JP2000339315A - Character dialogue support method and apparatus, recording medium recording character dialogue support program, and recording medium recording character dialogue support data - Google Patents

Abstract

(57) [Summary] A character dialogue for smooth communication between a user of a first language and a user of a second language connected by a communication line by using characters of their own language with each other. Provided are a support method and apparatus, a recording medium on which a character dialogue support program is recorded, and a recording medium on which character dialogue support data is recorded. Kind Code: A1 A word dictionary that records words that can be mechanically translated into English and word codes unique to each word as terms related to Japanese, a sentence pattern of a first language and a sentence pattern code unique to each sentence. And a means for selecting Japanese sentence pattern candidates from the predicate sentence pattern recording unit 11,
Means for selecting a word candidate necessary for each case constituting the designated sentence pattern from the word dictionary 14, and, when all the words constituting the sentence pattern are determined, the corresponding sentence pattern from the predicate sentence pattern storage unit 11 and the word dictionary 14; A feature of having means for reading out a code and a word code, and a code transmitting section 9a for transmitting the sentence pattern code and the word code to the communication line 8.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for supporting character dialogue, a recording medium on which a character dialogue support program is recorded, and a recording medium on which character dialogue support data is recorded. Dialogue support method for supporting a textual dialogue performed between a user of the second language and a user of the second language via an arbitrary communication line for communication between them, and directly used in the implementation thereof A character dialogue support apparatus, a recording medium storing a character dialogue support program for supporting the character dialogue using a computer, and character support data used for supporting the character dialogue using a computer It relates to a recording medium.

[0002]

2. Description of the Related Art As a means for realizing communication between remote points, there is a so-called chat system.

[0003] A chat system is an operation in which characters input by one user from his / her own terminal are sequentially displayed on the screen of the other user's terminal via a communication line. This system enables mutual communication between both users in real time.

[0004] When trying to communicate with such a system, it is a major premise that the language used by both users is a common language (for example, Japanese).
If the language understandable by one user is different from the language understandable by the other user, communication between the two naturally becomes inconvenient.

[0005] For this reason, in order for the conventional chat system to communicate with users of different languages that can understand each other through character dialogue, a sentence of a specific language used by one user is translated into another language. A means for translating into a sentence in another language used by a user is required.

FIG. 9 is a conceptual block diagram of a conventional chat system showing an example in which a translation means for communicating with each other through character dialogue is applied to the conventional chat system.

As shown in FIG. 1, in this case, in addition to the input devices (keyboards) 1a and 1b and the display devices 2a and 2b in the constituent terminals of the chat system α, both input devices 1a and 1b are provided by both users. Before transmitting the actual sentence input through the communication line 3 to the communication line 3,
a, 4b, and after mutually translating into a sentence in another language different from the language of the sentence (for example, if one translator 4a performs Japanese-English translation, the other translator 4b Japanese translation is performed), and the corresponding translation is displayed on each of the display devices 2a and 2b of the partner terminal.

Here, typical means for actual translation in each of the translation devices 4a and 4b include morphological analysis processing, syntax analysis processing, semantic analysis processing, and context analysis processing.
In addition, by creating a translation using a translation template, or by presenting a pre-registered example,
Techniques have been devised to support the required translation.

[0009]

However, in the case of translating an arbitrary input sentence in the above-mentioned case, no perfect means exists at present even if any translation means is employed in each of the translation devices 4a and 4b. Therefore, erroneous translation frequently occurs, and it is not possible to obtain sufficient translation accuracy for practical use.

To cope with this problem, it is conceivable to take measures such as automatically restricting the transmission of a sentence determined to be a mistranslation and optimizing the communication of intentions. Doing so results in a limited transmission of a significant percentage of sentences. As described above, unless many translations are transmitted, the intention of the sender is naturally not transmitted to the other party, and communication between the two parties is hardly established.

For this reason, at present, methods and apparatuses for character dialogue having an automatic translation function are not widely used, and there is a demand for a method and apparatus for character dialogue support having high translation accuracy.

Here, a first object of the present invention is to provide a first language user and a second language user between remote points connected by a communication line with each other in their own language. The present invention provides a novel character dialogue support method and apparatus capable of smoothly communicating with each other, a recording medium recording a character dialogue support program, and a recording medium recording character dialogue support data.

A second object of the present invention is to provide a character dialogue support method and apparatus capable of flexibly coping with character dialogues concerning arbitrary two types of languages, a recording medium storing a character dialogue support program, and a character dialogue support. It is intended to provide a recording medium on which data is recorded.

[0014] Other objects of the present invention will become apparent from the description of the specification, drawings, and particularly from the claims.

[0015]

According to the method of the present invention,
As terms related to the first language, words that can be mechanically translated into the second language are adopted in a limited manner, and words in the first language are
When a word constituting a predicate is specified in an input sentence of the first language that a user of the first language intends to create, a sentence pattern candidate of the first language in which the specified predicate can be used is presented;
A desired sentence pattern is designated from the presented sentence pattern candidates of the first language, and further, when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined, the fixed sentence pattern of the first language and A sentence pattern code and a plurality of word codes each representing a type of a plurality of fixed words are generated, and each of these codes is transmitted to a communication line as an intermediate expression in a first language and a second language. .

On the other hand, in the device of the present invention, as a term relating to the first language, a word code which is intended to be related to a corresponding word in the second language is recorded together with a word which can be mechanically translated into a second language. Word recording means, predicate sentence pattern recording means for recording a sentence pattern code for associating with a corresponding second language sentence pattern together with a first language sentence pattern, and a predicate designated from words in the first language may be used. Sentence pattern candidates in the first language
A sentence pattern candidate selection and display means for selecting and displaying from the predicate sentence pattern recording means, and selecting a necessary word candidate from the word recording means for each case constituting the designated first language sentence pattern from the first language sentence pattern candidates Word candidate selection display means for displaying
When all the specified words from the word candidates are confirmed,
It is characterized by having code reading means for reading a sentence pattern code and a plurality of word codes from a predicate sentence pattern recording means and a word recording means, and code transmission means for sending a sentence pattern code and a plurality of word codes to a communication line.

Also, in the program recording medium of the present invention, the user of the first language intends to create the character dialogue support program recorded on the computer from the words of the first language prepared in advance in the computer. When a word constituting a predicate is specified in the input sentence in the first language, a prepared sentence pattern candidate in the first language in which the specified predicate can be used is presented, and the presented sentence pattern candidate in the first language is presented. When a desired sentence pattern is designated from among them, and when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined,
A pre-prepared sentence pattern code and a plurality of word codes respectively representing the fixed sentence pattern of the first language and the plurality of fixed words are generated, and these codes are converted into the first language and the second word.
It has the characteristic that it is transmitted to a communication line as an intermediate expression over a language.

Still further, in the data recording medium of the present invention, the character dialogue support data recorded on the data recording medium is an area in which words that can be mechanically translated into a second language are recorded as terms relating to the first language. A word file including an area in which a word code assigned to each word is recorded;
A predicate sentence pattern file including an area in which a sentence pattern of a language is recorded together with types of predicates usable in the sentence pattern, and an area in which a sentence pattern code assigned to each sentence pattern is recorded;
And a predicate inflection type file having an area in which inflectional forms relating to words forming a predicate among words in the first language are recorded.

More specifically, in solving the above-mentioned problems, the present invention achieves the above object by adopting a novel characteristic configuration method and means ranging from a superordinate concept to a subordinate concept listed below. Done.

That is, a first feature of the method of the present invention is that a character for supporting a character dialogue between a user of a first language and a user of a second language via an arbitrary communication line is provided. In a dialogue support method, words that can be mechanically translated into a second language are limitedly employed as terms related to the first language when performing a character dialogue, and words in the first language are selected from among words in the first language. When a word constituting a predicate is specified in an input sentence in a first language that a user of one language intends to create, a sentence pattern candidate in a first language in which the specified predicate can be used is presented, and the presented When a desired sentence pattern is designated from among the sentence pattern candidates of the first language, for each case constituting the designated sentence pattern of the first language,
Word candidates necessary for performing proper expression according to the meaning of the designated predicate are presented, and a desired word is designated from each of the presented word candidates for each case constituting the designated sentence pattern of the first language, At the time when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined, a sentence pattern code and a plurality of word codes respectively representing the fixed sentence pattern of the first language and the plurality of fixed words are generated, The generated sentence pattern code and the plurality of word codes are transmitted to a communication line as an intermediate expression for a user of the first language to perform a required character conversation with a user of the second language. In other words, the configuration of the character dialogue support method is adopted.

A second feature of the method of the present invention is that, in addition to the first feature of the method of the present invention, when a word constituting a predicate is specified in an input sentence of the first language from among words of the first language. Prior to presenting a sentence pattern candidate of the first language, a candidate form of the designated predicate is presented, and a plurality of word codes are generated from the proposed candidate form of the designated predicate. Is specified, and all the words constituting the designated sentence pattern of the first language including the designated inflected predicate are determined, and the present invention is directed to adopting a character dialogue support method.

A third feature of the method of the present invention is that, in addition to the first or second feature of the above-described method of the present invention, a sentence relating to the first language at the time of performing a character dialogue has a semantic meaning to the second language. When a typical sentence that can be translated is adopted and a user of the first language intends to create an input sentence of the first language using the typical sentence, a typical sentence candidate of the first language is presented. A desired typical sentence is designated from among the typical sentence candidates in the first language, and when the typical sentence is determined, a sentence code representing the type of the fixed typical sentence in the first language is generated. The present invention resides in adoption of a character dialogue support method in which a code is transmitted to a communication line as an intermediate expression for a user of a first language to perform a required character dialogue with a user of a second language. .

A fourth feature of the method of the present invention is that, in addition to the first, second or third features of the above-described method of the present invention, a user of the second language is required to communicate with a user of the first language. The intermediate representation generated for performing the character dialogue is transmitted to the second
When an incoming call is received from a user of the language to a user of the first language, a character code string indicating a sentence of the first language corresponding to the type of the received intermediate expression is generated, and the generated character code is generated. The present invention resides in a configuration of a character dialogue support method in which a sentence in the first language represented by a column is displayed so as to be visible to a user in the first language.

On the other hand, a first feature of the apparatus of the present invention is that a character for supporting a character dialogue between a user of a first language and a user of a second language via an arbitrary communication line is provided. A dialogue supporting device, wherein a word that can be mechanically translated into a second language is used as a term related to the first language when performing a character dialogue.
Word recording means for classifying and recording the words according to their semantic attributes and recording a word code for each word for obtaining the relevance to the corresponding word in the second language; Predicate sentence pattern recording means for recording, for each sentence pattern, a sentence pattern for obtaining the association with the corresponding sentence pattern of the second language, while recording the sentence pattern of the sentence pattern together with the types of predicates usable in the sentence pattern; From the words of the first language recorded in the word recording means, a word constituting a predicate is specified in an input sentence of the first language that the user of the first language intends to create, and a word that can use the specified predicate is designated. A sentence pattern candidate selection display means for selecting and displaying a sentence pattern candidate of one language from the predicate sentence pattern recording means; and a desired sentence pattern selected from the first language sentence pattern candidates selected and displayed by the sentence pattern candidate selection display means, The first For each rating that make up the specified sentence pattern of the word,
A word candidate selection display means for selecting and displaying word candidates necessary for performing an appropriate expression according to the meaning of the designated predicate from the word recording means, and a word candidate selected and displayed by the word candidate selection display means , A desired word is designated for each case constituting the designated sentence pattern of the first language,
When all of the designated sentence pattern of the language and the words constituting the sentence pattern are determined, the sentence pattern code and the plurality of word codes respectively representing the fixed sentence pattern of the first language and the plurality of fixed words are stored in the predicate sentence pattern recording means. A code reading means for reading out each word by referring to the word recording means, and a sentence pattern code and a plurality of word codes obtained by the code reading means are exchanged between a user of the first language and a user of the second language by a user. The present invention resides in adopting a configuration of a character dialogue support device having code transmitting means for transmitting a message to a communication line as an intermediate expression for performing a character dialogue.

The second feature of the device of the present invention is that, in addition to the first feature of the above-described device of the present invention, the first feature recorded in the word recording means.
Among words of the language, there is a predicate conjugation form recording means for recording conjugation forms relating to the words constituting the predicate in the input sentence of the first language, and the sentence pattern candidate selection and display means for the same feature is provided by selecting When a word constituting a predicate is designated in an input sentence of the first language, prior to selectively displaying a sentence pattern candidate of the first language, an inflected candidate of the designated predicate is selected from predicate inflected form recording means. The code reading means having the same feature allows the user to specify a desired conjugation form from the conjugation form candidates selected and displayed by the sentence pattern candidate selection display means, and designates a first language including the specified conjugation form predicate. Another object of the present invention is to adopt a configuration of a character dialogue support device that reads out a plurality of necessary word codes when all words constituting a sentence pattern are determined.

A third feature of the apparatus of the present invention is that, in addition to the first or second feature of the above-described apparatus of the present invention, a sentence relating to the first language at the time of performing a character dialogue is semantically translated into a second language. A typical sentence recording means for recording, for each typical sentence, a typical sentence which can be translated and for recording a sentence code for obtaining a relation with the corresponding typical sentence in the second language; When trying to create an input sentence in the first language using the typical sentence,
A typical sentence selection display means for selecting and displaying the typical sentence candidate of the first language from the typical sentence recording means, and a code reading means having the same feature, wherein the code reading means selected and displayed by the typical sentence selection display means is displayed. A desired typical sentence is designated from among the typical sentence candidates of one language, and when the typical sentence is determined, the first typical sentence is determined.
The sentence code representing the type of the confirmed typical sentence of the language is read out with reference to the typical sentence recording means, and the code transmitting means having the same feature converts the sentence code obtained by the code reading means into the first language. The present invention resides in adopting a configuration of a character dialogue support device that transmits to a communication line as an intermediate expression for a user to perform a required character dialogue with a user of a second language.

A fourth feature of the device of the present invention is that, in addition to the first, second or third features of the device of the present invention, a user of the second language is required to communicate with a user of the first language. When an intermediate expression for performing a character dialogue arrives from a user of the second language via a communication line, the intermediate expression is stored in a word recording unit, a predicate sentence type recording unit, and a typical sentence recording unit. By referring to the language conversion means, the language conversion means converts the sentence in the first language corresponding to the type into a character code string indicating a sentence in the first language, and the sentence in the first language represented by the character code string obtained by the language conversion means is converted. And an incoming sentence display means for displaying the first sentence so as to be visible to the user of the first language.

Further, the first recording medium of the program recording medium of the present invention
Is characterized by recording a character dialogue support program for supporting a character dialogue between a user of a first language and a user of a second language via an arbitrary communication line using a computer. A computer-readable recording medium, wherein a character dialogue support program causes a computer to form a predicate in an input sentence in a first language that a user of the first language intends to create from words prepared in a first language. When a word is specified, a first prepared first that the specified predicate can be used
A sentence pattern candidate of a language is presented, and when a desired sentence pattern is designated from the presented sentence pattern candidates of the first language, according to the meaning of the designated predicate for each case constituting the designated sentence pattern of the first language. Word candidates necessary for performing the appropriate proper expression are presented, and a first word is selected from the presented word candidates.
A desired word is designated for each case constituting the designated sentence pattern of the language, and when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined, the fixed sentence pattern of the first language and a plurality of fixed words are determined. Generate a pre-prepared sentence pattern code and a plurality of word codes respectively representing each type of word,
These generated sentence pattern codes and multiple word codes are
A recording medium recording a character dialogue support program, which is transmitted to a communication line as an intermediate expression for a user of a first language to perform a required character dialogue with a user of a second language. It is in.

A second feature of the program recording medium of the present invention is that the character dialogue support program according to the first feature of the program recording medium of the present invention is stored in a computer by a typical sentence prepared by a user of a first language. When an input sentence of the first language is to be created, a typical sentence candidate of the first language is presented, a desired typical sentence is designated from the presented typical sentence candidates of the first language, and the typical sentence is designated. Is determined, a prepared sentence code representing the type of the confirmed typical sentence of the first language is generated, and the generated sentence code is used by the user of the first language and the user of the second language. The present invention is to adopt a configuration of a recording medium recording a character dialogue support program, which is transmitted to a communication line, as an intermediate expression for performing a required character dialogue between them.

Further, the first data recording medium of the present invention
Is characterized by character dialogue support data used when using a computer to support a character dialogue between a user of the first language and a user of the second language via an arbitrary communication line. A recording medium on which a word that can be mechanically translated into a second language is classified according to its semantic attribute as a term related to the first language when the character dialogue support data is used for the character dialogue. A word file including a recorded area and an area in which a word code assigned to each word in order to obtain the relevance of the corresponding word in the second language; An area in which a sentence pattern is recorded together with the types of predicates usable in the sentence pattern, and an area in which a sentence pattern code assigned to each sentence pattern is recorded in order to obtain the association with the corresponding second language sentence pattern. Predicate sentence pattern file and the first word Made and a predicate conjugations file having a region which records the inflected forms related words constituting the predicate of the word, in the configuration adopting a recording medium recording a character conversation assistance data.

A second feature of the data recording medium of the present invention is that the character dialogue support data in the first feature of the data recording medium of the present invention is used as a text according to a first language for performing a character dialogue. It has an area in which typical sentences that can be semantically translated into a language are recorded, and an area in which a sentence code assigned to each typical sentence is recorded in order to obtain the association with the corresponding typical sentence in the second language. The present invention resides in the adoption of a configuration of a recording medium having a typical sentence file and recording character dialogue support data.

[0032]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings, with reference to the drawings. In this embodiment, a user who uses Japanese as the first language (a user of one of the constituent terminals) and a user who uses English as the second language (a user of the other constituent terminal) A description will be given of an example in which a character dialogue is performed between the users.

(Example of Apparatus) FIG. 1 is a conceptual configuration diagram of a character support dialogue apparatus according to an embodiment of the present invention, and FIG. 2 is a detailed configuration diagram of a display unit shown in FIG.

As shown in both figures, the character support dialogue apparatus β according to this example of the device mainly includes a keyboard 5 and a mouse 6 constituting an input device of a terminal, and a display unit 7 constituting an output device. The display unit 7 includes an input sentence creation unit 71, an essay display unit 72, a confirmed transmission instruction unit 73, a transmitted sentence display unit 74, and an incoming sentence display unit 75 on the screen. It is composed.

In addition to the input / output devices described above, the character support dialogue device β is provided with a code transmission unit 9 on the output side of the communication line 8.
a, a language converter 9b is arranged on the input side of the communication line 8, and a typical sentence recording unit 10, a predicate sentence type recording unit 11, a predicate sentence recording unit, and a storage unit of the terminal. It has a unit 12, a Be sentence type recording unit 13, and a word dictionary 14.

Here, the input sentence creating unit 71 of the display unit 7
Further, when a Japanese user intends to create a Japanese input sentence using a typical sentence such as a greeting sentence, a typical sentence candidate which is selected from the typical sentence recording unit 10 and displayed. From the Japanese word recorded in the word dictionary 14, the sentence selection unit 71a includes a predicate (including "adjective" in addition to "verb") in a Japanese input sentence to be created by a Japanese user. A predicate sentence pattern selection unit 71b (part of a sentence pattern candidate selection display unit) that allows the user to specify a word to be composed and selects and displays a Japanese sentence pattern candidate in which the specified predicate can be used from the predicate sentence pattern recording unit 11; When a user of a word intends to create a Japanese sentence pattern corresponding to an English Be sentence pattern (including a sentence pattern with an "incomplete intransitive verb" other than a "Be verb"), the Japanese sentence pattern (candidate) is Select and display from the Be sentence pattern recording unit 13 Constructed and a e sentence pattern selection unit 71c (a part of the sentence pattern candidate selection displaying means).

The reason why the Be sentence type selection unit 71c is provided independently of the predicate sentence type selection unit 71b is that, for Japanese users, the object such as "~ ha ~ da" explains the subject. This is because in Japanese sentence pattern, the recognition that Japanese "da" is a word corresponding to a predicate is sparse.

The composition display section 72 displays the contents of a Japanese input sentence being input by a Japanese user in real time before transmitting various codes from the code transmission section 9a to the communication line 8. The confirmed transmission instructing unit 73 uses the mouse 6 to transmit various codes corresponding to the Japanese input sentence displayed on the composition display unit 72 by using the confirmed transmission button 73a on the display. This is an area to be determined by an operation or the like.

The sentence display section 74 is an area for historically displaying the contents of a Japanese input sentence corresponding to various codes already transmitted from the code transmission section 9a to the communication line 8,
Incoming sentence display unit 74 displays communication line 8 from the English user side.
This is an area for historically displaying the contents of Japanese (incoming sentence) corresponding to various codes generated from the English sentence sent to the language conversion unit 9b via the PC (both will be described later in detail).

The various codes described above include a “sentence code” generated when a Japanese user attempts to create a Japanese input sentence based on a typical sentence, and a “sentence code” generated when a predicate is input. There are a “sentence pattern code” and a “word code” to be generated, and these codes are recorded in the corresponding recording units, and the CPU (not shown) of the constituent terminal
(Code reading means).

More specifically, the sentence code is read out by specifying a desired typical sentence by a Japanese user from among the Japanese typical sentence candidates selected and displayed by the typical sentence selection unit 71a.
Thereafter, the operation is performed when the typical sentence is confirmed by operating the confirm transmission button 73a. On the other hand, the reading of the sentence pattern code and the word code is selectively displayed by the predicate sentence pattern selection unit 71b (including the Be sentence pattern selection unit 71c). A desired sentence pattern is designated by the Japanese user from the Japanese sentence pattern candidates, and thereafter, the designated sentence pattern of the first language and all the words constituting the sentence pattern are confirmed by operating the confirm transmission button 73a. Done at the time.

At this time, the code transmitting section 9a transmits the various codes generated as described above to a required character dialogue between a Japanese user using the present apparatus β and a non-illustrated English user. Functioning as an intermediate expression for performing the transmission to the communication line 8.

Further, the language conversion unit 9b is adapted to provide an intermediate expression for the user of the other party to perform a required character dialogue with a Japanese user using the present apparatus β, and to convert the English language of the other party into English. When an incoming call is received from the user's side via the communication line 8, the intermediate expression is referred to the corresponding recording units,
Type (sentence code or sentence pattern code and word code)
It functions to convert to a character code string indicating a Japanese sentence according to. Then, the character code string obtained by the language conversion unit 9b is transferred to the incoming sentence display unit 75 of the display unit 7, and the incoming sentence display unit 75 converts the Japanese sentence represented by the above character code string into a sentence. The information is displayed historically so that it can be visually recognized by a Japanese user.

On the other hand, each recording unit constituting the storage device of the terminal will be described. First, the typical sentence recording unit 10 performs a semantic translation into English as a sentence related to Japanese at the time of performing a character dialogue. A typical sentence such as the above-mentioned greeting sentence is recorded. More specifically, as data for performing the character dialogue support, the relationship between the above-described typical sentence recording area and the corresponding English typical sentence is recorded. In order to obtain a typical sentence, a typical sentence file including an area in which the above-described sentence code assigned to each typical sentence is recorded is stored.

The predicate sentence pattern recording unit 11 records a Japanese sentence pattern at the time of conducting a character dialogue together with types of predicates that can be used in the sentence pattern.
(A diagram showing a partial form of a predicate sentence pattern file held in the predicate sentence pattern recording unit 11 shown in FIG. 1), as an area for recording the above sentence pattern and the type of the predicate as data for supporting character dialogue, , A predicate sentence pattern file having an area for recording the above-mentioned sentence pattern code assigned to each sentence pattern in order to obtain the relevance with the corresponding English sentence pattern (in FIG. Corresponding "English sentence pattern"
Are also shown).

The predicate inflected form recording unit 12 records the inflected forms of the words constituting the predicate among the Japanese words recorded in the word record 14, and is described in detail in FIG. 4 (shown in FIG. 1). As shown in FIG. 2), a predicate utilization form having an area where the above utilization form is recorded as data for supporting the character dialogue, as shown in FIG. (In the figure, the corresponding "English utilization form" is also shown for reference.)

The Be sentence pattern recording unit 13 records a Be sentence pattern related to Japanese when a character dialogue is performed. However, the Japanese Be sentence pattern is based on the English Be sentence pattern,
Is a Japanese sentence pattern in the form of

The word dictionary 14 is a term in which words that can be mechanically translated into English are classified and recorded according to their semantic attributes as terms related to Japanese when performing a character dialogue. As shown in FIG. 5 (a diagram showing a partial form of a word file held in the word dictionary 14 shown in FIG. 1), as data for supporting character dialogue, an area in which the above-mentioned words are recorded and In order to obtain the relevance with the English word, a word file having an area in which the above-mentioned word code added to each word is recorded is stored. English words ”are also displayed.)

(Example of Method) Next, an example of a method applied to the character dialogue support apparatus β according to the example of the apparatus configured as described above will be described.

FIG. 6 is a flowchart for explaining the operation of the character dialogue support apparatus β shown in FIG. 1 relating to the code transmission.

As shown in the figure, when transmitting the code, the device β first generates a required Japanese input sentence by using a typical sentence or by inputting a predicate.
Alternatively, a request for selecting a creation format, such as whether to use a Be sentence pattern, is presented, and the process is started (see also FIG. 2).

Here, the Japanese user who operates the present apparatus β is operated by the typical section selecting section 71a in the input sentence creating section 71.
A, if you select designated by the cursor operation of the mouse 6, and the like, an apparatus β, "Good morning", "Hello", and select the typical sentence candidate of Japanese, such as "bye-bye" from the typical sentence storage section 10 display (ST1), a desired typical sentence is specified by the Japanese user (ST2),
At the time when the fixed transmission button 73a instructs the fixed transmission (ST3; Y), a sentence code corresponding to the specified Japanese typical sentence is read out from the typical sentence recording unit 10 and generated (ST4).

On the other hand, the user of Japanese inputs the input sentence
When the predicate such as “eat” is input by selecting and specifying the predicate sentence pattern selection unit 71b in ST1 (ST5), the apparatus β
The input form candidate of the input predicate is stored in the predicate use form recording unit 1
2 (ST6), and when the user specifies a desired inflected form such as "I want to eat" (ST7), a Japanese sentence pattern candidate in which the specified inflected form predicate can be used is displayed. It is selected and displayed from the predicate sentence pattern recording unit 11 (ST8). When the desired sentence pattern is designated in this way (ST9), every time a case constituting the designated sentence pattern is selected (ST10), the apparatus β changes the meaning (attribute) of the preceding designated predicate. A word candidate for performing the appropriate proper expression is selected from the word dictionary 14 and displayed (ST11), and a desired word is designated by the user (ST12), and until the word input is completed for all cases. The process up to is repeated.

The required word input is completed (ST13;
Y) Then, when the fixed transmission is instructed by the fixed transmission button 73a (ST14; Y), the apparatus β reads out the sentence pattern code corresponding to the specified Japanese sentence pattern from the predicate sentence pattern recording unit 11 and generates it. And a plurality of word codes corresponding to the specified Japanese word for all cases,
It is read out from the word dictionary 14 and generated (ST15).

On the other hand, the user of Japanese inputs the input sentence
When the Be sentence pattern selection unit 71c in No. 1 is selected, the device β outputs “da”, “da da”,
Inflection type candidates such as "Would" and "Maybe" are selected from the predicate inflection record and displayed (ST16), and when a desired usage form such as "Would" is designated by the user. (ST17), a Japanese sentence pattern candidate in which the specified inflected form predicate can be used is selected from the Be sentence pattern recording unit and displayed (ST18). When the desired sentence pattern is designated in this way (ST19), the device β every time a case constituting the designated sentence pattern is selected (ST20), the meaning (attribute) of the preceding designated utilization form. Are displayed (ST21), a desired word is designated by the user (ST22), and the above processing is repeated until word input is completed for all cases. Execute.

The required word input is completed (ST23;
Y) Then, when the confirmed transmission is instructed by the confirmed transmission button 73a (ST24; Y), the apparatus β reads out the sentence pattern code corresponding to the designated Japanese sentence pattern from the Be sentence pattern recording unit 13 and generates it. And a plurality of word codes corresponding to the specified Japanese word for all cases,
It is read out from the word dictionary 14 and generated (ST15).

When a Japanese input sentence is created, if it is a typical sentence, the above-mentioned processing ST
At the time when the desired typical sentence is determined in Step 2, if the word is determined by the predicate or the Be sentence pattern, the content is displayed on the composition display unit 72 one by one when the word is determined in the process ST12 or ST22. (In the case of creation by a typical sentence, “bye bye” shown in the figure). For this reason, the Japanese user who operates the present apparatus β can proceed with required input work while checking the state of creation of his / her own Japanese input sentence.

Next, the apparatus β performs processing S in the typical sentence input process.
When a sentence code is generated in T4, a character code string indicating a Japanese sentence is generated by referring to the typical sentence recording unit 10 from the sentence code (ST25), and a predicate input or Be sentence type input process is performed. When the sentence pattern code and the word code are generated in ST15, the predicate sentence pattern recording unit 11 or the Be sentence pattern recording unit 12 and the word dictionary 14 are respectively referred to from the sentence pattern code and the word code as necessary. Similarly, a character code string indicating a Japanese sentence is generated (ST26), and the Japanese character code string generated in each of these steps is historically displayed on the sentence display unit 74 as a Japanese sentence (ST27). .

At the same time, the code transmitting section 9a of the device β
Means that the Japanese user uses the sentence code generated in the process ST4 of the typical sentence input process or the sentence pattern code and the word code generated in the process ST15 of the predicate input or Be sentence type input process. A call is transmitted to the communication line 8 as an intermediate expression for performing a required text conversation with the user (ST28).

The above-described procedure for generating a Japanese character code string will be described by taking a predicate input as an example.
It looks like this: That is, the apparatus β is the predicate sentence pattern recording unit 1
First, by referring to the predicate sentence pattern file held in No. 1 (see “Japanese sentence pattern” in FIG. 3), first, a configuration defined by a file to which a sentence pattern code according to the specification of the Japanese user is added Recognize the word order, attributes, and necessity of the words, and the character code of the word that governs the relation between words such as "particles", and then, in each recognized sentence pattern information, indicate the Japanese language indicated by the word code Is sequentially applied to generate a required Japanese character code string, and the Japanese character code string is displayed by the sentence display unit 74.

Next, FIG. 7 is a flowchart for explaining the operation of the character dialogue support apparatus β shown in FIG. However, this figure shows the operation relating to the incoming code of the device used on the English user side.

As shown in the figure, when a code is received from the communication line 8, the device β of the other party used by the English user firstly receives the received code in the language conversion unit 9b constituting the input terminal. Is determined (ST31).

As a result of the discrimination, when the received code is a sentence code meaning a typical sentence (ST31; Ye
s) The language conversion unit 9b reads the typical sentence specified by the sentence code related to the incoming call from the typical sentence recording unit 10 and generates an English sentence (a character code string indicating an English sentence) (ST3).
2), the apparatus β displays the generated English sentence on the incoming sentence display unit 75 of the display unit 7 (ST33).

On the other hand, when the code received from the communication line 8 is not a sentence code (ST31; No),
The language conversion unit 9b determines that the code related to the incoming call is composed of a sentence pattern code and a word code, and determines that the predicate sentence pattern recording unit 11, the predicate utilization form recording unit 12, and the Be sentence pattern recording unit 1
3 and the word dictionary 14, respectively, and reads out words including the sentence pattern and the predicate conjugation form specified by the plurality of codes from each of these recording units to restore and generate a required English sentence (English character code string). (ST34). And
The device β displays the generated English sentence on the incoming sentence display unit 75 of the display unit 7 (ST33).

The procedure for generating an English character code string on the English user side will be described below. That is, the other party's device β refers to the predicate sentence pattern file held in the predicate sentence pattern recording unit 11 (see “English sentence pattern” in FIG. 3), and firstly, the sentence pattern code associated with the designation of the Japanese user. Recognize the word order, attributes, and necessity of the constituent words defined by the file to which is added, and the character code of the word that governs the relation between words such as "preposition", and then recognizes each recognized sentence pattern By sequentially applying character codes constituting English words indicated by the word codes to the information, a required English character code string is generated, and the English character code string is displayed by the incoming sentence display unit 75.

Next, the state of operation when the Japanese user actually operates the character dialogue support apparatus β will be described.

First, when a Japanese user who operates the present apparatus β inputs a Japanese input sentence, the Japanese input sentence is displayed on the composition display unit 72 of the display unit 7 in real time. When the user clicks the confirm transmission button 73a, the Japanese input sentence is converted in the code transmitting unit 9a and transmitted to the other party's device, and the contents of the Japanese text corresponding to the transmitted code are displayed. , Self equipment β
Are displayed historically in the order of their transmission times on the transmission sentence display section 74.

At this time, the sentence display unit 75 of the other party's device β displays historically the English sentences corresponding to various codes generated from Japanese, while the incoming sentence display unit 75 of the present device β , Japanese sentences corresponding to various codes generated from the other party's English are displayed in a history. In other words, the user of both devices can communicate with each other by a required character dialogue without being aware of the language used by the other device in his or her device.

When a typical sentence is used to create an actual Japanese input sentence, a typical sentence selection unit 71 of the input sentence creation unit 71
In a, a typical sentence recording unit 10 is set by a pull-down menu.
The typical sentence itself, such as a greeting sentence, recorded as a selection object is presented as a selection target, and the user of the apparatus β is caused to select and specify a typical sentence to be transmitted.

The predicate sentence pattern selection unit 71b first displays a predicate candidate, and when the predicate candidate is designated by a Japanese user, the template of the sentence pattern usable by the designated predicate is represented by its attribute and predicate. It is displayed in Japanese with.

For example, when the above-mentioned predicate "eat" is specified, the word input box 71b 'shown in FIG. 8 (a diagram showing a configuration example of the word input box of the predicate sentence type selection unit shown in FIG. 1) is entered. Displays a template such as "<[when]> [who] (eats) [food]." (The above template is also shown in the figure).

Here, the “(eat)” portion of the above-mentioned template can be further designated for its utilization form, ie, mode (form) and tense (tense). Specifically, when the user clicks on the “(eat)” portion to select the area, the inflected form corresponding to the predicate “eat” is read from the predicate inflected form recording unit 12, and the inflected form of the inflected form is read out. Everything is displayed by a pull-down menu. In this pull-down menu, there are displayed multiple usage forms such as "Eat", "Eat", "Eat", "Eat", "Eat?" Selection candidate for the person. However, when the sentence pattern changes depending on the mode such as the passive voice, a different template is read from the predicate sentence pattern recording unit 11 instead of the above template.

As described above, when the predicate (and its conjugation form) is determined, the semantic attributes and cases of the word candidates that differ depending on the meaning are determined, and enclosed in “[]” in the above sentence pattern. A selection range for the portion is determined. Here, if the Japanese user selects and specifies the portion of “[food]” as a candidate for the semantic attribute, the “food” that can be specified is
Appears in a pull-down menu, and the user
What is intended by the user can be selected and designated from the list. Note that “<>” in the drawing means that the selection of that part can be omitted.

The input to the Be sentence pattern selection unit 71c is almost the same as that in the predicate sentence pattern selection unit 71b, and the template is, for example, "<[when]><
[What]> [what] is [what] (da). ".

Here, the "(da)" part can designate a mode and a tense as its utilization form. Specifically, if you click on the “(da)” part and select the relevant area, its utilization forms include, for example, “da”, “was”, “was”, “may”, etc. Appearing in pull-down menus, these are selection candidates for Japanese users. In the case of "[what]",
Various "nouns" are selection candidates. However, since the number of nouns recorded in the candidate word dictionary 14 is enormous,
In this case, for each of the predetermined number of readings and characteristic spellings,
A multi-stage pull-down menu in which each is a selection candidate is presented. Note that presenting a multi-stage pull-down menu when the number of selection candidates is enormous is not limited to this example.

As described above, in the character dialogue support apparatus β, the input sentences that can be created include the typical sentence recording unit 10, the predicate sentence type recording unit 11, the predicate utilization type recording unit 12, and the Be sentence type recording unit 1.
3, and restrictions are placed so that it can be created only within sentence patterns and word candidates recorded in the word dictionary 14. However, the character input by the alphabet is performed by the composition display section 72 of the display section 7.
However, the input alphabet is accepted as a component of the sentence (English sentence). In this way, proper nouns that do not exist in the menu are written in alphabet,
Typical sentence selection unit 71a, predicate sentence type selection sentence 71b, and Be
The alphabet can be transmitted immediately without being controlled by the sentence pattern selection unit 71c.

As described above, according to the example of the apparatus and the example of the method according to the present embodiment, Japanese and English users can
It is possible to communicate with each other by using a required character dialogue while using each other's own language and without being aware of the language of the other party. Further, in the present apparatus example and method example, translation can be executed with a smaller processing amount as compared with the conventional automatic translation, so that the time required for the translation processing is significantly reduced, and Since there is no problem such as mistranslation or inability to translate, it is possible to communicate with the required characters in real time and reliably.

(Example of Recording Medium) Finally, a description will be given of a character dialogue support program for supporting a required character dialogue by using a computer, and a recording medium on which character dialogue support data used for the support are recorded. I do.

As is clear from the above description, the present invention can be realized by a general-purpose computer (terminal). Therefore, the method (procedure) for supporting the required character dialogue provided by the present invention is described on a CD-ROM.
Is recorded as a character dialogue support program on a recording medium such as a computer, and the character dialogue support program recorded on this recording medium is introduced into a computer having the hardware configuration of the apparatus provided by the present invention and executed. Thereby, the method and apparatus of the present invention can be easily realized.

Here, when the required Japanese input sentence is created by a predicate input, the character dialogue support program to be recorded on the above-mentioned recording medium is as follows:
When a word constituting a predicate is specified in a Japanese input sentence to be created by a Japanese user from pre-prepared Japanese words, the prepared Japanese language in which the specified predicate can be used. A step of presenting a sentence pattern candidate of the word;
(B) When a desired sentence pattern is designated from the presented Japanese sentence pattern candidates, a proper expression corresponding to the meaning of the designated predicate is performed for each case constituting the designated Japanese sentence pattern. A procedure for presenting necessary prepared word candidates;
(C) From the presented word candidates, a desired word is specified for each case constituting the designated Japanese sentence pattern, and when the Japanese designated sentence pattern and the words constituting the sentence pattern are all determined, (D) generating a prepared sentence pattern code and a plurality of word codes each representing each type of the determined sentence pattern and the plurality of determined words of the Japanese language;
These generated sentence pattern codes and multiple word codes are
As an intermediate expression for a Japanese user to perform a required character dialogue with an English user, a description having a description for executing a procedure for transmitting a call to the communication line 8 is adopted.

When a required Japanese input sentence is also created by a typical sentence, in addition to the above-described steps (a) to (d), (e) a Japanese user prepares the computer in advance. A procedure for presenting a Japanese typical sentence candidate when an attempt is made to create a Japanese input sentence based on the given typical sentence;
(F) When a desired typical sentence is designated from the presented Japanese typical sentence candidates, and when the typical sentence is determined, a prepared sentence code representing the type of the determined Japanese typical sentence is entered. And (g) transmitting the generated sentence code to the communication line 8 as an intermediate expression for a Japanese user to perform a required text conversation with an English user. May be added to the above-described character dialogue support program for inputting a predicate.

On the other hand, the present invention is realized by using data relating to sentence patterns and words recorded in each recording unit constituting a storage device of a computer. Therefore, these required data are recorded as character dialogue support data on a recording medium such as a CD-ROM, and the character dialogue support data recorded on this recording medium is introduced into each recording unit, and this is transferred to the above-described character The method and apparatus of the present invention can be easily realized by executing the dialogue support program.

Here, when a required Japanese input sentence is created by a predicate input, the character dialogue support data to be recorded on the above-mentioned recording medium includes (A) a term related to Japanese at the time of character dialogue. As an area in which words that can be mechanically translated into English are classified and recorded according to their semantic attributes,
A word file (introduced into the word dictionary 14) having an area in which a word code assigned to each word is recorded in order to obtain relevance to the corresponding English word; An area in which the sentence pattern of a word is recorded together with the types of predicates that can be used in the sentence pattern, and an area in which a sentence pattern code assigned to each sentence pattern in order to obtain the association with the corresponding English sentence pattern is provided. Predicate sentence type file (introduced into the predicate sentence type recording unit 11) and (C) a predicate inflection type file having an area in which an inflected type related to a word constituting a predicate among Japanese words is recorded (predicate inflection type recording unit 12 To be introduced).

When a Japanese input sentence is also created by a typical sentence, in addition to the above-mentioned files (A) to (C), (D) a sentence relating to Japanese when performing a character dialogue , A region having a typical sentence that can be semantically translated into English, and a region having a sentence code recorded for each typical sentence in order to obtain relevance to the corresponding English typical sentence. A sentence file (introduced into the typical sentence recording unit 10) may be added to the above-described character dialogue support data for inputting a predicate.

As described above, the embodiment of the present invention will be described with respect to the example of the apparatus, the example of the method, and the example of the recording medium by a user who uses Japanese as the first language and a user who uses English as the second language. The above description has been made by taking as an example the case where a character dialogue is performed between the two languages. However, if different languages are used as the first language and the second language, a dialogue supporting method and means having a translation function between the respective languages can be optionally used. Can be realized.

For example, in contrast to the above embodiment, if the first language is English and the second language is Japanese, the user who uses English will be able to communicate with the other user who uses Japanese. It is possible to communicate with each other through character dialogue.

Further, not only the first language and the second language but also other languages (the third language)
If the sentence, sentence pattern, predicate utilization form, and word of the language, the fourth language,...) Are recorded, the translation into the relevant language becomes possible.
At this time, by replacing the element for each language, it is possible to perform the conversion or the replacement of the language to be converted. Therefore, translation communication in multiple languages can be performed simply by changing a few settings. And a small amount of processing.

The present invention is not necessarily limited to the above-described method and means, but can be appropriately modified and implemented within the scope of achieving the object of the present invention and having the effects described below. It is something.

[0089]

As described above in detail, according to the present invention, although the use of a language in character translation is restricted, a situation in which an input sentence is incorrectly translated or cannot be translated in a target language is completely avoided. Since the translation can be completed with a small amount of processing, the time required for the translation processing can be greatly reduced, and communication with the required characters can be performed in real time and reliably. As a result, it is possible to smoothly communicate between the user of the first language and the user of the second language between the remote points connected by the communication line while using the characters of the own language. .

Further, according to the present invention, the sentences, sentence patterns, predicate utilization forms, and words relating to the first language and the second language recorded in each recording means are simply changed to those of another language. Since it is possible to easily extend the translation to multilingual translation, it is possible to flexibly cope with character conversations concerning arbitrary two types of languages.

[Brief description of the drawings]

FIG. 1 is a conceptual configuration diagram of a character support dialogue apparatus according to an embodiment of the present invention.

FIG. 2 is a detailed configuration diagram of a display unit shown in FIG.

FIG. 3 is a diagram showing a partial form of a predicate sentence type file held in a predicate sentence type recording unit shown in FIG. 1;

FIG. 4 is a diagram showing a partial form of a predicate inflection type file held in a predicate inflection type recording unit shown in FIG. 1;

FIG. 5 is a diagram showing a partial form of a word file held in the word dictionary shown in FIG. 1;

FIG. 6 is a flowchart for explaining an operation related to code transmission of the character dialogue support apparatus shown in FIG. 1;

FIG. 7 is a flowchart illustrating an operation of the character dialogue support apparatus shown in FIG.

FIG. 8 is a diagram showing a configuration example regarding a word input field of a predicate sentence type selection unit shown in FIG. 1;

FIG. 9 is a conceptual configuration diagram of a conventional chat system, showing an example in which a translation means for communicating with each other through character dialogue is applied to the conventional chat system.

[Explanation of symbols]

 α: Chat system 1a, 1b: Input device (keyboard) 2a, 2b: Display device 3: Communication line 4a, 4b: Translation device β: Character dialogue support device 5: Keyboard 6: Mouse 7: Display unit 71: Input sentence creation Part 71a ... Typical sentence selecting part 71b ... Predicate sentence type selecting part 71b '... Word input field 71c ... Be sentence type selecting part 72 ... Composition display part 73 ... Definite transmission instruction part 73a ... Definite transmission button 74 ... Transmission sentence display part 75 ... Reception Sentence display unit 8 Communication line 9a Code transmission unit 9b Language conversion unit 10 Typical sentence recording unit 11 Predicate sentence type recording unit 12 Predicate utilization type recording unit 13 Be sentence type recording unit 14 Word dictionary

Claims (12)

[Claims] 1. A character dialogue supporting method for supporting a character dialogue between a user of a first language and a user of a second language via an arbitrary communication line, wherein the character dialogue is provided. The term used in the first language is limited to a word that can be mechanically translated into a second language, and is created by the user of the first language from the words in the first language. When a word constituting a predicate is specified in an input sentence of the first language to be specified, a sentence pattern candidate of the first language in which the specified predicate can be used is presented, and the presented sentence pattern candidate of the first language is presented. When a desired sentence pattern is designated from among them, word candidates necessary for performing proper expression according to the meaning of the designated predicate are presented for each case constituting the designated sentence pattern of the first language. A specified sentence pattern of the first language from the word candidates thus selected When a desired word is designated for each case, and when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined, the fixed sentence pattern of the first language and each of the plurality of fixed words are respectively classified. A sentence pattern code and a plurality of word codes are generated, and the generated sentence pattern code and the plurality of word codes are converted by the user of the first language into a character required by the user of the second language. A character dialogue support method, wherein a call is transmitted to the communication line as an intermediate expression for performing a dialogue. 2. When a word constituting a predicate is specified in an input sentence in the first language from among words in the first language, the designation is performed prior to presenting a sentence pattern candidate in the first language. Presenting the inflected form candidate of the predicate; and generating the plurality of word codes, a desired inflected form is designated from the presented inflected form candidates of the designated predicate, and the first inclusive form including the designated inflected form predicate is included.
The character dialogue supporting method according to claim 1, wherein the method is performed when all words constituting the designated sentence pattern of the language are determined. 3. A typical sentence that can be semantically translated into a second language is used as a sentence related to the first language when performing the character dialogue, and a user of the first language uses the typical sentence. When an input sentence of the first language is to be created by the method, a typical sentence candidate of the first language is presented, and a desired typical sentence is designated from the presented typical sentence candidates of the first language, When the typical sentence is determined, a sentence code representing the type of the confirmed typical sentence of the first language is generated, and the generated sentence code is used by the user of the first language to use the second language. The character dialogue support method according to claim 1, wherein the call is transmitted to the communication line as an intermediate expression for performing the required character dialogue with a person. 4. An intermediate expression generated by a user of the second language for performing a required character dialogue with a user of the first language, the intermediate expression of the second language via the communication line. When a call is received from a user to the user of the first language, a character code string indicating a sentence in the first language corresponding to the type of the received intermediate expression is generated, and the generated character The method according to claim 1, 2 or 3, wherein the first language sentence represented by the code string is displayed so as to be visible to a user of the first language. 5. A character dialogue support apparatus for supporting a character dialogue between a user of a first language and a user of a second language via an arbitrary communication line, wherein the character dialogue is provided. The words that can be mechanically translated into a second language are classified and recorded according to their semantic attributes as the terms related to the first language when performing the above, and the relevance with the corresponding words in the second language is recorded. A word recording means for recording a word code for obtaining each word, and a sentence pattern of the first language at the time of performing the character dialogue, together with a type of predicate usable in the sentence pattern, and A predicate sentence pattern recording means for recording a sentence pattern code for obtaining the relevance with the sentence pattern of the second language for each sentence pattern; and from words of the first language recorded in the word recording means,
The user of the first language is caused to specify a word constituting a predicate in the input sentence of the first language to be created, and a sentence pattern candidate of the first language in which the specified predicate can be used is recorded from the prewritten word sentence pattern recording means. A sentence pattern candidate selection display means for selecting and displaying, and a desired sentence pattern is designated from among the first language sentence pattern candidates selected and displayed by the sentence pattern candidate selection display means, and a case constituting a designated sentence pattern of the first language is selected. Word candidate selection display means for selecting and displaying, from the word recording means, a word candidate necessary for performing an appropriate expression according to the meaning of the designated predicate; From the word candidates, a desired word is designated for each case constituting the designated sentence pattern of the first language, and when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined, The sentence pattern code and a plurality of word codes representing said first language determined sentence patterns and a plurality of defined words each type, respectively,
Code reading means for reading by referring to the pre-written word sentence pattern recording means and the word recording means, respectively; and the sentence pattern code and the plurality of word codes obtained by the code reading means are read by the user of the first language. A character dialogue support device, comprising: a code transmission means for transmitting to the communication line as an intermediate expression for performing the required character dialogue with a user of a second language. 6. A predicate inflection type recording means for recording inflection forms relating to words constituting a predicate in an input sentence in the first language among words in the first language recorded in the word recording means, The sentence pattern candidate selection display means is configured to specify a word constituting a predicate in an input sentence in the first language from among words in the first language,
Prior to selectively displaying the sentence pattern candidate of the language, the inflected form candidate of the designated predicate is selected from the preceding word inflected form recording means and displayed, and the code reading means is selected by the sentence pattern candidate selection display means. A desired conjugation form is designated from the selected conjugation form candidates, and when all the words constituting the specified sentence pattern of the first language including the designated conjugation form predicate are determined, the plurality of required word codes are determined. The character dialogue support device according to claim 5, wherein the character dialogue is read. 7. A typical sentence that can be semantically translated into a second language is recorded as a sentence related to the first language at the time of performing the character dialogue, and a corresponding typical sentence in the second language is recorded. A typical sentence recording means for recording a sentence code for each typical sentence for obtaining relevance; and when the user of the first language attempts to create an input sentence of the first language by the typical sentence, A typical sentence selection display means for selecting and displaying a typical sentence candidate in the first language from the typical sentence recording means; and the code reading means comprises a first sentence selected and displayed by the typical sentence selection display means.
A desired typical sentence is designated from among the typical sentence candidates of the language, and when the typical sentence is determined, a sentence code representing the type of the determined typical sentence of the first language is read out with reference to the typical sentence recording means. Wherein the code transmitting means transmits the sentence code obtained by the code reading means to the user of the first language for performing the required character dialogue with the user of the second language. The character dialogue support device according to claim 5, wherein the intermediate expression is transmitted to the communication line as an intermediate expression. 8. An intermediate representation for a user of the second language to perform a required character dialogue with a user of the first language,
When an incoming call is received from the user of the second language via the communication line, the intermediate expression is referred to by referring to the word recording unit, the pre-written word sentence type recording unit, and the typical sentence recording unit, Language conversion means for converting into a character code string indicating a sentence in the first language according to the type; and a sentence in the first language represented by the character code string obtained by the language conversion means, The character dialogue support device according to claim 5, further comprising: an incoming sentence display unit that displays the sentence so as to be visible to a user in one language. 9. A character dialogue support program for supporting a character dialogue between a user of a first language and a user of a second language via an arbitrary communication line using a computer. A recorded medium, wherein the character dialogue support program stores, in a computer, an input sentence in a first language that a user of the first language intends to create from words of the first language prepared in advance. When a word constituting the predicate is designated, a prepared sentence pattern candidate of the first language in which the designated predicate can be used is presented, and a desired sentence pattern is selected from the presented sentence pattern candidates of the first language. At the time of designation, for each case constituting the designated sentence pattern of the first language, a word candidate prepared in advance for performing an appropriate expression according to the meaning of the designated predicate is presented, and the presented Said word From among them, a desired word is designated for each case constituting the designated sentence pattern of the first language, and when the designated sentence pattern of the first language and the words constituting the sentence pattern are all determined, the first word of the first language is determined. A pre-prepared sentence pattern code and a plurality of word codes respectively representing a fixed sentence pattern and a plurality of fixed words are generated, and the generated sentence pattern code and the plurality of word codes are generated by a user of the first language. Wherein a character conversation support program is transmitted to the communication line as an intermediate expression for performing the required character conversation with the user of the second language. 10. The character dialogue support program according to claim 1, wherein when the user of the first language attempts to create an input sentence of the first language using a typical sentence prepared in advance, the first language of the first language is used. Is selected, and a desired typical sentence is specified from the presented typical sentence candidates in the first language, and when the typical sentence is determined, the type of the determined typical sentence in the first language is changed. Generating a pre-prepared sentence code representing the sentence code, and using the generated sentence code as an intermediate representation for the user of the first language to perform the required character dialogue with the user of the second language The recording medium according to claim 9, wherein the character communication support program is transmitted to the communication line. 11. A character dialogue performed between a user of a first language and a user of a second language via an arbitrary communication line,
It is a recording medium which records character dialogue support data used when supporting using a computer, wherein said character dialogue support data is a term relating to said first language when performing said character dialogue, to a second language An area in which words that can be mechanically translated are classified according to their semantic attributes and recorded, and an area in which a word code assigned to each word in order to obtain relevance with the corresponding word in the second language is recorded A word file comprising: a region in which a sentence pattern of the first language when performing the character dialogue is recorded together with types of predicates usable in the sentence pattern; And a predicate sentence pattern file having an area in which a sentence pattern code assigned to each sentence pattern is recorded, and a predicate having an area in which an inflected form relating to a word constituting a predicate among words in the first language is recorded. A recording medium recording character dialogue support data, characterized by having a utilization file. 12. The character dialogue support data corresponds to an area where a typical sentence that can be semantically translated into a second language is recorded as a sentence related to the first language when performing the character dialogue. 12. A typical sentence file having an area for recording a sentence code assigned to each typical sentence in order to obtain the association with the typical sentence of the second language, comprising a typical sentence file. A recording medium on which character dialogue support data is recorded.