered for each category.

For example, "action", "place", "occupation" and the like are registered as categories, and corresponding words are registered in each category. For example, words such as "date", "movie", "shopping", and "sports" are registered in the "action" category. In the category of "place", words such as "Tokyo", "sea", "theme park", and "theater" are registered. further,
In the category of "occupation", words such as "salaryman", "programmer", "doctor", and "public servant" are registered.

On the other hand, in the information area indicating the semantic information, information indicating the semantic content is stored for each category. For example, semantic information such as-is registered for the category of "action". For example, semantic information is registered for "doing one person, or two or more people, or one or more people?"
In addition, the verb form is registered, for example, semantic information such as "to do or to go" is registered. Furthermore, the semantic information of “whether the relationship is more than a lover or less” is registered. These pieces of semantic information are related to the above-mentioned “action” category, and are, for example, naturally conceived languages and ideas.

Similarly, regarding the category of “place”, semantic information such as “〜” is registered. For example, the semantic information includes “whether the“ place ”is a place name or not a place name” In the information, the "place" is registered as "famous place or not a famous place", and as the semantic information, the "place" is registered as semantic information "whether far or near".

The same applies to “occupation”.
For example, the semantic information shown in FIG. In addition to the above categories, other categories are registered, and words corresponding to the categories and many pieces of semantic information are registered.

FIG. 4 is a diagram for explaining the system configuration of the server computer 3 of the service provider. A data bus 30 is also provided in the server computer 3, and a CPU 31, a RAM 32, a communication interface 33 and the like are connected to the data bus 30. The CPU 31 transmits the shared personality information D associated with the reception of the personality information DB data from the user.
B performs data update processing and the like.
It functions as a work area used for the processing of 31. Further, the real-time clock (RTC) 37 outputs current time information used for data update information and calendar information such as a day of the week of an arbitrary date. Further, the communication interface 33 is connected to the above-described communication device 4 and transmits and receives data via the Internet 21.

The processing operation of the system having the above configuration will be described below. <First Embodiment> This embodiment is an example in which a natural conversation is performed using the database shown in FIG. Hereinafter, the processing will be described with reference to the flowchart shown in FIG. In addition, the flowchart on the left side shown in the figure shows processing on the mermaid "Seaman" side, and the flowchart on the right side shows processing on the operator side.

First, when a mermaid "Seaman" generates a language, words to be used are read out (step (hereinafter referred to as ST).
1). Next, a category related to the read word is determined (ST2). In this determination, for example, if the word is "movie", it is understood that the word is a category of "action" (ST3). Also, if the word is "Tokyo", it can be understood that the category is "Place".

Next, a question sentence for the mermaid "Seaman" is created from the above category. This question is created based on the semantic information corresponding to the category. For example, if the word is "movie", the category is "action", and for example, the conversation information is created by selecting the semantic information of the category. For example, "You went to the movie yesterday,"
Create a conversation saying "Today go to the movie."

Next, the speech of the above-mentioned conversation is synthesized and output from the speaker 19 as the voice of the mermaid "Seaman" (ST5). At this time, the text data of the conversation is output to the display device (display) 14, and the conversation is also displayed. FIG. 6 shows a display state of the mermaid "Seaman" at this time.

The operator knows the question by displaying the mermaid "Seaman" on the display and listening to the sound from the speaker 19 (ST6). Then, consider the response to the question from the mermaid "Seaman" (ST
7). For example, the question might be "Yes, yesterday, the movie went,"
Or, if it is "Today, go to a movie," consider a response such as "Yeah, go" or "Yeah, go" and speak to the microphone 20 (ST8).

Next, when there is the above response, it is determined whether there is any more question (YES in ST9, ST10). Here, if there is no question, the process is terminated, but if there is a further question, a question sentence is created (NO in ST10, ST4). In this case, the next question is also created with reference to the semantic information of the category “action”.

For example, using the semantic information of the category of “action”, a conversation sentence “movie, go alone, go” or “movie, go alone, go” is created, and the speaker 1
9 and output it as the voice of the mermaid "Seaman" (ST
5). Thereafter, as described above, the operator considers the response and responds to the mermaid "Seaman" (ST6 to ST9).

Thereafter, the above processing is repeated, and when all the questions are completed (ST10: YES), the processing is completed. As described above, according to the present embodiment, a conversation sentence is created by referring to the semantic information set in advance depending on the word to be used, and the conversation is performed in a natural language.

In the above example, the example of the word "movie" has been described. However, when words such as "date" and "shopping" are used, for example, "yesterday, date, did", "yesterday, Conversation sentence such as "I did shopping." Also, "Tokyo",
If it is a category related to "place" such as "sea", it will be a conversation sentence such as "there, place name", "there is a famous place" or the like.

Similarly, conversation can be performed in a natural language corresponding to the word to be used for other categories such as "occupation". <Second Embodiment> Next, a second embodiment will be described.

In this example, the conversation is branched and the conversation is advanced. Also in this example, the processing will be described using the database shown in FIG. 3 and according to the flowchart shown in FIG.

Also in this example, first, a mermaid "Seaman"
Reads a word to be used when generating a language (step (hereinafter, referred to as STP) 1), and determines a category to which the word is related (STP2). In this case, for example, the word "programmer" is a category of "occupation" (STP3). From the above category, for example, semantic information is selected, and a question saying “It seems to be a programmer or a very busy job” is sent to the operator (STP
4, STP5).

The operator listens to the display of the mermaid "Seaman" and the question from the speaker 19, considers the answer, and gives a response such as "Yeah, it's quite busy" (STP6 to STP).
TP8). If there is a further question to the operator, a new question is asked in the same manner as in the above-described example.

However, in this example, after receiving the response,
The word included in the response is determined (STP10). If a new word is included, the process proceeds to the above-described category determination (YES in ST10, STP2). For example, in the category of "occupation", semantic information of "workplace, where" is selected (see FIG. 8).
When the operator answers "Tokyo" or "Yokohama", a new word is included, and the category shifts to the "place" category in the category judgment (STP2).

Therefore, the conversation branches and proceeds thereafter. For example, in the above case, a new category related to "places",
It develops into conversations such as "There is near."

That is, by processing as described above, it is possible to branch to a new conversation and proceed with the conversation. Therefore, when a new word, for example, a word such as “shopping” or “sports” is used during a conversation related to the category, the category is shifted to another category of “action”.

Thereafter, the above processing is repeated, and when all the questions are completed (YES in ST11), the processing is completed in the same manner as described above. As described above, according to the present example, during the conversation between the mermaid "Seaman" and the operator, if a new word is included, the category shifts to the category corresponding to the new word, and the conversation branches and proceeds. And the conversation can be progressed more naturally. <Third Embodiment> Next, a third embodiment will be described.

This example describes a process for registering a new word. If a new word is included, the database is checked. If the new word is not registered in the dictionary,
This is a configuration for performing registration processing of the word. Hereinafter, description will be given according to the flowchart shown in FIG.

First, during the conversation between the mermaid "Seaman" and the operator, it is determined whether a new word is included in the response of the operator (step (hereinafter, referred to as S) 1). If a new word is included, it is determined whether the word is registered in the database (S2). Here, if it is registered in the database (YES in S2), a question is created from the semantic information of the category related to the word as described in the second embodiment.

On the other hand, if the word is not registered in the database (S2: NO), a question is created to register the word in the database, and the operator is asked a question (S3). Then, when the operator responds to the question, the category of the word is determined from the response content (S4 to S6, S7,
S8).

For example, in the case of the new word "Hokkaido", when the operator answers "Yes" to the question "Is there a place name?", The mermaid "Seaman" is replaced by the word "Hokkaido". Location category.
Register the word "Hokkaido" in the "Place" category. On the other hand, when the operator asks the word "public service" for "busy job" and the operator answers "yes", the mermaid "Seaman" has the word "public employee" in the category related to "occupation". Judge that there is, and register the word "public servant" in the "occupation" category.

If the operator gives a negative response such as "No" or "No" to the question of the mermaid "Seaman", it is determined that the word is not included in the category including the meaning information. And ask the following questions:

As described above, according to the present embodiment, when a new word is detected, the new word can be automatically registered in the database. It can be a production system.

In the above description of the first to third embodiments, the example in which the mermaid fish "Seaman" is used as the electronic pet has been described. However, the present invention is not limited to the mermaid fish "Seaman". The present invention is not limited to the electronic pet displayed on the display, but may be applied to a robot-type electronic pet such as a dog or a cat.

The semantic information shown in FIG. 3 may be configured to select a numerical value and register the degree of busyness at a level. Further, in the description of the above embodiment, the example using the personal computer 1 (1a, 1b) has been described.
Needless to say, the above processing can be performed by the game machine 6 or the mobile phone 7 shown in FIG.

[0061]

As described above, according to the electronic pet natural language generation system of the present embodiment, the conversation between the electronic pet and the operator can be performed in a more natural language. You can also proceed.

In addition, a word can be automatically registered, and a natural language generation system for an electronic pet that is easier to use can be provided.

[Brief description of the drawings]

FIG. 1 is a system diagram of a natural language generation system for an electronic pet according to the present invention.

FIG. 2 is a block diagram showing a computer owned by a user used in the embodiment of the present invention.

FIG. 3 is a diagram showing a configuration of a database registered in a storage device.

FIG. 4 is a block diagram illustrating a server computer used in the present embodiment.

FIG. 5 is a flowchart illustrating a first embodiment.

FIG. 6 is a diagram showing a sounding state of a mermaid “Seaman”.

FIG. 7 is a flowchart illustrating a second embodiment.

FIG. 8 is a diagram showing a sounding state of a mermaid “Seaman”;

FIG. 9 is a flowchart illustrating a third embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Personal computer (user) 2 Communication device 3 Server computer 4 Communication device 5 Communication device 6 Computer game machine 7 Cellular phone 8 Television 9 Microphone 10 Data bus 11 Central payment processing device (CPU) 12 RAM 13 Communication interface 14 Display device 15 Storage device 19 Speaker 20 Microphone 21 Internet network 30 Data bus 31 Central payment processing device (CPU) 32 RAM 33 Communication interface 35 Storage device

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (Reference) G10L 21/06 F term (Reference) 2C150 AA14 CA01 CA02 CA04 DA04 DA05 DA24 DA25 DA26 DA27 DA28 DF03 DF04 DF06 DF33 ED42 ED47 ED52 EF07 EF16 EF17 EF23 EF29 EF33 EF36 5B091 AA11 CA14 CA24 CB32 5D045 AB11

Claims (10)

[Claims] 1. A natural language generation system for an electronic pet, comprising: a plurality of terminal devices; and a server computer connected via a network and connected to enable data communication, wherein a word corresponding to each category is generated. A storage unit that stores semantic information corresponding to the category; and, when generating a language of the electronic pet, searching for a category corresponding to the word stored in the storage unit. A language generating means for generating a natural language on the basis of: a natural language generating system for an electronic pet. 2. The electronic pet natural language generation system according to claim 1, wherein the language generation unit selects a language following the word based on semantic information stored in the storage unit. . 3. A natural language generation system for an electronic pet, comprising: a plurality of terminal devices; and a server computer connected to the terminal devices via a network and capable of performing data communication, wherein a word corresponding to each category is provided. Storage means for storing semantic information corresponding to the category; confirmation means for confirming whether a new word is included in the response of the operator to the question from the electronic pet; And a registration unit for asking a question based on the semantic information from the storage unit, and registering the new word in a corresponding category. 4. The electronic pet natural language generation system according to claim 1, wherein the category is an action or the like. 5. The electronic pet natural language generation system according to claim 1, wherein the category is a place or a similar place. 6. The electronic pet natural language generation system according to claim 1, wherein the category is occupation or the like. 7. A method for generating a natural language of an electronic pet used in a system including a plurality of terminal devices and a server computer connected via a network and connected to enable data communication, the electronic pet comprising: When the language of the pet is generated, the language corresponding to each category and the category corresponding to the language stored in the storage unit storing the semantic information corresponding to the category are searched, and the semantic information corresponding to the category is searched. A natural language generation method for an electronic pet, comprising: performing a language generation process for generating a natural language based on the natural language. 8. A natural language generation method for an electronic pet used in a system including a plurality of terminal devices and a server computer connected via a network and connected to enable data communication, the method comprising: And a process of storing semantic information corresponding to the category. If a new word is included in the response of the operator to the question from the electronic pet, a question based on the semantic information is generated. And registering the new word in a corresponding category. 9. A program used in a natural language generation system for an electronic pet, comprising: a plurality of terminal devices; and a server computer connected via a network and connected to enable data communication. When the language of the pet is generated, a word corresponding to each category and a category corresponding to the language stored in a storage unit in which the meaning information corresponding to the category is stored are searched, and the meaning information corresponding to the category is searched. A natural language generation program for an electronic pet, which performs language generation processing for generating the natural language based on the natural language. 10. A program used in an electronic pet natural language generation system including a plurality of terminal devices and a server computer connected to the terminal devices via a network and capable of data communication, the program comprising: A process for storing a word corresponding to each of the categories and semantic information corresponding to the category; and a question based on the semantic information when a new word is included in a response of the operator to the question from the electronic pet. And a process of registering the new word in a corresponding category.