WO2010010920A1

WO2010010920A1 - Information processing apparatus and information processing method

Info

Publication number: WO2010010920A1
Application number: PCT/JP2009/063174
Authority: WO
Inventors: 直子篠崎; 稔幸奥西; 晃一杉山
Original assignee: シャープ株式会社
Priority date: 2008-07-25
Filing date: 2009-07-23
Publication date: 2010-01-28
Also published as: JP2010033156A

Abstract

A storing unit (420) stores a template database (422) having a plurality of templates in which first exemplary sentences expressed in a first language are associated with second exemplary sentences expressed in a second language. Each of the first exemplary sentences has a first changeable part in which first candidate words can be replaced, while each of the second exemplary sentences has a second changeable part in which second candidate words, which are obtained by expressing the respective first candidate words in the second language, can be replaced. A keyword selecting unit (442) designates, based on a first instruction, a first keyword from among a plurality of predetermined keywords. An exemplary sentence selecting unit (444) extracts first exemplary sentences in which the first keyword can be replaced. The exemplary sentence selecting unit (444) then displays, on a display unit (462), the first exemplary sentences, in which the first keyword has been replaced, such that the first exemplary sentences become selectable. The exemplary sentence selecting unit (444) then selects, based on a second instruction, one of the first exemplary sentences displayed on the display unit. A translation output unit (446) displays a second exemplary sentence, which is corresponding to the selected first exemplary sentence, on the display unit after putting a second candidate word, which is obtained by expressing the first keyword in the second language, into the second changeable part.

Description

Information processing apparatus and information processing method

The present invention relates to an information processing apparatus and method for translating a sentence, and more particularly to an information processing apparatus and method for storing an example sentence and its translation in association with each other and displaying a translation of a specified example sentence.

Conventionally, by storing example sentence data in which a plurality of example sentences each having a description written in different languages and corresponding to each other are stored, when a user selects an example sentence in a certain language, the selected example sentence is selected based on the example sentence data. There is a translation device that displays example sentences (called translated sentences) in other languages corresponding to example sentences.

If one of such translation devices is provided with a replaceable word (variable part) in an example sentence, and the variable part of the example sentence is changed, the variable part of the translation corresponding to the changed variable part is changed according to the change of the example sentence. There is something to change. Patent Document 1 (Japanese Patent Laid-Open No. 9-251462) discloses a translation apparatus that classifies a word and displays it with a hierarchical structure when displaying the word as a replacement candidate for the variable part.

JP-A-9-251462

In the invention described in Patent Document 1, it is not easy for the user to search for a desired example sentence and obtain a desired translation as described below.

In the invention described in Patent Document 1, in order for a user to search for a desired example sentence, for example, it is necessary to include the following procedure. First, the user searches and selects an example sentence having contents close to a desired example sentence from a predetermined example sentence collection. And the desired example sentence is acquired by changing the variable part of the selected example sentence to another word. Thus, it takes a lot of time to search for a desired example sentence.

Furthermore, even if the word of the variable part of the example sentence selected by the user is changed, a desired example sentence may not be obtained. In this case, the user must further perform a process of selecting another example sentence and replacing the words in the variable part.

The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an apparatus that allows a user to easily search for a desired example sentence and translation.

According to a certain aspect of the present invention, an information processing apparatus is provided. An information processing apparatus includes a template database that includes a plurality of templates that associate an input unit, a display, and a first example sentence described in a first language with a second example sentence that describes the first example sentence in a second language. And a storage unit for storing. The first example sentence has at least one first variable unit that can replace each of the plurality of first candidate words in the first language. The second example sentence has at least one second variable unit capable of replacing each of the second candidate words in which the plurality of first candidate words of the corresponding first example sentence are described in the second language. . The information processing apparatus further includes a processing unit for controlling the operation of the information processing apparatus. The processing unit replaces the first keyword as a first candidate word with a keyword selection unit for designating the first keyword from a plurality of predetermined keywords based on a first instruction from the input unit Extract at least one first example sentence having a possible first variable part, and select a first example sentence in which a first keyword is put in the first variable part of the extracted first example sentence on the display And an example sentence selection unit for selecting one first example sentence from the first example sentence displayed on the display based on the second instruction from the input unit, and a selection based on the template. A second candidate corresponding to the first example sentence in the second language, the second example sentence corresponding to the first example sentence, and the second variable part corresponding to the first variable part containing the first keyword. Put a word on the display And a translation output of the order.

Preferably, the storage unit further stores index data in which a keyword is associated with a template having a first example sentence that can replace a candidate word that matches the keyword. The example sentence selection unit extracts the first example sentence based on the first keyword and the index data.

Preferably, the index data includes a keyword category data in which keywords and categories are associated, a template and a template related to a first example sentence having a first variable unit capable of replacing candidate words that match the keywords corresponding to the categories. And category template data in which is associated with each other. The example sentence selection unit acquires a category corresponding to the first keyword based on the keyword category data, and extracts a first example sentence included in the template corresponding to the acquired category based on the category template data.

Preferably, the input unit accepts input of characters. The keyword selection unit displays a keyword corresponding to the input character on the display so as to be selectable, and designates the first keyword from the displayed keywords based on the first instruction.

Preferably, the storage unit further stores a predicted keyword list that associates the input characters with keywords that partially or wholly match the input characters. A keyword selection part extracts the keyword corresponding to the input character based on a prediction keyword list, and displays the extracted keyword on a display.

Preferably, the keyword selection unit extracts a first example sentence having a first variable unit capable of replacing the first keyword, and the keyword selected as the first variable unit of the extracted first example sentence A first example sentence with a character and a keyword corresponding to the input character are displayed on the display.

Preferably, the keyword selection unit specifies the second keyword from among a plurality of predetermined keywords based on the third instruction from the input unit. The example sentence selecting unit extracts a first example sentence having a first variable part capable of replacing at least one of the first and second keywords, and the first example sentence extracted from the first example sentence is stored in the first variable part of the extracted first example sentence. A first example sentence containing one keyword is displayed on the display in a selectable manner.

Preferably, the keyword selection unit specifies the second keyword from among a plurality of predetermined keywords based on the third instruction from the input unit. The example sentence selecting unit extracts a first example sentence having a first variable part capable of replacing at least one of the first and second keywords, and a first example capable of replacing both the first and second keywords. And the first example sentence that can replace either of the first and second keywords are displayed on the display.

According to another aspect of the present invention, the input unit, the display, the processing unit, and the first language are described in the first language, and each of the plurality of first candidate words in the first language can be replaced. A first example sentence having one variable part, a first example sentence described in the second language, and a plurality of first candidate words of the corresponding first example sentence each described in the second language Information processing method executed by an information processing apparatus including a storage unit that stores a template database having a plurality of templates that associate a second example sentence having at least one second variable unit that can replace each of the candidate words Is provided. In the information processing method, the processing unit accepts the selection of the first keyword from a plurality of predetermined keywords via the input unit, and the processing unit replaces the first keyword as the first candidate word. Extracting at least one first example sentence having a possible first variable part; and a first part in which the processing part puts a first keyword into the first variable part of the extracted first example sentence A step of displaying an example sentence in a selectable manner on the display; and a step of accepting selection of one first example sentence from at least one first example sentence displayed on the display via the input unit; Based on the template, the processing unit outputs the second example sentence corresponding to the selected first example sentence, and the first keyword to the second variable part corresponding to the first variable part containing the first keyword. In the second language After having put the second candidate words and predicates, and displaying on the display.

Preferably, the storage unit further stores index data in which a keyword is associated with a template having a first example sentence that can replace a candidate word that matches the keyword. The step of extracting the first example sentence includes the step of extracting the first example sentence based on the first keyword and the index data.

Preferably, the index data includes a keyword category data in which keywords and categories are associated, a template and a template related to a first example sentence having a first variable unit capable of replacing candidate words that match the keywords corresponding to the categories. And category template data in which is associated with each other. The step of extracting the first example sentence based on the index data includes a step of acquiring a category corresponding to the first keyword based on the keyword category data, and a template corresponding to the acquired category based on the category template data. And extracting a first example sentence included in.

Preferably, the information processing method further includes a step in which the processing unit receives input of a character via the input unit, and a step in which the processing unit displays a keyword corresponding to the input character on the display in a selectable manner. Including.

Preferably, the storage unit further stores a predicted keyword list that associates the input characters with keywords that partially or wholly match the input characters. The step of selectively displaying the keyword on the display includes a step of extracting a keyword corresponding to the input character based on the predicted keyword list and displaying the extracted keyword on the display.

Preferably, the step of displaying the first example sentence on the display in a selectable manner includes the first example sentence in which the first keyword is entered in the first variable part of the extracted first example sentence, and the input characters. Displaying a corresponding keyword on the display.

Preferably, the information processing method further includes a step in which the processing unit receives selection of a second keyword from a plurality of predetermined keywords via the input unit. The step of extracting the first example sentence includes the step of extracting at least one first example sentence having a first variable part capable of replacing at least one of the first and second keywords.

The information processing method further includes a step in which the processing unit receives selection of a second keyword from a plurality of predetermined keywords via the input unit. The step of extracting the first example sentence includes a step of extracting a first example sentence having a first variable part capable of replacing at least one of the first and second keywords. The step of selectively displaying the first example sentence on the display includes a first example sentence that can replace both the first and second keywords and a first example that can replace either the first or second keyword. The example sentence is distinguished and displayed on a display.

According to the present invention, it is possible to search for an example sentence having a variable part that can be replaced with a keyword designated by the user. As a result, the user can easily search for a desired example sentence and translation.

It is a figure which shows the hardware constitutions of the conversation assistance apparatus 100 which concerns on this Embodiment in block diagram format. 1 is a perspective view of a conversation assist device 100. FIG. FIG. 6 is a diagram showing a component arrangement in a second housing 220. It is a figure which shows the functional structure of the conversation assistance apparatus 100 in a block diagram format. It is a figure which shows the specific example of the template 500. FIG. It is a figure which shows the example of the keyword category data 425. It is a figure which shows the example of the category template data 426. It is a figure which shows the specific example of a category list. It is a figure which shows the specific example of the dictionary 423. FIG. It is a 1st figure which shows the specific example of the data contained in the prediction keyword list | wrist 428. FIG. It is a 2nd figure which shows the specific example of the data contained in the prediction keyword list | wrist 428. FIG. It is a figure which shows the specific example of the data contained in registration example sentence information. It is a figure for demonstrating the menu screen 1010. FIG. It is a figure for demonstrating the initial screen 1020. FIG. It is a figure for demonstrating the keyword input screen 1030. FIG. It is a figure for demonstrating the keyword input screen 1030 which displays a long keyword and an example sentence. It is a figure for demonstrating the sentence list display screen 1040. FIG. It is a figure for demonstrating the sentence list display screen 1040 which displays a long example sentence. It is a 1st figure for demonstrating the screen displayed on the display 130 in setting of a 1st additional keyword. It is a 2nd figure for demonstrating the screen displayed on the display 130 in setting of a 1st additional keyword. It is a 3rd figure for demonstrating the screen displayed on the display 130 in the setting of a 1st additional keyword. It is a 4th figure for demonstrating the screen displayed on the display 130 in setting of a 1st additional keyword. It is a 1st figure for demonstrating the screen displayed on the display 130 in setting of a 2nd additional keyword. It is a 2nd figure for demonstrating the screen displayed on the display 130 in setting of a 2nd additional keyword. It is a 3rd figure for demonstrating the screen displayed on the display 130 in setting of a 2nd additional keyword. It is a 4th figure for demonstrating the screen displayed on the display 130 in setting a 2nd additional keyword. It is a figure which shows the conversation sentence one display screen 1510 of an English setting. It is a figure which shows the conversation text one display screen 1520 of audio | voice reproduction | regeneration mode. It is a figure which shows the screen 1530 during audio | voice reproduction | regeneration. It is a figure which shows the conversation text 1 case display screen 1540 of a Chinese setting. It is a figure which shows 1 conversation sentence display screen 1550 of audio | voice reproduction | regeneration mode. It is a figure which shows the word replacement screen 1610 in which the variable part is selected. It is a figure which shows the word replacement screen 1620 in which the variable part is not selected. It is a figure which shows the word replacement screen 1630 in which the variable part is selected. It is a figure which shows the word replacement screen 1640 which changed the word which replaces a variable part to "whale." It is a figure which shows the word replacement screen 1650 after inputting a character "ko". It is a figure which shows the word replacement screen 1660 which changed the word which replaces a variable part into "koala". It is a figure which shows the word replacement screen 1710 displayed when the variable part of a number is selected. It is a figure which shows the word replacement screen 1810 displayed when the variable part of a number is selected. It is a figure which shows an example of the word replacement screen corresponding to the selected variable part. It is a figure for demonstrating how to arrange the word by the translation output part 446 which concerns on a 1st modification. It is a figure for demonstrating how to arrange the word by the translation output part 446 which concerns on a 2nd modification. It is drawing which shows the example by which the word which the top 3 characters which concern on a 3rd modification in common are grouped and displayed. It is a figure which shows the example by which the number of words corresponding to a representative display is displayed on the right of the representative display which concerns on a 3rd modification. It is a figure which shows an example of the word replacement screen which concerns on a 4th modification. It is a figure which shows an example of the word replacement screen which concerns on a 5th modification. It is a figure for demonstrating the state of the screen displayed on a display, the transition of a screen, and the handwriting pad by the function of the word book of a conversation assistance application. It is a figure for demonstrating the state of the screen displayed on a display, the transition of a screen, and the state of a handwriting pad by the function of the log | history example sentence of a conversation assistance application. It is a flowchart which shows the outline | summary of the flow of the process which the conversation assistance apparatus 100 performs. 6 is a flowchart showing a flow of selected keyword determination processing executed as a subroutine of a conversation assist application by the conversation assist device 100. It is a flowchart which shows the flow of the keyword candidate display process performed as a subroutine of a conversation assistance application by the conversation assistance apparatus. It is a flowchart which shows the flow of the example sentence candidate display process performed as a subroutine of a conversation assistance application by the conversation assistance apparatus. It is a flowchart which shows the flow of the log | history example sentence management process performed as a subroutine of a conversation assistance application by a conversation assistance apparatus. It is a flowchart which shows the flow of the registration example sentence management process performed as a subroutine of a conversation assistance application by a conversation assistance apparatus.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

(1. Hardware configuration)
With reference to FIG. 1, the hardware configuration of conversation assistance device 100 according to the present embodiment will be described. FIG. 1 is a block diagram showing a hardware configuration of conversation assistance device 100 according to the present embodiment.

In the present embodiment, the following description is given on the assumption that the conversation assist device 100 is an electronic dictionary having a conversation assist function described in detail below. However, the conversation assist device 100 may be another type of device. For example, the conversation assist device 100 may be a dedicated machine for conversation assistance, a mobile terminal having a conversation assist function (such as a mobile phone or a PDA (Personal digital assistant)), or a personal computer having a conversation assist function. It may be. Further, it may be a network server that presents example sentences / translation results in response to keyword input from a remote place.

The conversation assist device 100 includes an input key 110, a handwriting pad 120, a display 130, a display controller 132, a speaker 140, a voice controller 142, a CPU (Central Processing Unit) 150, and a RAM (Random Access Memory) 160. And a ROM (Read Only Memory) 170 and a hard disk 180.

The input key 110 outputs a signal corresponding to the input key 110 when pressed from the outside. The handwriting pad 120 detects contact of an object (such as a stylus pen) with the surface of the handwriting pad 120 and outputs a signal corresponding to a contact portion between the handwriting pad 120 and the object. The handwriting pad 120 displays a screen corresponding to the operation state of the conversation assist device 100. In addition, the handwriting pad 120 may be configured to perform keyword recognition by performing character recognition from an input stroke with a stylus pen through a handwritten character recognition program (not shown) or the like. Further, a configuration may be adopted in which speech recognition is performed and a keyword is input through a speech input unit (such as a microphone) and a speech recognition program (not shown).

The display 130 displays a screen based on the data inside the conversation assist device 100. The display controller 132 controls the operation of the display 130. As the display 130, for example, a liquid crystal display, an LED (Light-Emitting Diode) display panel, or the like can be used.

Speaker 140 outputs sound based on the data inside conversation assist device 100. The audio controller 142 controls the operation of the speaker 140.

CPU 150 executes a calculation for controlling the operation of conversation assist device 100. The RAM 160 stores temporary information generated when the CPU 150 performs processing. The ROM 170 stores information such as a conversation assist program 172 executed by the CPU 150. Information stored in the ROM 170 will be described in detail later.

FIG. 2 is a perspective view of the conversation assist device 100. However, FIG. 2 shows an example of the appearance of the conversation assist device 100, and the appearance of the conversation assist device 100 is not limited to that shown in FIG.

Referring to FIG. 2, conversation assistance device 100 includes a first casing 210, a second casing 220, and a stylus pen 230. A display 130 is disposed in the first housing 210. In the second housing 220, a handwriting pad 120, a speaker 140, and an input key 110 are arranged. The first casing 210 and the second casing 220 are connected so as to be openable and closable.

With reference to FIG. 3, the arrangement of components in the second casing 220 will be described. FIG. 3 is a diagram showing a component arrangement in the second housing 220. However, the arrangement of the keys and the handwriting pad 120 of the conversation assist device 100 shown in FIG. 3 is an example, and the arrangement of the keys and the handwriting pad 120 of the conversation assistance device 100 is limited to that shown in FIG. Do not mean.

The second casing 220 is provided with a handwriting pad 120, a speaker 140, and various keys. The keys arranged on the second casing 220 include a character key 302, an upper key 304a, a lower key 304b, a right key 304c, a left key 304d (hereinafter, an upper key 304a, a lower key 304b, a right key 304c, and a left key). 304d is collectively referred to as a direction key 304), a direction key 304, a search / decision key 306, a return key 308, a clear key 310, a backward key 312, a conversation assist key 314, a switching key 316, and a voice key 318. , Menu key 320, previous page key 322, and next page key 324.

The following is a brief description of some keys. The character key 302 is used for inputting characters. The direction key 304 is used for moving the cursor in the screen displayed on the display 130. When the conversation assist key 314 is pressed, the conversation assist key 314 outputs a signal for starting a conversation assist application. When pressed, the menu key 320 outputs a signal for displaying a menu screen on the display 130.

(2. Functional configuration)
With reference to FIG. 4, the functional structure of the conversation assistance apparatus 100 is demonstrated. FIG. 4 is a block diagram showing the functional configuration of the conversation assist device 100. As shown in FIG.

The conversation assist device 100 includes an input unit 410, a storage unit 420, a processing unit 440, and an output unit 460.

The input unit 410 receives an instruction from the outside. In the conversation assist device 100 shown in FIG. 1, the input key 110 and the handwriting pad 120 function as the input unit 410. However, other input devices such as a mouse may be used for the input unit 410 depending on the type of the conversation assist device 100.

The output unit 460 outputs information inside the conversation assist device 100. The output unit 460 includes a display unit 462 that outputs an image and an audio output unit 464 that outputs audio. In the conversation assist device 100 illustrated in FIG. 1, the display 130 functions as the display unit 462. Further, in the conversation assist device 100 illustrated in FIG. 1, the speaker 140 functions as the audio output unit 464. Note that a device other than the speaker 140, for example, an earphone may be used as the audio output unit 464.

The storage unit 420 stores various types of information. In the conversation assist device 100 shown in FIG. 1, the RAM 160, the ROM 170, and the hard disk 180 function as the storage unit 420. However, other storage media such as a flash memory may be used as the storage unit 420.

In the present embodiment, storage unit 420 stores conversation assist program 172, template database 422, dictionary 423, index data 424, predicted keyword list 428, registered example sentence information 430, and historical example sentence information 432. To do.

Hereinafter, the structure of various data stored in the storage unit 420 will be described in detail.
<Template database>
The template database 422 includes a plurality of templates 500 for associating a plurality of example sentences respectively described in a plurality of different languages. The template database 422 is used when the conversation assist device 100 creates a translation of the example sentence selected by the user.

The template 500 will be specifically described with reference to FIG. FIG. 5 shows a specific example template 500 a of the template 500. Hereinafter, the template 500 will be specifically described based on the template 500a.

In the category sentence (Japanese) 530, characters other than those enclosed in {} are Japanese. In the category sentence (English) 550, characters other than those enclosed in {} are English. In the category sentence (Chinese) 570, characters other than those enclosed in {} are Chinese. In the representative word sentence (Japanese) 520, characters other than {} symbols, symbols in {}, and numbers in {} are Japanese. In the representative word sentence (English) 540, characters other than the {} symbol, the symbol in {}, and the number in {} are English. In the representative word sentence (Chinese) 560, characters other than the numbers in {} symbols, {}: symbols, and {} are in Chinese.

In addition, here, although the template for three languages for translating from Japanese to English and from Japanese to Chinese is shown, the combination of languages is not limited to this. In addition, it is clear that the following contents can be applied to a bilingual template or a template of four or more languages.

The template 500 includes a template number 510 for specifying the template 500, a representative word sentence (Japanese) 520, a category sentence (Japanese) 530, a representative word sentence (English) 540, and a category sentence (English) 550. A representative word sentence (Chinese) 560, a category sentence (Chinese) 570, and a keyword 580.

The category sentence (Japanese) 530, the category sentence (English) 550, and the category sentence (Chinese) 570 have the same meanings although the languages are different. Categories are set for some words. The category sentence (English) 550 will be described as an example. The category sentence (English) 550 includes four parts ({1: QPRON1}, {2: GOODS1 + pl}, {3) that specify categories of words to be included in the sentence. : VB_BUY1 + inf}, {4: PLACE1}) and a part represented by a fixed word ("can't I", "in"). Hereinafter, a part for specifying a category is referred to as “variable part”, and a part other than the variable part is referred to as “fixed part”. Each variable part is specified by the first number in {}. In the following, when differentiating the variable parts, the variable part whose number in {} is n is referred to as the nth variable part.

The structure of the variable part will be described. Capital letters in {} (such as QPRON1) indicate category names of words that should be included in the variable part. In addition, depending on the sentence structure, information specifying the shape of a word entering the variable part may be included in the variable part. For example, + inf of {3: VB_BUY1 + inf} indicates that a verb infinitive infinitive is placed in this variable part. As another example, + pl of {1: GOODS1 + pl} in the category sentence (English) 550 indicates that a plural form of a noun is included in the variable part.

The representative word sentence (Japanese) 520, the representative word sentence (English) 540, and the representative word sentence (Chinese) 560 are a category sentence (Japanese) 530, a category sentence (English) 550, and a category sentence (Chinese), respectively. In each variable part 570, a preferred word is put. The representative word sentence (English) 550 will be described as an example. In the representative word sentence (English) 550, the first variable part of the category sentence (English) 550 is set to “Which”, the second variable part is set to “bags”, and the third variable is set. The part is replaced with buy and the fourth variable part is replaced with Japan.

The representative word sentence (Japanese) 520, the representative word sentence (English) 540, the representative word sentence (Chinese) 560, the category sentence (Japanese) 530, the category sentence (English) 550, the category sentence (Chinese) ) 570 may be formed of only a fixed portion. In that case, the representative word sentence and the category sentence may hold the same data, or one of them may be empty data.

<Dictionary>
Returning to FIG. 4, the dictionary 423 is data that associates words of a plurality of languages with each other. Specific examples of the dictionary will be described later.

<Index data>
The index data 424 is data for associating user input with example sentences, and is used for example sentence search by user input. In the present embodiment, index data 424 includes keyword / category data 425 that associates a user input with a keyword category corresponding to the user input, and category / template data 426 that associates a category with a template.

The index data 424 will be described with reference to FIGS. 6A and 6B. 6A and 6B are diagrams for explaining the index data 424. FIG. In FIG. 6A, the character corresponding to Japanese input and the character corresponding to Japanese notation are Japanese.

(Keyword category data)
FIG. 6A is a diagram illustrating an example of the keyword / category data 425. FIG. 6A shows three keyword / category data 425 (keyword / category data 425a, keyword / category data 425b, and keyword / category data 425c).

Referring to FIG. 6A, each keyword / category data 425 correlates a keyword Japanese input 612, a word number 613, a keyword Japanese notation 614, and a keyword category number 616. For example, the Japanese input 612 of the keyword / category data 425a is “Japan”, the word number 613 is “2312”, the Japanese notation 614 is “Japan”, and the category number 616 is “C3001”.

Japanese input 612 is Japanese corresponding to the keyword. The Japanese notation 614 is a display on the display 130 of keywords corresponding to the Japanese input 612. That is, when the input key 110, the handwriting pad 120, or the voice input microphone (not shown) receives the input of the character string described in the Japanese input 612, the CPU 150 and the display controller 132 display the Japanese notation 614 on the display unit 462. Displays a character string as a keyword. Explaining in accordance with the keyword / category data 425a, when Japanese input “Nippon” is made, the conversation assisting apparatus 100 recognizes that the keyword “Japanese” written in Japanese is input. It is also possible to accept input of Japanese notation 614 directly with the handwriting pad 120 or the like.

The word number 613 is the ID of the corresponding word in the dictionary.
The category number 616 is data unique to each category representing the category to which the keyword belongs. A keyword belonging to a plurality of categories (in the example shown in FIG. 6A, keyword / category data 425c) has a plurality of category numbers 616.

The keyword / category data 425 related to the Japanese keyword has been described above assuming that the user searches for an example sentence using the Japanese keyword. However, there may be keyword / category data 425 regarding other languages. With such keyword / category data 425, the user can search for example sentences from other languages.

(Category / Template data)
FIG. 6B is a diagram showing an example of category template data 426. FIG. 6B shows two category template data 426 (category template data 426a and category template data 426b).

Referring to FIG. 6B, category / template data 426 associates category number 616 with template number 510. That is, according to each category / template data 426, the template 500 related to a certain category number 616 can be specified. For example, according to the category / template data 426a, it is understood that the template 500 having the template numbers 10100, 10151, and 10200 is associated with the category number C0010.

(Search by index data)
The conversation assist device 100 searches for an example sentence corresponding to the character string input by the user by using the index data 424 as described above. That is, the conversation assist device 100 acquires the keyword category number 616 corresponding to the character string input by the user based on the keyword / category data 425. Next, the conversation assistance device 100 acquires a template number 510 corresponding to the acquired category number 616 based on the category / template data 426. Finally, the conversation assist device 100 searches for and acquires an example sentence corresponding to the acquired template number 510 based on the template database 422. The example sentence search process corresponding to the keyword will be described in detail later.

The conversation assist device 100 according to the present embodiment can search not only the template 500 having the input character string as a keyword, but also the template 500 having a variable part for designating a category including the input character string. Therefore, a large number of example sentences related to the input character string can be searched. For example, if the number of example sentences is 1000, the average of the variable parts included in the example sentence is 3, and the average number of candidates per variable part is 100, 1000 × 100 × 100 × 100 = 1,000, 000,000 example sentences can be searched.

Further, since the conversation assist device 100 uses the index data 424, the above-described example sentence search can be executed in a shorter time than when the index data 424 is not prepared.

That is, when the index data 424 is not prepared, it is necessary to perform a character string search for all example sentences in order to search for example sentences from the input keywords. In this character string search, since pattern matching between character strings is performed, the processing time increases in proportion to the amount of example sentences. On the other hand, the conversation assist device 100 according to the present embodiment can quickly acquire a keyword category based on the keyword / category data 425 included in the index data 424. Further, the conversation assist device 100 can search a template corresponding to the category at high speed based on the category / template data 426.

(About creating index data)
The index data 424 according to the present embodiment is created based on the template 500, the dictionary 423, and the category list 700 that associates category names and category numbers.

A specific example of the category list 700 will be described with reference to FIG. FIG. 7 is a diagram illustrating a specific example of the category list. Referring to FIG. 7, the first entry 700a of category list 700 indicates that the category number of the category whose category name is “PLACE1” is “C3001”. The second entry 700b of the category list 700 indicates that the category number of the category whose category name is “GOODS1” is “C0010”.

A specific example of the dictionary 423 will be described with reference to FIG. FIG. 8 is a diagram illustrating a specific example of the dictionary 423. In FIG. 8, the character corresponding to Japanese input and the character corresponding to Japanese notation are Japanese. The character corresponding to the Chinese notation is Chinese.

The first entry 800a of the dictionary indicates that the Japanese notation “bag”, the Japanese input “bag”, the English notation “bag”, and the Chinese notation “pack” correspond to each other. Yes. The first entry 800a indicates that the word numbers of “bag”, “bag”, “bag”, and “pack” are “5432” and the category numbers are “C0010” and “C0011”.

For one word included in the dictionary 423, the Japanese input 612 and the Japanese notation 614 of the keyword / category data 425 are created from the Japanese notation and the Japanese input of the entry of the dictionary 423. The word number 613 and category number 616 of the keyword / category data 425 are created from the word number and category number of the entry in the dictionary 423, respectively.

Also, in order to register other than variable part candidates as keywords, it is possible to create Japanese input and Japanese notation of the keyword / category data 425 from Japanese input and Japanese notation of the keyword 580 of the template 500.

In this embodiment, it is assumed that one or a plurality of Japanese inputs are associated with one Japanese notation and registered. For example, since the Japanese input “Japan” and “Japan” are associated with the Japanese notation “Japan”, the user can enter the category number “C3001” regardless of whether the user inputs “Japan” or “Japan”. Via this, the template number “10100” can be searched.

Further, the category / template data 426 is created from the category name of the variable part included in the template 500 and the category list 700.

As described above, the index data 424 created based on the template 500, the category list 700, and the dictionary 423 appropriately reflects the correspondence between the input words and the template 500.

<Predicted keyword list>
In the predicted keyword list 428, a plurality of predetermined keywords are registered by being classified according to the first one or more characters of each keyword. The predicted keyword list 428 is used to assist the keyword input by the user.

Specific examples of the predicted keyword list 428 will be described with reference to FIGS. 9 and 10 are diagrams illustrating specific examples of data included in the predicted keyword list 428, respectively. In FIG. 9 and FIG. 10, the characters listed below the reading (input characters 910) are Japanese hiragana characters. That is, the characters listed on the left are Japanese hiragana characters. The characters listed below the candidate 922 are Japanese kanji, katakana, hiragana, or their collocations. That is, the characters listed on the right are Japanese kanji, katakana, hiragana, or their collocations.

Referring to FIG. 9 and FIG. 10, the predicted keyword list 428 stores a search result 920 of predicted words that match the input character 910 with a forward match and a complete match, respectively. Here, “front match” means that all character strings of the input character 910 are included when referred from the beginning of the predicted word character string. “Completely matched” means that the predicted word character string and the input character 910 completely match.

Also, the predicted words included in the search result 920 are not limited to those that match the input character 910 in a forward match and a complete match. A case where the input character 910 is included in the middle of the predicted word character string is referred to as “partial match”. The predicted keyword list 428 may store a search result 920 of predicted words in any or all of these forward match, partial match, and complete match.

Here, FIG. 9 shows the case of forward matching, and FIG. 10 shows the case of perfect matching.
Referring to FIG. 9, the search result 920 includes predicted word candidates 922, candidate numbers 923, and word numbers 924. The word number 924 is a number that identifies the candidate 922 and matches the word number shown in FIG. In the example shown in FIG. 9, the total number of candidates that coincide with the reading “A” is 964, and the value of the word number 924 that specifies “ART” is 2415.

Similarly, in the example shown in FIG. 10, the total number of candidates that exactly match the reading “Artichoke” is one, and the value of the word number 924 that identifies “Artichoke” is 711. .

<History example sentence information and registered example sentence information>
The registered example sentence information 430 is information about example sentences registered by the user among example sentences determined as output targets. The history example sentence information 432 is information about the history of example sentences determined as output targets.

The registered example sentence information 430 will be described with reference to FIG. FIG. 11 is a diagram illustrating specific examples of data included in the registered example sentence information 430.

Referring to FIG. 11, registered example sentence information 430 includes keywords specified from input characters, example sentence search results, translation results, and translation dictionary data in association with each other. The translation dictionary data (corresponding to the template 500) includes a sentence number (corresponding to the template number 510), the number of variable parts, the number of words for each variable part, a word number and a character string indicating a word for each variable part, Are included in association with each other.

The input keyword is a word input by the user or a word predicted from characters input by the user in order to search for example sentences.

The example sentence search result is an example sentence as a result determined by the user among at least one example sentence searched based on the input keyword. The example sentence search result may include a portion sandwiched between a full-width black square and a half-width underbar, and a portion sandwiched between a full-width square and a half-width underbar. These parts are the variable parts described above. The part sandwiched between the full-width black square and the half-width underbar is the variable part that is in focus, and the part sandwiched between the full-width square and the half-width underbar indicates the variable part that is not in focus. Yes. The description of the example sentence search result is not limited to this example, and a method of changing the typeface of the word in the variable part (adding an underline, changing the color of the character, making it bold, etc.) may be used.

Translation results are translated sentences based on example sentence search results. The sentence number is a number assigned in advance to each example sentence. The number of variable parts is the number of variable parts included in the example sentence. The word for each variable part is at least one word that can be replaced with the variable part. The number of words for each variable part is the number of words registered as words that can be replaced in the variable part.

For example, as registered example sentence information 430 corresponding to the input keyword “Asian”, the example sentence search result “This tour is very popular with Asian customers”, “This tour is very popular with our Asian customers.”, And Data of the sentence number “5826” is included.

Also, as the data of the sentence number “5826”, the number of variable parts “3”, the number of words in the variable part 1 “2”, and the character string “(11: here / this / 12: there / That /) ”, the number of words in the variable part 2“ 2 ”, the character string“ (2331: tour / tour / 2345: cruise / cruise /) ”indicating the word in the variable part 2, and the word in the variable part 3 The number “19”, the character string indicating the word of variable part 3 ((301: Asian / Asian / 302: African / African / 303: Middle East / Middle Eastern / 304: East Asian / East Asian / 305: West Asian / West Asian / 306: South American / South American / 307: North American / North American / 308: Oceanian / 309: Europe / European / 310: Persian / Persian / 311: Western / Western / 312: Japanese / Japanese / 313: American / American / 314: German / German / 315: British / British / 316: French / French / 317: Italian / Italian / 318: Korea Roh / South Korean / 319: China / Chinese /) "are included. In the character string indicating the words of the variable parts 1 to 3, the word number and the character string (Japanese + English) are included.

Also, depending on the example sentence, there are cases where three variable parts are included, and there are cases where no variable part is included. In particular, a case where no variable part is included is referred to as an example sentence having only a fixed part. Even for example sentences with only a fixed part, it is possible to search for predicted keywords, example sentences, and obtain translation results from input characters.

Note that the data included in the history example sentence information 432 is the same as the data included in the registered example sentence information 430 described with reference to FIG.

As the registered example sentence information 430 and the history example sentence information 432, a part or all of the data structure recorded in the translation dictionary data is stored. For this reason, since the example sentences registered by the user among the example sentences determined as output targets and the history of the example sentences determined as output targets can be reused as they are, not only the trouble of searching for the same example sentence again can be saved. As will be described later, it is possible to acquire various example sentences and translated sentences by changing the variable part of those example sentences.

As described above, the storage unit 420 stores various data. Next, the processing unit 440 will be described.

Returning to FIG. 4, the processing unit 440 includes a keyword selection unit 442, an example sentence selection unit 444, a translated sentence output unit 446, a registered example sentence management unit 448, and a history example sentence management unit 449.

The keyword selection unit 442 selects a keyword based on the instruction received by the input unit 410. The keyword selection unit 442 includes a prediction conversion unit 443 that extracts words predicted from the characters received by the input unit 410 as keyword candidates based on the prediction keyword list 428.

The keyword selection unit 442 displays the keyword candidates extracted by the prediction conversion unit 443 on the display unit 462 so as to be selectable as keyword candidates. More specifically, the keyword selection unit 442 searches the keyword having the Japanese input 612 that matches the input character string with reference to the index data 424. Then, the predictive conversion unit 443 displays the searched keyword notation 614 on the display unit 462. The keyword selection unit 442 selects one keyword candidate from the keyword candidates displayed on the display unit 462 based on the instruction received by the input unit 410.

Specifically, the keyword selection unit 442 displays keyword candidates as shown in FIG. 10 on the display unit 462. FIG. 10 is a diagram showing a predicted keyword search result that completely matches the input character string as described above.

The example sentence selection unit 444 selects an example sentence based on the keyword selected by the keyword selection unit 442, the index data 424, and the instruction received by the input unit 410. Specifically, the example sentence selection unit 444 searches the index data 424 and extracts the template 500 related to the selected keyword. Details of the search have already been described in the description of the index data 424. Further, the example sentence selection unit 444 displays an example sentence on the display unit 462 so as to be selectable based on the searched template 500. Further, the example sentence selecting unit 444 selects one example sentence from the example sentences displayed on the display unit 462 based on the instruction received by the input unit 410.

The translated sentence output unit 446 displays the example sentence selected by the example sentence selecting unit 444 and the translated sentence of the example sentence on the display unit 462. In particular, the translation output unit 446 displays the variable part of the example sentence so as to be selectable, and replaces the words of the variable part based on the instruction received by the input unit 410. Details of this replacement will be described later. In addition, the translation output unit 446 causes the voice output unit 464 to output the translation based on the instruction received by the input unit 410.

The registered example sentence management unit 448 manages an example sentence for which the user has performed an operation of registering among example sentences selected by the example sentence selection unit 444. More specifically, the registered example sentence management unit 448 has input keywords, example sentence search results, translation results, translation dictionary data (sentence number, number of variable parts) related to example sentences selected by the example sentence selection unit 444 and registered by the user. The number of words for each variable part, the word number indicating the word of the variable part, and the character string) are associated with each other and stored in the storage unit 420 as the registered example sentence information 430 described in FIG. For example sentences with no variable part (only the fixed part), the translation dictionary data has a variable part number and the number of words per variable part is "0", and the word number and character string indicating the variable part word are empty (NULL ) Is stored in the storage unit 420.

Also, the registered example sentence management unit 448 can read the registered example sentence stored in the storage unit 420 as the registered example sentence information 430 based on the instruction received by the input unit 410, and can select the registered example sentence from the list of registered example sentences read out. As shown in FIG. Furthermore, the registered example sentence management unit 448 selects one registered example sentence from the registered example sentences displayed on the display unit 462 based on the instruction received by the input unit 410.

The translated sentence output unit 446 displays the registered example sentence selected by the registered example sentence management unit 448 and the translated sentence of the registered example sentence on the display unit 462. The display of registered example sentences is the same as the display of example sentences selected by the example sentence selection unit 444 described above. Note that the example sentence selected by the example sentence selecting unit 444 is displayed based on the template 500, but the registered example sentence is displayed based on data stored as the registered example sentence information 430.

History example sentence management unit 449 manages the history of example sentences selected by the example sentence selection unit 444. Specifically, the history example sentence management unit 449 relates to an example sentence selected by the example sentence selection unit 444, input keywords, example sentence search results, translation results, translation dictionary data (sentence number, number of variable parts, number of words for each variable part) , A word number indicating a word of the variable part and a character string) are associated with each other and stored in the storage unit 420 as history example sentence information 432 similar to the registered example sentence information 430 described in FIG.

Further, the history example sentence management unit 449 can read the history example sentence stored in the storage unit 420 as the history example sentence information 432 based on the instruction received by the input unit 410, and can select the history example sentence from the list of read history example sentences As shown in FIG. Further, the history example sentence management unit 449 selects one history example sentence from the history example sentences displayed on the display unit 462 based on the instruction received by the input unit 410.

The translation output unit 446 displays the history example sentence selected by the history example sentence management unit 449 and the translation of the history example sentence on the display unit 462. The display of the history example sentence is the same as the display of the example sentence selected by the example sentence selection unit 444 described above. The history example sentence is displayed based on the data stored as the history example sentence information 432.

(3. Operation of conversation assist device)
The operations performed by the conversation assist device 100 are roughly divided as follows:
(1) Activation of conversation assist application (2) Acceptance of keyword (3) Determination of example sentence (4) Addition of keyword (5) Output of one conversation sentence (6) Word replacement (7) Selection of registered example sentence (8) Selection of History Example sentence Hereinafter, each of (1) to (8) will be described in order.

(1) Activation of Conversation Assist Application In the present embodiment, conversation assist device 100 is selected when “conversation assist” is selected on menu screen 1010 shown in FIG. 12 or when conversation assist key 314 is pressed. , Launch a conversation assist application. However, the menu screen illustrated in FIG. 12 is an example, and the menu screen displayed on the display 130 by the conversation assist device 100 is not limited to the one described below.

Referring to FIG. 12, the menu screen 1010 displays a content name (such as a dictionary name) or an application name (such as conversation assist, which is preceded by a note mark). A text for assisting the user's operation is displayed at the bottom of the menu screen 1010.

The content name and application name displayed on the menu screen 1010 can be selected. In response to pressing of the up key 304a (or down key 304b), the processing unit 440 newly selects content that is one level above (or one level below) the currently selected content name.

The selected content name or application name is displayed in a form that shows that it is selected, such as changing the color or enclosing it in a frame. On the menu screen 1010, the conversation assist is selected, and the conversation assist is surrounded by a frame.

When the search / decision key 306 is pressed, the processing unit 440 activates an application corresponding to the selected content name or application name.

The processing unit 440 activates the conversation assist application based on the conversation assist application activation instruction, and displays the initial screen 1020 shown in FIG. That is, when the processing unit 440 detects that the search / decision key 306 is pressed or “conversation assist key 314” is pressed when “conversation assist” is selected on the menu screen 1010, the initial screen is displayed. 1020 is displayed on the display 130. When the return key 308 is pressed while the initial screen 1020 is displayed, the processing unit 440 displays the menu screen 1010 on the display 130.

Referring to FIG. 13, initial screen 1020 includes a title 1022, a battery mark 1023, a hint 1024, a character input field 1026, and a language mode setting 1028.

Title 1022 represents the name of the currently activated application, that is, “conversation assist”, and the language mode. Here, the “language mode” is information for designating which language is translated into which language in conversation assistance.

In the present embodiment, two language modes (a mode for translating Japanese into English and a mode for translating Japanese into Chinese) can be selected, and characters representing the selected language mode are displayed in the title 1022. Is displayed. When the mode for translating Japanese into English is selected, “Japanese → English” is displayed in the title 1022. On the other hand, when the mode for translating Japanese into Chinese is selected, “Japanese → Chinese” is displayed in the title 1022.

The battery mark 1023 represents the remaining battery level for driving the conversation assist device 100. The display form of the battery mark 1023 changes according to the remaining battery level.

The hint 1024 displays a sentence explaining an operation that the user should or can perform.

The character input field 1026 displays characters in response to pressing of the character key 302 or input to the handwriting pad 120. As described above, the character input may be a keyword input by performing character recognition from an input stroke by a stylus pen through a handwritten character recognition program (not shown) or the like. Further, a configuration may be adopted in which speech recognition is performed and a keyword is input through a speech input unit (such as a microphone) and a speech recognition program (not shown).

The language mode setting 1028 represents the selected language mode. Specifically, a square color provided to the left of the characters representing each language mode indicates which language mode is selected. An initial screen 1020 shown in FIG. 13 is displayed when a mode for translating Japanese into English is selected, and a square next to “Japanese → English” is displayed in black.

The user can select the language mode by pressing the switch key 316 while the initial screen 1020 is displayed on the display 130. That is, the processing unit 440 switches the language mode when detecting the pressing of the switching key 316 during the display of the initial screen 1020.

It is assumed that the default language mode, that is, the language mode from when the conversation assist application is first started up until the switch key 316 is pressed, is a mode for translating Japanese into English. The processing unit 440 stores the language mode at the end of the conversation assist in the storage unit 420, and selects the language mode at the end of the previous conversation assist when the conversation assist application is activated for the second time or later. .

The types of language modes are not limited to the two described above. For example, there may be a mode for translating a language other than Japanese into another language. The language mode that can be handled by the conversation assist device 100 is set at the time of manufacture. The manufacturer of the conversation assist device 100 may set the language mode handled by the conversation assist device 100 as appropriate (for example, according to the sales area of the conversation assist device 100).

In addition, the number of language modes that the conversation assist device 100 can handle is not limited to two. The conversation assist device 100 may handle three or more language modes. Conversely, the conversation assist device 100 may handle only one language mode. In this case, it is unnecessary to display the language mode setting 1028 on the initial screen 1020.

(2) Keyword Acceptance When a character is input when the initial screen 1020 is displayed, the processing unit 440 displays a keyword input screen 1030 shown in FIG. When the return key 308 or the clear key 310 is pressed while the keyword input screen 1030 is displayed, the processing unit 440 displays the initial screen 1020 on the display 130.

FIG. 14 shows a screen for translating Japanese (first language) into English (second language). In FIG. 14, the keyword input in the character input field 1026, the text displayed in the candidate keyword list 1032, and the text displayed in the example sentence preview 1034 are in Japanese (first language).

Referring to FIG. 14, the keyword input screen 1030 includes a title 1022, a battery mark 1023, a hint 1024, and a character input field 1026, similar to the initial screen 1020. A keyword input screen 1030 shown in FIG. 14 is displayed when the character string “pen” is input. Therefore, “pen” is displayed in the character input field 1026.

Note that various screens described below also include a title 1022, a battery mark 1023, a hint 1024, and a character input field 1026. However, since the roles of these displays are the same as those of the initial screen 1020, these displays are not basically referred to in the following description of the screens.

The keyword input screen 1030 includes a candidate keyword list 1032 and an example sentence preview 1034 in addition to the above display.

The candidate keyword list 1032 displays keywords corresponding to the characters input in the character input field 1026. Here, “pen”, “penguin”, “pen growth saddle dome”, “pensylvania”, and “pence”, which are keywords corresponding to the candidate keyword character string “pen”, are displayed in the candidate keyword list 1032.

The keywords included in the candidate keyword list 1032 are those searched by the prediction conversion unit 443 included in the keyword selection unit 442 based on the prediction keyword list 428. Here, the predictive conversion unit 443 searches for a keyword whose input character string matches the first character string based on the candidate keyword list 1032.

The user can select each keyword included in the candidate keyword list 1032 using the up key 304a, the down key 304b, the previous page key 322, the next page key 324, and the like. In addition, when the predicted word cannot be displayed on the display 130, the keyword selection unit 442 scrolls and displays the predicted word included in the candidate keyword list 1032 in response to pressing of various keys.

In this embodiment, the number of predicted words included in the candidate keyword list 1032 is not particularly limited. However, the number of predicted words included in the candidate keyword list 1032 may be limited depending on the performance of the conversation assist device 100.

The currently selected keyword is displayed on the display 130 so that it can be seen that it has been selected. The keyword input screen 1030 is displayed when the keyword “penguin” is selected, and lines including “penguin” are shaded. Here, for the display of the currently selected (focused) keyword, it is possible to change the typeface (add an underline, make it bold, change the color of the character), or add a symbol (mark). The same applies to an example sentence list 1044 described later.

Note that if the predictive conversion unit 443 cannot search for a keyword corresponding to the input character, in the present embodiment, the keyword selection unit 442 indicates that there is no keyword corresponding to the input character on the display 130. A message (for example, “no candidate”) is displayed. It is assumed that the input unit 410 does not accept keyword selection and determination instructions while this message is displayed. That is, during this message, even if the user presses the up key 304a, the down key 304b, and the search / decision key 306, the screen displayed on the display 130 does not change. As another form, the candidate keyword list may be blank, indicating that the keyword could not be searched.

Note that the predictive conversion unit 443 can search for a word that does not completely match the input character string but has a similar notation. For example, the prediction conversion unit 443 searches for a predicted word without distinguishing between katakana and hiragana. That is, even when “pen” is input in hiragana, the predictive conversion unit 443 extracts the word “penguin” (in katakana notation) as a predicted word, similarly to the case where “pen” is input in katakana. Other examples and specific description of how to associate the input character string with the keyword will be described later.

The example sentence preview 1034 displays an example sentence including the selected keyword. The keyword selection unit 442 searches the template database 422 for example sentences relating to the selected keyword based on the index data 424. The specific process performed when the keyword selection unit 442 determines an example sentence to be displayed in the example sentence preview 1034 is the same as the process of the example sentence selection unit 444 when creating a sentence list display screen described later. Therefore, detailed description of this process is omitted here. The example sentence preview 1034 makes it easy for the user to select a desired example sentence. The user can select a keyword by looking at related example sentences, so that it is less likely to select an undesired keyword. That is, it is less likely that there are no example sentences expected by the user in the example sentences related to the keyword selected by the user.

Note that if the number of characters of the keyword exceeds the number of characters that can be displayed in one line of the screen, the keyword selection unit 442 displays the first character of the keyword in the candidate keyword list 1032. When the number of characters in the example sentence exceeds the number of characters that can be displayed in one line on the screen, the keyword selection unit 442 displays the first character of the example sentence in the candidate keyword list 1032.

An example of such a screen is shown in FIG. FIG. 15 is a diagram showing a screen displayed by the keyword selection unit 442 on the display unit 462 when the prediction conversion unit 443 “Kelly Tarlton's Antitactic Encounter & Underworld” is extracted as a prediction word. is there. FIG. 15 also shows a screen for translating Japanese (first language) into English (second language). In FIG. 15, the keyword input in the character input field 1026, the text displayed in the candidate keyword list 1032, and the text displayed in the example sentence preview 1034 are in Japanese (first language).

Referring to FIG. 15, candidate keyword list 1032 includes characters from the beginning of the predicted word to (number of characters that can be displayed on display unit 462-1), that is, “Kelly Tarlton's Antitactic Encounter & "Is displayed, followed by" ... ". The keyword selection unit 442 may display characters from the beginning of the predicted word (the number of characters that can be displayed on the display unit 462) without displaying “...”. The example sentences displayed in the example sentence preview 1034 are also omitted in the same manner as the predicted words. This example corresponds to a case where there is a restriction on the number of characters that can be displayed on the display panel. However, the size of the character font may be reduced, or folded display may be performed.

(3) Example sentence determination When one of the keywords displayed in the candidate keyword list 1032 is selected at the time of displaying the keyword input screen 1030, the processing unit 440 displays a figure to assist the user in determining an example sentence. A sentence list display screen 1040 shown in FIG. FIG. 16 shows a screen for translating Japanese (first language) into English (second language). In FIG. 16, the keyword displayed in the keyword display field 1042, the text displayed in the example sentence list 1044, and the first text in the translation preview 1046 are in Japanese (first language). The second sentence in the translation preview 1046 is English (second language).

Referring to FIG. 16, the sentence list display screen 1040 includes a keyword display column 1042, an example sentence list 1044, and a translated sentence preview 1046.

The keyword display field 1042 displays the keyword selected on the keyword input screen 1030. As will be described later, the keyword selection unit 442 can set an additional keyword. The keyword display field 1042 displays a set additional keyword in addition to the keyword selected on the keyword input screen 1030. Such a display example will be described later. It is assumed that conversation assist device 100 according to the present embodiment can accept up to three keywords. However, the maximum number of keywords is not limited to three.

The example sentence list 1044 displays the example sentences searched by the example sentence selection unit 444. The example sentences displayed in the example sentence list 1044 are related to the keywords displayed in the keyword display field 1042, and are extracted by the example sentence selection unit 444. Here, the example sentence related to the keyword refers to an example sentence having a variable part capable of replacing the keyword.

In this embodiment, it is assumed that the example sentence selecting unit 444 displays up to 20 example sentences searched for in the example sentence list 1044. Since the number of example sentences to be displayed is limited, the user can easily confirm the example sentences. However, the upper limit number of example sentences displayed in the example sentence list 1044 is not limited to 20. Further, the example sentence selection unit 444 may not limit the number of example sentences to be displayed.

Returning to FIG. 4, the example sentence extraction process by the example sentence selection unit 444 will be described in detail. In the present embodiment, the example sentence selection unit 444 extracts an example sentence related to the keyword based on the index data 424. More specifically, the example sentence selection unit 444 extracts an example sentence related to the keyword as follows.

First, the example sentence selection unit 444 refers to the keyword / category data 425 and acquires the category number 616 of the selected keyword. That is, the example sentence selection unit 444 extracts an entry including the notation 614 that matches the selected keyword, and acquires the category number 616 of the extracted entry as the category number 616 of the selected keyword.

Next, the example sentence selecting unit 444 acquires the template number 510 corresponding to the acquired category number 616 based on the category / template data 426. When there are a plurality of template numbers 510 corresponding to the category number 616, the example sentence selecting unit 444 acquires all the template numbers 510 corresponding to the category number.

Finally, the example sentence selection unit 444 searches for and extracts a template having the template number 510 acquired from the template database 422. When a plurality of template numbers 510 are acquired, the example sentence selecting unit 444 searches for a template corresponding to each acquired template number 510. Then, the example sentence selection unit 444 replaces the variable part that can be replaced with the keyword related to the keyword among the variable parts in the category sentence of the searched template, and replaces the other variable part with the word of the representative word sentence. Are extracted as example sentences related to the keyword.

With reference to FIG. 6A and FIG. 6B, the processing of the example sentence selection unit 444 when extracting the template number from the input character string will be described more specifically.

For example, when a keyword having the notation 614 “bag” is selected, the example sentence selecting unit 444 extracts the keyword / category data 425 c having the notation 614 “bag”, and the category number “ C0010 "and" C0011 "are acquired.

Then, the example sentence selection unit 444 acquires the template number 510 (10100, 10151, 10200) corresponding to the acquired category number “C0010” and the template number 510 corresponding to “C0011” based on the category / template data 426. . Then, the template 500 having the acquired template number 510 is searched and acquired from the template database 422.

Also, for example, when a keyword having the notation 614 “Japan” is selected, the example sentence selection unit 444 extracts keyword / category data 425a and 425b having the notation 614 “Japan”. Then, the example sentence selecting unit 444 acquires the category number “C3001” of the notation 614 “Japan” from the extracted keyword / category data 425a and 425b.

Then, the example sentence selecting unit 444 searches the category / template data 426b regarding the acquired category number “C3001”, and acquires the template numbers (10100, 17151, 18000, 18500) corresponding to the acquired category number “C3001”. Then, the template 500 having the acquired template number 510 is searched and acquired from the template database 422.

The example sentence selection unit 444 creates an example sentence by replacing the variable part of the category sentence (Japanese) 530 of the acquired template 500 with a word. The example sentence selection unit 444 replaces the variable unit having the keyword selected by the keyword selection unit 442 as a candidate word with the selected keyword. Other variable parts are replaced with words described in the representative word sentence 520. As described above, an example sentence when there is no variable part (only a fixed part) can be created in the same manner.

Referring to FIG. 16 again, one example sentence among the example sentences shown in the example sentence list 1044 is a selection target and is shaded and displayed. FIG. 16 shows a case where an example sentence “Is there a tour to see penguins or dolphins?” Is selected. The example sentence selection unit 444 changes an example sentence to be selected in response to an instruction such as pressing the up key 304a, the down key 304b, the previous page key 322, and the next page key 324.

When the search / decision key 306 is pressed while the sentence list display screen 1040 is displayed, the example sentence selecting unit 444 displays a conversation sentence one display screen including a translation of the example sentence selected in the example sentence list 1044 on the display 130. . The operation of the conversation assist device 100 after the display of the conversation sentence single display screen and the conversation sentence single display screen will be described later.

If the number of example sentences displayed in the example sentence list 1044 exceeds the number of characters that can be displayed in one line on the screen, the example sentence selecting unit 444 displays the first letter of the example sentence in the example sentence list 1044. An example of such a screen is shown in FIG. The way of omitting the characters in the example sentence is the same as that by the keyword selection unit 442 described with reference to FIG. FIG. 17 also shows a screen for translating Japanese (first language) into English (second language). In FIG. 17, the keyword displayed in the keyword display field 1042, the text displayed in the example sentence list 1044, and the first text in the translation preview 1046 are in Japanese (first language). The second sentence in the translation preview 1046 is English (second language).

The translated sentence preview 1046 displays the example sentence selected in the example sentence list 1044 and the translated sentence of the selected example sentence. The displayed translation is created by the example sentence selection unit 444 based on the selected example sentence and the template database 422. The translation preview 1046 makes it easy for the user to select a desired example sentence. Since the user can select an example sentence by looking at the translated sentence, it is less likely to select an inappropriate example sentence.

(4) Keyword Addition / Erase The example sentence selection unit 444 controls the handwriting pad 120 when the sentence list display screen 1040 is displayed so that the handwriting pad 120 receives instructions for keyword addition and keyword deletion. When the handwriting pad 120 receives an instruction, the keyword selection unit 442 starts processing for adding or deleting a keyword. Note that the example sentence selection unit 444 also starts the keyword addition process when the right key 304c is pressed while the sentence list display screen 1040 is displayed.

When the handwriting pad 120 receives a keyword erasing instruction, the keyword selection unit 442 displays an alert window 1050 on the display 130. Upon receiving a signal indicating that the Y key has been pressed when the alert window 1050 is displayed, the keyword selection unit 442 deletes the input keyword and displays the initial screen 1020 on the display 130. On the other hand, upon receiving a signal indicating that the N key has been pressed when the alert window 1050 is displayed, the keyword selection unit 442 stops displaying the alert window 1050 and displays the sentence list display screen 1040 on the display 130 again.

When the handwriting pad 120 receives the keyword addition instruction, the keyword selection unit 442 starts the first additional keyword setting process. The first additional keyword setting process will be described below with reference to FIGS. FIGS. 18 to 21 are diagrams for explaining the screens displayed on the display when setting the first additional keyword.

18 to 20, the keyword input in the character input field 1026, the text displayed in the candidate keyword list 1032, and the text displayed in the example sentence preview 1034 are in Japanese (first language). In FIG. 21, the keyword displayed in the keyword display field 1042, the text displayed in the example sentence list 1044, and the first text in the translated text preview 1046 are in Japanese (first language). The second sentence in the translation preview 1046 is English (second language).

When the handwriting pad 120 receives a keyword addition instruction, the keyword selection unit 442 first displays a keyword addition screen 1310 shown in FIG. The keyword addition screen includes a character input field, a candidate keyword list, and an example sentence preview. Since no character is input after the keyword addition instruction, no character is displayed in the character input field, and no keyword is displayed in the candidate list keyword list. In the example sentence preview, an example sentence related to “Penguin”, which has been determined, is displayed.

When the return key 308 is pressed while the keyword addition screen 1310 is displayed, the keyword selection unit 442 displays the sentence list display screen 1040 on the display. That is, the processing unit 440 ends the keyword addition process and restarts the example sentence selection process.

When the character “T” is input when the keyword addition screen 1310 is displayed, the keyword selection unit 442 displays the keyword addition screen 1320 shown in FIG. Referring to FIG. 19, the candidate keyword list on keyword addition screen 1320 displays keywords related to “tsu”.

When the return key 308 is pressed while the keyword addition screen 1320 is displayed, the keyword selection unit 442 displays the sentence list display screen 1040 on the display. That is, the processing unit 440 ends the keyword addition process and restarts the example sentence selection process.

When the character “A” is input when the keyword addition screen 1320 is displayed, the keyword selection unit 442 displays the keyword addition screen 1330 shown in FIG. Referring to FIG. 20, the candidate keyword list on keyword addition screen 1330 displays keywords related to “Tsu”. Here, a screen in a state where the keyword “tour” is selected is shown.

Also, the example sentence preview on the keyword addition screen 1330 displays example sentences related to at least one of “penguin” and “tour”. Here, the keyword selection unit 442 distinguishes and displays the example sentence including only one of “penguin” and “tour” and the example sentence including both in the example sentence preview. Specifically, the keyword selection unit 442 displays an example sentence including only one of the keywords with an “analog” mark at the beginning. Also, an example sentence including both keywords is brought higher in the example sentence preview than an example sentence including only one keyword. With such a display, the user can easily grasp how the example sentence displayed in the example sentence preview relates to the input keyword.

Note that the keyword selection unit 442 may arrange the example sentences in the order based on the number of the selected keywords included in the example sentence, and distinguish the example sentence including all the selected keywords from other example sentences. Further, the degree of relationship between the keyword and the example sentence may be determined in advance, and the example sentence may be arranged in an order based on the degree of relation.

When the return key 308 is pressed while the keyword addition screen 1330 is displayed, the keyword selection unit 442 displays the sentence list display screen 1040 on the display. That is, the processing unit 440 ends the keyword addition process and restarts the example sentence selection process.

When the search / decision key 306 is pressed while the keyword addition screen 1330 is displayed, the keyword selection unit 442 displays a second sentence list display screen 1340 shown in FIG. Second sentence list display screen 1340 includes a keyword display field, an example sentence list, and a translated sentence preview.

In the keyword display field, the keyword “penguin” set first and the keyword “tour” selected on the keyword addition screen 1330 are displayed. In the keyword display field, “>>” is displayed between the initially set keyword and the additional keyword. The separation of additional keywords is not limited to this example, and may be “,” / “&” / “+”, etc., and may change the typeface for each keyword or change the character color. .

When the search / decision key 306 is pressed when the second sentence list display screen 1340 is displayed, the example sentence selection unit 444 displays one conversation sentence including the translation of the example sentence selected in the example sentence list on the keyword addition screen 1330. The screen is displayed on the display 130. The operation of the conversation assist device 100 after the display of the conversation sentence single display screen and the conversation sentence single display screen will be described later. When the return key 308 is pressed when the second sentence list display screen 1340 is displayed, the example sentence selection unit 444 displays the keyword addition screen 1330 on the display 130 again.

When the second sentence list display screen 1340 is displayed, the example sentence selection unit 444 controls the handwriting pad 120 so that the handwriting pad 120 receives instructions for adding additional keywords (second additional keywords) and deleting keywords. To do.

Hereinafter, the second additional keyword setting process will be described with reference to FIGS. FIGS. 22 to 25 are diagrams for explaining screens displayed on the display when setting the second additional keyword.

22 to 24, the keyword input in the character input field 1026, the text displayed in the candidate keyword list 1032, and the text displayed in the example sentence preview 1034 are in Japanese (first language). In FIG. 25, the keyword displayed in the keyword display field 1042, the text displayed in the example sentence list 1044, and the first text in the translated text preview 1046 are in Japanese (first language). The second sentence in the translation preview 1046 is English (second language).

When the keyword selection unit 442 receives a keyword addition instruction when the second sentence list display screen 1340 is displayed, the keyword selection unit 442 displays a keyword addition screen 1410 shown in FIG. When the return key 308 is pressed when the keyword addition screen 1410 is displayed, the keyword selection unit 442 displays the second sentence list display screen 1340.

23. The keyword addition screen 1420 shown in FIG. 23 is a screen that the keyword selection unit 442 displays on the display 130 when “RI” is input when the second sentence list display screen 1340 is displayed. The keyword addition screen 1420 has the same configuration as the screen displayed when setting the first additional keyword.

The keyword display field of the keyword addition screen 1420 displays keywords related to “RI”. Here, the keyword “Re-Shan” is selected. The example sentence preview displays example sentences related to keywords “Penguin” and “Tour”, and “Re-Shan”. The method of displaying example sentences by the keyword selection unit 442 is the same as the method of displaying FIG.

The keyword addition screen 1430 shown in FIG. 24 is a screen that the keyword selection unit 442 displays on the display 130 when the down key 304b is pressed when the keyword addition screen 1420 is displayed. On the keyword addition screen 1430, the keyword immediately below the keyword selected on the keyword addition screen 1420 is selected.

When the search / decision key 306 is pressed when the keyword addition screen 1430 is displayed, the keyword selection unit 442 displays a third sentence list display screen 1440 shown in FIG. The operation of the conversation assist device 100 when the third sentence list display screen 1440 is displayed is the same as that when the sentence list display screen 1040 or the second sentence list display screen 1340 is displayed. For example, when the handwriting pad 120 receives an instruction to delete a keyword, the keyword selection unit 442 displays an alert window 1450 on the display 130. However, the handwriting pad 120 does not accept addition of keywords. This is because the maximum number of keywords that can be set has already been set.

(5) Output of one conversation sentence When the search / decision key 306 is pressed when the sentence list display screen 1040, the second sentence list display screen 1340, or the third sentence list display screen 1440 is displayed, The translated sentence output unit 446 displays a conversation sentence one-item display screen 1510 as shown in FIG. Since the language mode from Japanese to English is set, the translated text output unit 446 displays the conversation text display screen 1510 set in English. In FIG. 26, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language).

Referring to FIG. 26, one conversation sentence display screen 1510 includes a sentence list display screen 1040, a second sentence list display screen 1340, or an example sentence selected on the third sentence list display screen 1440, and The translation of the set language (here, English) corresponding to the selected example sentence is included. A single conversation sentence display screen 1510 is a screen that is displayed when an example sentence “Is there a tour to see penguins or dolphins?” Is selected.

The translation is created by the translation output unit 446 based on the template database 422. Specifically, the translated sentence output unit 446 replaces the variable part of the category sentence (English) 550 with the English word corresponding to the Japanese word to be replaced with the variable part of the category sentence (Japanese) 530. Create as a translation.

It is assumed that data representing the correspondence between Japanese words and English words is stored in the storage unit 420 in advance. Specifically, an index is attached to the candidate word whose variable part can be replaced, and the translation output unit 446 searches for an English word corresponding to a Japanese word based on the index. However, the method in which the translation output unit 446 creates the translation of the example sentence is not limited to this.

Further, when an example sentence is selected, the processing unit 440 inputs the input keyword, the example sentence search result, the translation result, and the data structure of the conversation assist translation dictionary related to the selected example sentence (sentence number, variable part number, variable part) Storage unit 420 as history example sentence information 432 similar to the registered example sentence information 430 described in FIG. 11 in association with the number of words in each variable part and a character string indicating the variable part word for each variable part). Remember me.

When the voice key 318 is pressed when the single conversation sentence display screen 1510 is displayed, the translated sentence output unit 446 displays a single conversation sentence display screen 1520 in the voice reproduction mode shown in FIG. Also in FIG. 27, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language). On the single conversation sentence display screen 1520, the output volume setting is displayed in the title. In addition, on the single conversation sentence display screen 1520, the translated sentence is shaded.

When the clear key 310 or the return key 308 is pressed during the display of the single conversation text display screen 1520 in the voice playback mode, the translated text output unit 446 displays the single conversation text display screen 1510 on the display 130 again.

When displaying the single conversation text display screen 1520, the translation output unit 446 controls the handwriting pad 120 so that the handwriting pad 120 receives an instruction to set the speed of voice output of the translation. Here, it is assumed that five types of speed settings can be set. FIG. 27 shows a case where an output speed of “standard” is selected, and “standard” is highlighted on the handwriting pad 120.

When the search / decision key 306 is pressed while the single conversation sentence display screen 1520 is displayed, the translated sentence output unit 446 displays the voice reproduction screen 1530 shown in FIG. 28 on the display 130 and sets the volume and speed. Therefore, the translation is output using TTS (speech synthesis technology, abbreviation for Text to Speech). Also in FIG. 28, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language). Audio playback screen 1530 includes a display 1532 indicating that audio playback is in progress. In addition to TTS, voice data recorded in advance in association with example sentences / translations may be used.

Referring to FIG. 26 again, at the time of displaying one conversation sentence display screen 1510, the translation output unit 446 controls the handwriting pad 120, and the handwriting pad 120 adds keywords, replaces words, switches language modes, and example words. Accept book registration instructions.

When the handwriting pad 120 receives a keyword addition instruction, the translated text output unit 446 displays a keyword addition screen and starts an additional keyword setting process. The additional keyword setting process is the same as that already described. Note that when the maximum number of keywords has already been set, the handwriting pad 120 does not accept a keyword addition instruction.

When the handwriting pad 120 receives a word replacement instruction, the translation output unit 446 starts a word replacement process. The word replacement process will be described later.

When “Register in word book” displayed on the handwriting pad 120 is selected when the conversation text display screen 1510 is displayed, the processing unit 440 inputs an input keyword, an example sentence search result, and a translation related to the displayed example sentence. As a result, data of the data structure of the conversation assist translation dictionary (sentence number, number of variable parts, number of words in the variable part for each variable part, and character string indicating the word in the variable part for each variable part) Thus, the stored example sentence information 430 described in FIG.

When the handwriting pad 120 receives the language mode switching instruction, the translated text output unit 446 switches the language mode, and displays on the display 130 a single conversation text display screen 1540 in the language mode after switching shown in FIG. In FIG. 29, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is in Chinese (third language). Note that the translated text output unit 446 also switches the language mode when the switch key 316 is pressed, and displays the single conversation text display screen 1540 (in Chinese setting) in the language mode after switching on the display 130.

Referring to FIG. 29, the Chinese-speaking one conversation sentence display screen 1540 displays the Chinese translation of the example sentence. When the voice key 318 is pressed while the one conversation sentence display screen 1540 is displayed, the translated sentence output unit 446 displays the one conversation sentence display screen 1550 in the voice reproduction mode shown in FIG. In FIG. 30, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is in Chinese (third language).

(6) Word replacement The conversation assist device 100 according to the present embodiment can replace words in the variable part of the example sentence displayed on the single conversation sentence display screen. Hereinafter, variable word replacement processing will be described.

The user can replace the words in the variable part by performing a predetermined process when displaying the screen for displaying one conversation sentence. The operation for exchanging words is roughly divided into key input for designating a variable portion and word replacement instruction input to the handwriting pad 120.

First, key input to specify the variable part will be described. Referring again to FIG. 26, translation output unit 446 displays a symbol for identifying the variable part in association with the variable part of the example sentence on the single conversation sentence display screen 1510. Here, [1] [2] [3] added before the word of the variable part is a symbol for specifying the variable part. However, the symbol for specifying the variable part is not limited to this, and may be an alphabetic character or a number. In addition, the translation output unit 446 may indicate the variable part by a difference in format such as font or underline type.

When the key corresponding to the symbol specifying the variable part is pressed, the translation output unit 446 starts the word replacement process of the variable part corresponding to the pressed key. Thus, the user can start the word replacement process of the variable part with a simple operation.

Specifically, when the [1] key is pressed when the single conversation sentence display screen 1510 is displayed, the translation output unit 446 displays the word replacement screen 1610 shown in FIG. In FIG. 31, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). On the word replacement screen 1610, the variable part corresponding to the pressed key is displayed in a shaded manner. Also, candidate words that can replace the variable part corresponding to the pressed key are displayed at the bottom of the screen. The candidate word is also Japanese (first language).

Note that the translation output unit 446 may emphasize the variable part corresponding to the pressed key by means other than shading, for example, underline, color change, and font change. Also, the translated sentence output unit 446 may perform shading, underlining, color change, font change, etc. so that the user can easily recognize the variable part that is not selected.

On the other hand, the translation output unit 446 also starts the word replacement process when the handwriting pad 120 receives a word replacement instruction when the single conversation text display screen 1510 is displayed. At this time, the translation output unit 446 displays the word replacement screen 1620 shown in FIG. In FIG. 32, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language). The word replacement screen 1620 is a word replacement screen before selection of the variable part. Note that the translation output unit 446 also displays the word replacement screen 1620 on the display 130 when the return key 308 is pressed when the word replacement screen 1610 is displayed.

Referring to FIG. 32, on the word replacement screen 1620, the selected variable part is displayed in a shaded manner. FIG. 32 shows a screen when the first variable part is selected. When the right key 304c or the left key 304d is pressed while the word replacement screen 1620 is displayed, the translation output unit 446 changes the variable unit to be selected.

When the search / decision key 306 is pressed when the word replacement screen 1620 is displayed, the translation output unit 446 displays the word replacement screen 1610 on the display 130. In addition, when the key corresponding to the symbol specifying the variable part is pressed, the translation output unit 446 displays a word replacement screen of the variable part corresponding to the pressed key on the display 130. When the [1] key is pressed, the translation output unit 446 displays the word replacement screen 1610 on the display 130.

Referring to FIG. 31 again, the replacement of the words of the selected variable part will be described. The method for replacing words is roughly divided into a method for directly selecting and determining a keyword displayed on the word replacement screen, and a method for selecting and determining a keyword after inputting characters to narrow down the keyword.

First, the former method will be described. For example, when the down key 304b is pressed while the word replacement screen 1610 is displayed, the translation output unit 446 displays the word replacement screen 1630 in which the selected keyword is changed as shown in FIG. In FIG. 33, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). The keyword selected on the word replacement screen 1630 is “whale”.

When the search / decision key 306 is pressed while the word replacement screen 1630 is displayed, the translation output unit 446 displays the word replacement screen 1640 shown in FIG. In FIG. 34, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language). The word replacement screen 1640 includes an example sentence in which the selected variable part is replaced with the word (whale) selected on the word replacement screen 1630 and a translation of the example sentence. When the return key 308 is pressed while the word replacement screen 1640 is displayed, the translation output unit 446 displays a single conversation sentence display screen for the example sentence after the word replacement on the display 130.

33, when the return key 308 is pressed when the word replacement screen 1610 is displayed, the translation output unit 446 displays the word replacement screen 1620 before the selection of the variable part on the display 130. Of course, the word replacement screen 1610 is displayed when the up key 304a is pressed.

Next, the latter method, that is, a method of narrowing down keywords by inputting characters will be described. The translation output unit 446 is selected when a character is input during display of the word replacement screen (for example, the word replacement screen 1610 shown in FIG. 31 and the word replacement screen 1630 shown in FIG. 33) in which the variable unit is selected. Among the candidate words of the variable part, those related to the inputted character are extracted, and a word replacement screen for displaying the extracted word is displayed on the display 130. As will be described later with reference to FIGS. 37 and 38, the variable portion is configured not only to select / determine a keyword but also to allow the user to input an arbitrary character string (numbers such as date and number). Such keyword input is called free input.

The translation output unit 446 extracts a word related to the input character by the same processing as the prediction conversion unit 443. That is, the translation output unit 446 extracts a word that partially matches the input character string as a word related to the input character string. In the following description, the translation output unit 446 will be described as extracting candidate words that match the input character string in the forward direction. However, the translation output unit 446 may extract candidate words based on other criteria such as complete match or partial match.

Specifically, when the character “ko” is input when the word replacement screen 1610 or the word replacement screen 1630 is displayed, the translation output unit 446 displays the word replacement screen 1650 shown in FIG. In FIG. 35, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). The word replacement screen 1650 displays a word having a letter “ko” at the head among candidate words that can replace the selected variable part.

When the search / decision key 306 is pressed while the word replacement screen 1650 is displayed, the translation output unit 446 displays the word replacement screen 1660 shown in FIG. In FIG. 36, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language). The word replacement screen 1640 includes an example sentence in which the selected variable part is replaced with the word (koala) selected on the word replacement screen 1650 and a translation of the example sentence. When the return key 308 is pressed while the word replacement screen 1660 is displayed, the translation output unit 446 displays a single conversation sentence display screen for the example sentence after the word replacement on the display 130.

In the example described above, the translated sentence output unit 446 does not change the example sentence and the translated sentence on the word replacement screen until the user selects a candidate for replacing the variable part and presses the search / decision key 306. For example, in the screen of FIG. 33, although the candidate word “whale” is selected, the displayed example sentence and translation remain as when the variable part is a penguin.

However, the translation output unit 446 may change the displayed example sentence and translation in accordance with the change of the selected candidate word. In this case, for example, as shown in FIG. 33, when the selected candidate word is changed to “whale”, the translated sentence output unit 446 displays an example sentence and a translated sentence when the variable part is a whale (shown in FIG. 34). Stuff).

If the displayed example sentences and translations are changed when the candidate word is determined, the flickering of the screen is reduced and the screen is easy to see. On the other hand, if the displayed example sentence and the translated sentence are changed according to the change of the selected candidate word, the user can determine the candidate word after knowing the translated sentence after the change.

Also, when replacing the variable part with a word, the translation output unit 446 changes the word to be replaced, if necessary, according to the words before and after the variable part. For example, when replacing the variable part of the category sentence “Which is not {VB_BUY1 + v.kanou} in Japan?” With the word “buy”, the category sentence “+ v.kanou” (possible expression Based on the “specified” part, “buy” is transformed into “buy” and replaced. By this processing, the displayed example sentence becomes a correct appearance.

On the other hand, the translation output unit 446 displays the pre-deformation word, not the post-deformation word, when displaying the word that can replace the variable part. That is, the translation output unit 446 displays “buy” and “go” as candidate words instead of “buy” and “go”, for example. This saves processing time for displaying candidate words. In addition, the display of candidate words is lexicographic and natural.

Of course, even if the candidate word is displayed in the basic form, in the sentence of the replacement result of the variable part, the replaced word is transformed into an appropriate form. As an example, replace the variable part “Go” (possible, with the designation of continuous form) of “Unfortunately I cannot go”. If the replacement candidate “Yes” is selected, the replacement result is “unfortunately not possible”. The replacement result will not be "do not do" or "do not do".

By the way, in some variable parts, the character types of replaceable words are limited, and it is more realistic for the user to input than selecting from the replacement candidates. For example, it is a variable part of numbers such as an amount and a postal code, or a variable part of English letters such as a label and an area display. When such a variable part is selected as a change target, the translation output unit 446 automatically changes the character input mode according to the character type that can be input to the variable part when there are a plurality of character input modes. Here, the character input mode is a setting that determines the type of character to be input when a certain key is pressed. When there are a plurality of character input modes, the user can input a plurality of types of characters (Hiragana, alphabets, numbers) with one key by switching the input mode. Hereinafter, this operation will be specifically described with reference to FIGS. 37 and 38.

FIG. 37 is a diagram showing a word replacement screen 1710 displayed when a numeric variable part is selected. In FIG. 37, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). When displaying the word replacement screen 1710, the translation output unit 446 switches the input mode to numeric input. In addition, the translation output unit 446 displays 10 keys on the handwriting pad 120 and accepts numeric input. When the search / decision key 306 is pressed after accepting the number input, the translated sentence output unit 446 displays the example sentence in which the selected variable part is replaced with the inputted number and the translated sentence of the example sentence.

FIG. 38 is a diagram showing a word replacement screen 1810 displayed when a numeric variable part is selected. In FIG. 38, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). When displaying the word replacement screen 1810, the translation output unit 446 switches the input mode to numeric input. In addition, the translation output unit 446 displays 10 keys on the handwriting pad 120 and accepts numeric input. When the search / decision key 306 is pressed after accepting the number input, the translated sentence output unit 446 displays the example sentence in which the selected variable part is replaced with the inputted number and the translated sentence of the example sentence.

Also, when a variable part that restricts the types of characters that can be replaced, such as date and month name, is selected, the translated text output part 446 displays a word replacement screen corresponding to the selected variable part on the display 130. When the word replacement screen is displayed, the translated text output unit 446 does not accept input of characters that cannot replace the selected variable unit.

FIG. 39 shows an example of a word replacement screen corresponding to the selected variable part. In FIG. 39, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). In the word replacement screen 1910 shown in FIG. 39, the variable part “MM month” is selected. The variable part “MM month” is a variable part that can be replaced with a month name. On the word replacement screen 1910, words (January, February,...) That can replace the variable part “MM month” are displayed, and the user selects one word from the displayed words.

Alternatively, the user can also specify a month name to replace the variable part “MM month” by entering a number. However, even if a number such as “13” is entered, the month name cannot be specified. When a number that does not correspond to the month name is input, the translation output unit 446 displays or voices that the input of such a number is inappropriate, and invalidates the input number.

(Modification of the word replacement screen)
The word replacement screen is not limited to the one described above. For example, the translation output unit 446 may display a word replacement screen as described below on the display 130 so that the user can easily replace words in the variable part.

As a first modification, the translation output unit 446 can create a word replacement screen in which words that can replace the variable unit are arranged in the order of notation. That is, it is possible to create a word replacement screen in which words having common notation of characters at predetermined positions are collectively displayed.

Referring to FIG. 40, description will be given of how words are arranged by the translated sentence output unit 446 according to the first modification. Each word shown in FIG. 40 is composed of Japanese kanji. The words in the leftmost column in FIG. 40 are not arranged in the notation order (arranged in the reading order). The words in the center column of FIG. 40 are classified and displayed according to characteristics such as the notation “elementary”, “vinegar”, and “form” of the first letter of the word, while the words in the rightmost column of FIG. Cooking methods are categorized and displayed according to characteristics such as “Fried”, “Boiled”, and “Yaki”. When the words are arranged in the order of notation as shown in the center column and the rightmost column, the user can easily grasp similar words. In particular, when classified according to the notation of the last character, words of the same category are easily displayed together like XYZ hotel and ABC company, so that the user can easily grasp the category.

As a second modified example, the translation output unit 446 can create a word replacement screen in which words that can replace the variable unit are arranged in order of division. That is, it is possible to create a word replacement screen in which words with common categories are displayed together.

Here, it is assumed that the word classification is stored in the storage unit 420 in advance. For example, it is assumed that the classification of each word is registered in the dictionary 423 in association with each word. Note that the classification is a further category classification. By setting the division in this way, words that can replace the variable part specified in the category can be further classified. As another example, it is also possible to register the frequency and importance of words as segment data.

Referring to FIG. 41, description will be given of how words are arranged by the translated sentence output unit 446 according to the second modification. Each word shown in FIG. 41 is composed of Japanese kanji or katakana. The words shown in the left column are arranged according to the reading. On the other hand, the words shown in the right column are arranged according to the classification of “the person indicated by occupation name” in the first half and “the person indicated by family relationship” in the second half. When a word is shown as in the right column, the user can easily grasp the category. In the display of the reading order, it is easy to search when the target word is completely fixed, but in the display according to the second modification, the user can easily grasp the word classification, and the meaning and concept of the word. It is easy to determine the target word.

Here, if the basic display, the first modified example, and the second modified example can be replaced by the user's specification, the degree of freedom can be further increased.

As a third modification, an example will be described in which words that can change the variable part are grouped and displayed. The translation output unit 446 according to the third modified example collects the words having the same predetermined number of characters at the beginning into one representative display when the number of words that can replace the selected variable unit exceeds a predetermined number. To display.

42A and 42B show display examples of words according to the third modification. FIG. 42A shows an example in which words having the same first three letters are grouped and displayed. The character string (Nice, Kneader, etc.) prefixed with “+” displayed at the left of FIG. 42A and “...” at the end is a representative display of a plurality of grouped words. When the representative display is selected, the translation output unit 446 displays a plurality of words corresponding to the representative display. When a predetermined number of characters or more are input, the translation output unit 446 expands the grouping and displays a word corresponding to the input characters.

This type of display is effective when words cannot be displayed on the screen. For example, the translation output unit 446 performs such grouping display when the number of words exceeds the number that can be displayed on the screen. Further, according to this grouping display, the user can easily select a desired word. This display is particularly advantageous when the number of words is large.

42B is a diagram showing an example in which the number of words corresponding to the representative display is displayed on the right of the representative display in addition to the display of FIG. 42A. According to this display, although the representative display is selected, the corresponding word is scarce or too many, and the user can search for the desired word smoothly.

As a fourth modification, an example will be described in which the number of words that can replace each variable part is displayed in association with each variable part on the word replacement screen. When displaying the word replacement screen, the translation output unit 446 according to the fourth modification obtains the number of words that can replace the variable part included in the displayed example sentence based on the template database 422 and the dictionary 423.

Specifically, the translation output unit 446 acquires a category of words whose variable part can be replaced based on the template database 422. Then, the translation output unit 446 searches the dictionary 423 for words included in the acquired category, and obtains the total number of searched words as the number of words that can replace the variable unit. The translation output unit 446 may use index data 424 instead of the dictionary 423. In addition, in a case where data in which each category is associated with a word included in each category is stored in the storage unit 420, the translation output unit 446 may use the data instead of the dictionary 423.

FIG. 43 shows an example of a word replacement screen according to the fourth modification. In FIG. 43, the first sentence in the conversation sentence display area 1511 is Japanese (first language). The second sentence in the conversation sentence display area 1511 is English (second language). Referring to FIG. 43, the number of words that can replace each variable part is displayed to the left of each variable part word displayed on the word replacement screen. By displaying the number of candidates before replacement, the user can know in advance how many variations of the example sentences are present.

As a fifth modification, an example will be described in which translated words of replacement words are simultaneously displayed as reference information of replacement words. When the replacement word candidate is displayed, the translation output unit 446 according to the fifth modification retrieves the translation of the candidate word from the dictionary 423 and displays it together with the candidate word.

FIG. 44 shows an example of a word replacement screen according to the fifth modification. In FIG. 44, the first sentence in the conversation sentence display area 1512 is Japanese (first language). The second sentence in the conversation sentence display area 1512 is English (second language). In the candidate word area 1513 displayed below the conversation sentence display area 1512, the words listed on the left are Japanese (first language), and the words listed on the center are English (second And the word listed on the right is Chinese (third language).

44, English (second language) and Chinese (third language) corresponding to the candidate word are displayed next to the candidate word (Japanese which is the first language). By knowing the translated words in advance before the replacement, the user can work efficiently. In particular, a situation in which the translated words are the same even though the words of the variable part are replaced (for example, such a situation is shown in FIG. 44, in the example shown in FIG. Can be avoided). Also, with this function, the conversation assist device 100 can be used as a learning tool with a word memo.

Each of the above modifications may be performed alone or in combination as appropriate. According to these modified examples, the user can change the word of the variable part to a desired word more easily.

(7) Selection of registered example sentence Selection of a registered example sentence will be described in detail with reference to FIG. FIG. 45 is a diagram for explaining the screen displayed on the display 130, the transition of the screen, and the state of the handwriting pad 120 in the word book function of the conversation assist application. In FIG. 45, the sentence in the registered example sentence list display field 5214 is Japanese (first language).

In the present embodiment, conversation assist device 100 selects “conversation assist” from the list on the list screen of the word book displayed by performing an operation for viewing the word book of all applications. When “conversation assist” is selected by operating the search / decision key 306 in the state, a word book screen 5210 of the conversation assist application is displayed.

Referring to FIG. 45, the word book screen 5210 includes a title 5212 and a registered example sentence list display field 5214. Here, the “word book” means a function for storing example sentences / translations and translation dictionary data acquired by the user. This is realized by selecting the item “Record in word book” on the handwriting pad of FIG. As a result, the user can intentionally hold the acquired example sentence / translation and reuse it later.

Title 5212 represents the name of the currently activated application, that is, “conversation assist”. In the registered example sentence list display field 5214, the example sentence search result is read from the data stored in the storage unit 420 as the registered example sentence information 430, and a list number is assigned to the head of each registered example sentence of the read example sentence search result. Are displayed as a list of registered example sentences.

45. In the registered example sentence list display field 5214 of the word book screen 5210 shown in FIG. 45, registered example sentences to which list numbers from “1” to “3” are assigned are displayed.

When the character key 302 indicating the list number is pressed while the word book screen 5210 is displayed, the processing unit 440 displays the one-item display screen 5220 of the registered example sentence of the list number on the display 130. This one case display screen 5220 is the same as the one conversation sentence display screen described in FIG. In FIG. 45, the first sentence 5224 in the conversation sentence display area 1511 is Japanese (first language). The second sentence 5225 in the conversation sentence display area 1511 is English (second language).

When the return key 308 or the clear key 310 is pressed while the word book screen 5210 is displayed, the processing unit 440 displays the initial screen 1020 described with reference to FIG.

When the return key 308 or the clear key 310 is pressed while the one-item display screen 5220 is displayed, the processing unit 440 redisplays the word book screen 5210 on the display 130.

(8) Selection of history example sentence Selection of a history example sentence will be described in detail with reference to FIG. FIG. 46 is a diagram for explaining the screen example displayed on the display 130, the transition of the screen, and the state of the handwriting pad 120 in the history example sentence function of the conversation assist application. In FIG. 46, the sentence in the history example sentence list display field 5314 is Japanese (first language). Here, the “history example sentence” means a function for storing example sentence / translation sentence and translation dictionary data acquired by the user. The difference from the above-mentioned “word book” is that the conversation assist device automatically holds the example sentences and translations acquired by the user. Since it is automatically recorded, it is possible to reuse a record of example sentences and translations acquired retroactively without the user intentionally saving them.

In this embodiment, the conversation assist device 100 displays a history example sentence screen 5310 of the conversation assist application when an operation for viewing a history example sentence is performed.

Referring to FIG. 46, the history example sentence screen 5310 includes a title 5312 and a history example sentence list display field 5314.

Title 5312 represents the name of the currently running application, ie, “conversation assist”, and the name of the function currently being executed, ie, “history example sentence”. In the history example sentence list display field 5314, the example sentence search result is read from the data stored in the storage unit 420 as the history example sentence information 432, and the list number is assigned to the head of each history example sentence of the read example sentence search result. , Displayed as a list of historical example sentences.

46. The history example sentence list display field 5314 of the history example sentence screen 5310 shown in FIG. 46 displays a list of history example sentences each assigned with a list number from “1” to “3”.

When the history example sentence screen 5310 is displayed, when the character key 302 indicating the list number or the search / decision key 306 is pressed while the list number is selected, the processing unit 440 displays the history of the list number. A display screen for displaying one conversation sentence of the example sentence is displayed on the display 130. This one conversation sentence display screen is the same as the one conversation sentence display screen described in FIG.

When the return key 308 or the clear key 310 is pressed when the conversation sentence one-item display screen is displayed and when the history example sentence screen 5310 is displayed, the processing unit 440 displays the initial screen 1020 described with reference to FIG. To display.

(4. Keyword ingenuity)
The conversation assist device 100 according to the present embodiment is devised as follows for keyword input or result display according to keyword input.
(I) Addition of reading variations (ii) Supplementary explanation to ambiguous word entries (iii) Keyword end sentence expression (iv) Segmentation of keywords Hereinafter, these ideas will be described.

(I) Addition of Reading Variations The conversation assist device 100 according to the present embodiment can deal with reading variations input by the user by using data in which a plurality of readings are assigned to one keyword. Can do. Hereinafter, the data structure will be described with reference to FIG.

The keyword 580 of the template 500a shown in FIG. 5 indicates that the template 500a has four keywords respectively corresponding to the four variable parts of the category sentence. Here, the keyword corresponding to the nth variable portion is represented in the form of “n: (keyword notation) = (character input (Japanese input))”. Note that another keyword “?” Included in the keyword 580 will be described later.

* One keyword is assigned to each keyword. Here, four keywords having the notations “Japan”, “buy”, “bag”, and “which” are shown.

When the character string indicated in the character input is input during the keyword selection process, the conversation assist device 100 considers that the keyword corresponding to the character input has been selected. In the present embodiment, a plurality of character strings are assigned as character input to a keyword having a plurality of character input variations. For example, two character strings “Nippon” and “Nihon”, which are readings of “Japan”, are assigned as character inputs to a keyword whose notation is “Japan”.

With this data structure, the example sentence selection unit 444 can display many example sentences corresponding to the character input accepted by the input unit 410 on the display unit 462. Therefore, the conversation assist device 100 according to the present embodiment is highly likely to display example sentences related to the keywords intended by the user.

When using data in which only one character string is associated with one keyword, if the character string entered by the user is slightly different from the character string assigned to the keyword, the example sentence intended by the user is searched. I can't. On the other hand, the conversation assist device 100 according to the present embodiment can cope with such a case.

More specifically, the operation of the conversation assist device 100 when a character string is input will be described. The conversation assist device 100 searches for an example sentence using the index data 424 created based on the template 500 instead of directly using the keyword 580 included in the template 500.

When the input unit 410 receives input of a character string, the keyword selection unit 442 acquires a keyword notation 614 having a Japanese input 612 that matches the input character string based on the keyword / category data 425.

The example sentence selection unit 444 acquires the category number 616 corresponding to the notation 614 acquired by the keyword selection unit 442 based on the keyword / category data 425. Then, a template number 510 corresponding to the acquired category number 616 is acquired based on the category / template data 426.

The operation of the processing unit 440 after obtaining the template number 510 is the same as that described in other places. That is, the example sentence selecting unit 444 searches for the template 500 having the acquired template number 510 and displays the example sentence (Japanese) designated by the searched template 500 on the display unit 462. The translation output unit 446 displays the translation (English or Chinese) of the example sentence selected from the displayed example sentences on the display unit 462.

As described above, the index data 424 is used instead of directly using the keyword 580 of the template 500 in order to speed up the search. However, the conversation assist device 100 may search for example sentences using the keyword 580 directly. In this case, the example sentence selection unit 444 acquires the template 500 in which a keyword having a character input that matches the input character string is set. Then, the example sentence selection unit 444 displays the example sentence on the display unit 462 based on the acquired template 500.

The character input corresponding to the keyword is not limited to the Japanese character string as shown in FIG. For example, romaji input may be set as character input for a keyword whose notation is Japanese. Some examples are shown below.
(Example 1) Window: Window = Windo; Windo = wiNdo:; uiNdou
(Example 2) Edinburgh: Edinburgh = Einbara; Ejinbara = edimbara; ejimbara
(Example 3) Japan: Japan = Japan; Japan = niqpoN; nihoN
Example 1 shows data related to a keyword whose notation is “window”. According to this data, when any of the character strings “Windo”, “Windo”, “wiNdo:”, and “uiNdou” is input, the example sentence selection unit 444 sets the keyword whose notation is “window”. Extracted template 500 is extracted. The same applies to Example 2 and Example 3.

In addition to this, for example, a character input such as “bus = bus; bus = basu; basu” may be set so that a keyword can be searched by performing either hiragana input or katakana input.

The device for setting the character input corresponding to the keyword as described above is not limited to the conversation assist device 100 according to the present embodiment, but the keyword is extracted according to the character input and the content associated with the keyword is output. Applicable to devices in general. For example, it can be applied to an electronic dictionary. In the case of an electronic dictionary, the headword of a phrase corresponds to a keyword, and the semantic data of the phrase corresponds to content.

(Ii) Supplementary Explanation to Polynomial Entry The conversation assist device 100 according to the present embodiment uses keywords as necessary so that the user can easily select an appropriate keyword when displaying the keyword on the display 130. A supplementary explanation is added to the notation to indicate a keyword.

Supplementary explanation is a character string that is uniquely set for each of a plurality of different keywords with the same notation. Character strings set for keywords having the same notation are different from each other for distinguishing keywords.

It is preferable to use a supplementary explanation character string that helps the user to select an appropriate keyword. Hereinafter, various types of supplementary explanations and specific examples thereof will be described.

A) Supplementary explanation due to difference in meaning For keywords having the same notation but different meaning, a character string representing the meaning of each keyword can be used as a supplementary explanation.

In Table 1, words that have the same katakana notation in Japanese (first language) but different kanji display, that is, different meanings are listed. That is, the English (second language) corresponding to each is different. The corresponding Chinese (third language) is also different.

The contents shown in the above example are displayed on the display (heading): keyword notation = keyword character input (reading) = keyword character input [keyword English notation (English translation, ie translation in the second language) / Chinese notation of keywords (Chinese translation, that is, translation in the third language)]. In this example, keywords are distinguished by supplementary explanations of (fish) (vehicle) (bath).

In particular, it is possible to use subdivisions of meaning as supplementary explanations for Japanese notation of words that are not distinguished in Japanese but are distinguished in foreign languages.

(Example)
Cousin (cousin, paternal): cousin = cousin = itoko
Cousin (cousin, maternal): cousin = cousin = itoko
Cousin (cousin, paternal): cousin = cousin = itoko
Cousin (cousin, maternal): cousin = cousin = itoko
Cousin (cousin, paternal): cousin = cousin = itoko
Cousin (cousin, maternal): cousin = cousin = itoko
Cousin (cousin, paternal): cousin = cousin = itoko
Cousin (cousin, mother): cousin = itoko
In Japanese, concepts that are grouped together with the word “cousins” are distinguished into different words in Chinese. Therefore, in addition to the Japanese notation, displaying a supplementary explanation indicating the detailed classification of the meaning corresponding to the translated word makes it easier for the user to select an appropriate keyword.

In particular, in the conversation assist, it is preferable that each foreign language notation is one for each original language (Japanese) notation. This is because when the variable part candidate word displayed on the display 130 is selected, the translated word can be immediately replaced. Therefore, unlike a dictionary that displays a plurality of meanings and translated words for one word, it is particularly preferable that the conversation assist is such that the words are distinguished in detail at the stage of displaying the original words.

B) Supplementary explanations for different usages Supplementary explanations for different usages can be set. Such an example is shown below.

In Table 2, words that have the same Hiragana notation in Japanese (first language), have no corresponding kanji display, and are used in different situations are listed. That is, the English (second language) corresponding to each is different. The corresponding Chinese (third language) is also different.

As in the above explanation, it is particularly preferable that the words are distinguished in detail at the stage of displaying the original language in this way, particularly in the conversation assist.

C) Supplementary explanation for different usage scenes Supplementary explanations for different usage scenes can also be set. Such an example is shown below.

In Table 3, words that have the same katakana notation in Japanese (first language), have no corresponding kanji display, and are used in different situations are listed. That is, the English (second language) corresponding to each is different. The corresponding Chinese (third language) is also different.

D) Supplementary explanation in notation When keywords can be distinguished by notation other than the set notation, notation other than the set notation may be used as supplementary explanation. Such an example is shown below.

In Table 4, words that have the same katakana notation in Japanese (first language) and corresponding kanji characters are listed. That is, the English (second language) corresponding to each is different. The corresponding Chinese (third language) is also different.

Such a supplementary explanation can be used in a language in which one word uses a plurality of notations such as Japanese.

E) Supplementary explanation with reading As a modified example of d), the reading of a keyword may be used as supplementary explanation. In a normal dictionary, it is rare that a plurality of keywords with different readings are displayed simultaneously. However, this may occur when displaying candidate words in the variable part. Therefore, the distinction by reading may be effective. An example of supplementary explanation by reading is shown below.

(Example 1)
Empty: empty = to = kara [empty / empty]
Sky: Sky = sora [sky / heaven]
(Example 2)
Haneda: Haneda = Hada = hada [Hada / Haneda]
Haneda: Haneda = Haneda = haneda [Haneda / Haneda]
In particular, this distinction is effective when the Roman name is different for Japanese names as in Example 2.
(Iii) Keyword end sentence expression Further, the conversation assist device 100 according to the present embodiment sets a keyword corresponding to the sentence end pattern of the category sentence of the template 500 as the keyword 580 of the template 500. Such a keyword setting may be performed by the conversation assist device 100 analyzing the end sentence pattern of the category sentence of the template 500 or may be given to the template 500 in advance.

Here, the sentence end pattern is an expression that determines the type of sentence (aspiration, request, denial, question, etc.). In Japanese, the expression at the end of a sentence corresponds to the type of sentence, so the expression that determines the type of sentence is called in this way for convenience. However, there are languages that express sentence types with expressions other than sentence end expressions. For a sentence in such a language, the type of sentence may be determined based on a predetermined standard, and a keyword corresponding to the type of sentence may be given. For example, in the case of English, if the words “Please” and “Would you ...?” Are included, the request is determined to be a request.

An example of the correspondence between the sentence end pattern and the set keyword is shown below. However, this is an example, and the keyword set in the keyword 580 may be determined using a correspondence relationship different from that shown here.

[End-of-sentence pattern]
In the following example, a sentence end expression is shown on the right side of |, and a keyword corresponding to the sentence end expression is shown on the left side of |.
Wish (aspiration) | I wanna
Please / Can you please (Request) | |! }
Can you please / can I get it (request)?
Can you please / request (request)?
No / No (Negative) | No.
No / No (Negative) | No.
Do you have any questions?
Do you have any questions?
Was it (question)?
May I ask (question)?
(Question) |
Was it (question)?
Is it (question)?
(Question) |
By using this keyword, the user can narrow down the types of example sentences to be searched. For example, the user can narrow down the example sentences to be searched by specifying the combination of “Go” + “Please / Can you (request)” or “Go” + “None / No (Negation)” .

Specific examples of example sentences including each of the specific sentence ending patterns shown above and keywords given to the example sentences are shown below. In the following example, the sentence prefixed with JD is a representative word sentence (Japanese) 520. A sentence prefixed with JP is a category sentence (Japanese) 530. A sentence with ED at the beginning is a representative word sentence (English) 540. The sentence prefixed with EN is a category sentence (English) 550. A character string prefixed with BK is a description of the keyword 580.

In the following, the characters corresponding to JD, JP, and BK are Japanese (first language). The letters corresponding to ED and EN are English (second language).

(Example 1)
I want to do JD {1: city tour}.
I want to do JP {1: & SIGHTSEEING-SURU}.
ED I'd like to take a {1: city sightseeing tour}.
EN I'd like to take a {1: & SIGHTSEEING-SURU}.
BK 1: City / Sightseeing = No / Kankou = shinai / kaNkou, Tai (Wish) = Tai = tai
(Example 2)
Please tell me the place of JD.
Please tell me the JP location.
ED Could you tell me where it is?
EN Could you tell me where it is?
BK place = basho = basho, teach = tell me = oshieru [teach = tell me], please / can you ask me (request) = please; can you do = kudasai; itadakemasuka
(Example 3)
Can you tell me JD?
JP Can you give me guidance?
ED Would you coach me, please?
EN Would you coach me, please?
BK Guidance = Shido = shidou, please / can you ask (request) = please; can you do = kudasai; itadakemasuka
(Example 4)
JD I don't have {1: gasoline} anymore.
JP There is no {1: & FUEL}.
ED There's no more {1: gas}.
EN There's no more {1: & FUEL}.
BK already = already = mou, 1: gasoline = is sorin = gasoriN, is = some = aru [yes = yes], no / not (denial) = not; not = nai; maseN
(Example 5)
Do you have to wait {1: how long} in JD {2: waiting room}?
Do I have to wait at {1: & QADV_DURATION-BASE} at JP {2: & ROOM}?
ED {1: How long} will I have to wait in the {2: waiting room}?
EN {1: & QADV_DURATION-BASE} will I have to wait in the {2: & ROOM}?
BK 2: Waiting room = Town Aitsu = machiaishitsu, 1: Which / about = what / about = dono / gurai, wait = matsu = matsu [waited = again], (wait = taiki = taiki) ) = Would you like = maseNka
(Example 6)
Do you have JD {1: small change}?
Is JP {1: & CURRENCY-CHANGE} available?
ED Do you have any {1: small change}?
EN Do you have any {1: & CURRENCY-CHANGE}?
BK 1: Coin = Kozeni = kozeni, there = some = aru [yes = there], do you (question) = do you do = masuka
(Example 7)
Has there been an announcement of JD {1: boarding} {2: already}?
Was there {2: & AV_YET-mou} announcement of JP {1: & ACT-BOARD-gen}?
ED Has {1: boarding} been announced {2: yet}?
EN Has {1: & ACT-BOARD-gen} been announced {2: & AV_YET-mou}?
BK 1: Boarding = Tojo = to: jou, Announce = Anausu = anauNsu, 2: Another = mou, Yes = Yes = aru [Yes = Yes]
(Example 8)
JD {1: First} Would you like to drink something?
JP {1: & AV_TO-START} Would you like to drink something?
ED {1: To start}, how about a drink?
EN {1: & AV_TO-START}, how about a drink?
BK 1: First = first = mazu, something = something = nanika, drink = nomu = nomu [drink = only], let's (question) = let's do = masho: ka
(Example 9)
When is the regular holiday of JD {1: here}?
When is the regular holiday of JP {1: & THIS_THAT-PLACE}?
ED Which days is {1: this place} closed?
EN Which days is {1: & THIS_THAT-PLACE} closed?
BK 1: Here = here = koko, regular holiday = departure = teikyu: bi, when = when = itsu, (question) = is = desuka
(Example 10)
Was it before JD {1: meal} or after?
Was it before or after JP {1: & FOOD}?
ED Was it before or after the {1: meal}?
EN Was it before or after the {1: FOOD}?
BK 1: Meal = Shokuji = Shokuji, before = mae, or = soretomo, after = after = ato, (question) = was it = deshitaka
(Example 11)
What line is JD?
JP What line is it?
ED What track number is it?
EN What track number is it?
BK What line = Nanbansen = nambaNseN, (question) = I wonder = desho: ka
(Example 12)
JD I don't need to {1: pay}, right?
JP I don't need {1: & VB_PAY + v.mi1}, right?
ED I don't need to {1: pay}, do I?
EN I don't need to {1: & VB_PAY + inf}, do I?
BK I = I; Watashi = watashi; watakushi, 1: Pay = Harau [Pay = Harawa], No = No = nai [None = No], Good = Good = ii Ne = desune
(Iv) Keyword Division In the conversation assist device 100 according to the present embodiment, there is a keyword for which “division keyword designation” has been made. A keyword designated as a split keyword is associated with a plurality of partial character inputs. When all the partial character inputs are input, the processing unit 440 extracts a keyword corresponding to the input partial character inputs.

By specifying a “split keyword” for a keyword, if a part of the keyword is selected as a keyword for search, additional input of a partial character corresponding to another part of the keyword You can search for keywords or example sentences corresponding to the keywords.

For example, consider a keyword set as follows.
(Example)
Embassy of Japan: Japan / Embassy = Nihon / TaishikaN
The keyword “Japanese Embassy” is divided into two parts, “Japan” and “Embassy”. Corresponding to these two parts, partial character input “Nihon” “Taikan” or “nihoN” “taishikaN” is set. The keyword may be divided into two or more parts.

The user can search for this keyword by designating the keyword “Japan” with the character input “Nihon” or “nihoN” and then designating the keyword “Embassy” with the character input “Taikan” or “taishikaN”. Conversely, the user can search for this keyword even after specifying “Embassy” and then specifying “Japan”. That is, the user can search for the keyword “Embassy of Japan” without entering “Nihontaikan” or “nihoNtaishikaN” from the beginning.

Especially, it is effective to specify a split keyword for a long keyword or a keyword that a user is supposed to input characters in a plurality of parts when searching.

By the way, partial character input may not be simply divided character input before division. Corresponding to such a case, it is preferable to devise the setting of partial character input. The following is a specific example of how to devise partial character input settings.

A) Muddy sound | Semi-voiced sound ⇒ Kiyoku Words that become muddy or semi-muddy sound due to combination with other words may return to clear sound by dividing. Corresponding to such a situation, it is preferable to provide variations in partial character input. Such an example is shown below.

(Example 1)
Reservation person: Reservation / person in charge = Yoyaku / Gakari; Takes = yoyaku / gakari; kakari
When inputting “reservation” and “relation” separately, it is conceivable that the user inputs “required” in the “relation” portion. Therefore, keyword setting is performed as described above.

Simply,
Reservation person: Reservation / person in charge = Yoyaku / Gakari = yoyaku / gakari
If the split keyword is specified, the above situation cannot be handled.

(Example 2)
Infant shoes: Infant / shoes = toothpick / gutsu; shoes = yo: ji / gutsu; kutsu
(Example 3)
Cooking book: Cooking / book = Ryo / bon; Hon = ryo: ri / boN; hoN
B) Reading aloud ⇒ Reading A word whose reading is limited to either reading aloud or reading aloud depending on the type of other words combined is input either as reading aloud or reading aloud in the case of divided input. Corresponding to such a situation, it is preferable to provide variations in partial character input. Such an example is shown below.

(Example 1)
YakuhiN / teN; mise: YakuhiN / teN; mise
Simply,
Pharmaceutical store: Pharmaceutical / Store = Yakuhin / Ten = yakuhiN / teN
Not. This is because when the “medicine” and the “store” are input separately, the user may input “show” in the “store” portion.

(Example 2)
Apple cider: Apple / Sake = apple / shu; salmon = riNgo / shu; sake
(Example 3)
Constipation: Constipation / drug = benpi / yaku; medicine = bempi / yaku; kusuri
C) Suffix ⇒ Noun Instead of a part of a keyword, the user may enter another word having the same meaning as that part. Corresponding to such a situation, it is preferable to provide variations in partial character input. Such an example is shown below.

(Example 1)
Insurance premiums: Insurance / Fee = Hoken / Ryoukin = hokeN / ryou; ryoukin
Simply,
Insurance fee: Insurance / Fee = Hoken / Ryou = hokeN / ryou
Not.

(Example 2)
Reservation: Reservation / book = Yokuaku / sho; Shorui = yoyaku / sho; shorui
(Example 3)
Ice cream shop: Ice cream / ya = ice cream / ya; mise = aisukuri-mu / ya; mise
D) “No”
In the present embodiment, negative “not” can be part of the keyword. For example, a keyword “not so many (words)” can be set. In this keyword, the word before “None” changes in shape before and after the division. Corresponding to such a situation, it is preferable to provide variations in partial character input. Such an example is shown below.

(Example 1)
Not too high: Not too high; Not too high; Not too high = Not too high; Tough / Not = amari / takaku; takai / nai
Simply,
Not too high: Not too high / Not too high = Not too high / Not high = amari / takaku; takai / nai
Not.

(Example 2)
Not well known: not / known; known / not = less / done; being / not = amari / shiraretei; shirareteiru / nai
E) “Going, coming, coming, being, trying” etc. [Part 1]
The sentence may contain expressions that overlap multiple verbs, such as “Go to / To / To / To”. If you want to search for such a sentence using keywords such as “going / coming / being / seeing” as auxiliary keywords, “going / coming / being / seeing” is also used as a split keyword. An example is shown below.

(Example 1)
Walking: walking; walking / going = going; going; going / going; going = aruite; aruku / iku; yuku
In this example, the “walking” part has a variation. "Walking: walking / going = there is / going; going = aruite / iku; yuku" is not.

(Example 2)
Visit: Visit; Visit; Visit / Come = Ask; Ask / Come = tazunete; tazuneru / kuru
F) “Going, coming, coming, being, trying” etc. [Part 2]
Unlike e), it may not be preferable to use keywords such as “go / come / is / see”. In this case, “go / come / be / see” is not used as a split keyword. An example is shown below.

(Example 1)
Closed: Closed; Closed = Closed; Closed = heiteNshiteiru; heiteN
This is the keyword "Closed: Closed = Closed = heiteNshiteiru" plus a split specification.

This keyword
Closed: Closed; Closed; Closed / Being = Heating; Being / Heating = heiteNshite; heiteN / iru
It can also be divided as follows. According to this division, the keyword is searched from the words “closed” and “is”. However, such a search result may not be preferable. "I" here is an auxiliary verb that means "Continue to do". On the other hand, “is” is sometimes used as a main verb representing the meaning of “exists”. Therefore, as described above, if “is” is part of the keyword, it may be difficult to search for an example sentence that uses “is” as the main verb.

(Example 2)
Try to ride: Try to get on; Ride = Try to ride; Noru = noqtemiru; noru
This is the keyword "Try to get on: Try to get on = Try to get on = noqtemiru" with the division specification.

This keyword
Ride: Ride; Ride / Look = Notte; Noru / Look = noqte; noru / miru
It can also be divided as follows. According to this division, this keyword is searched from the words “ride” and “see”. However, such a search result may not be preferable. As described above, when “view” is part of the keyword, it may be difficult to search for an example sentence using “see” as the main verb (that is, in the meaning of “see”).

(5. Process flow)
Hereinafter, the flow of processing performed by the conversation assist device 100 will be described.

FIG. 47 is a flowchart showing an outline of the flow of processing performed by the conversation assist device 100.

Referring to FIG. 47, processing unit 440 launches a conversation assist application based on an instruction received by input unit 410, and then, in step S <b> 101, processing unit 440, processing unit 440 is input unit 410. 1 selects and determines one language mode from among a plurality of language modes.

However, when the conversation assist device 100 has only one language mode (for example, a mode for translating Japanese into English), the processing unit 440 performs the process of step S101 after starting the conversation assist application. Without proceeding, the process proceeds to step S103.

In step S103, the processing unit 440 determines one selected keyword from a plurality of predetermined keywords based on the instruction received by the input unit 410. The selection keyword determination process will be described later with reference to FIG.

In step S105, the processing unit 440 searches for the template 500 related to the selected keyword determined in step S103. Then, processing unit 440 displays example sentences in the first language (Japanese in the present embodiment) in the searched template 500 as example sentence candidates. The example sentence candidate display process will be described later with reference to FIG.

In step S106, the processing unit 440 determines whether the input unit 410 has received a keyword addition instruction. The keyword addition instruction includes input to the handwriting pad 120 and pressing of the right key 304c while the sentence list display screen is displayed.

When it is determined that the input unit 410 has received an addition instruction (when determined Yes in step S106), the processing unit 440 repeats the processing from step S103. In other words, the processing unit 440 determines an additional keyword in step S103. In addition, in step S105, the processing unit 440 displays the keyword determined before the keyword addition and example sentence candidates related to the additional keyword.

When it is determined that the input unit 410 has not received an addition instruction (when it is determined No in step S106), the processing unit 440 proceeds to the process of step S107.

In step S107, the processing unit 440 selects one example sentence from the displayed candidate sentences based on the instruction received by the input unit 410. Specifically, the processing unit 440 selects one example sentence in response to an input of the up key 304a and the down key 304b and an input of the search / decision key 306.

In step S109, the processing unit 440 outputs the example sentence selected in step S107 and the translation of the selected example sentence. Specifically, the processing unit 440 causes the display 130 to display a single conversation sentence display screen that displays the selected example sentence and the translated sentence of the selected example sentence. In addition, when the voice key 318 is pressed while the single conversation text display screen is displayed, the processing unit 440 causes the speaker 140 to output the translated text as a voice.

In step S111, the processing unit 440 executes a history example sentence management process. The historical example sentence management process will be described later with reference to FIG.

In step S113, the processing unit 440 determines whether or not the input unit 410 has received a variable unit replacement instruction. The variable part replacement instruction includes pressing of the character key 302 and a word replacement instruction to the touch panel.

If it is determined that the input unit 410 has received the variable part replacement instruction (Yes in step S113), the processing unit 440 creates an example sentence and a translation in which the variable part is replaced based on the instruction received by the input unit 410 in step S115. And displayed on the display unit 462.

More specifically, the processing unit 440 selects a variable unit that replaces words based on the instruction. Then, processing unit 440 replaces the selected variable part of the example sentence with the word determined based on the instruction, and creates the example sentence after the replacement. Furthermore, the processing unit 440 replaces the variable part corresponding to the selected variable part of the example sentence in the translated sentence with the translation of the determined word, and creates a translated sentence after the replacement.

In step S117, the processing unit 440 determines whether or not “Register in word book” displayed on the handwriting pad 120 of the input unit 410 has been selected. If it is determined that “Register in word book” is not selected (No in step S117), the process to be executed proceeds to the process in step S121.

On the other hand, if it is determined that “Register in word book” is selected (Yes in step S117), in step S119, the processing unit 440 executes a registered example sentence management process. The registered example sentence management process will be described later with reference to FIG.

In step S121, the processing unit 440 determines whether the input unit 410 has received an instruction to end the conversation assist application. The conversation assist application end instruction includes, for example, pressing of an application switching key or pressing of a power key.

When it is determined that the input unit 410 has received an end instruction (when determined Yes in step S121), the processing unit 440 ends the conversation assist process.

On the other hand, when it is determined that the input unit 410 has not received an end instruction (when it is determined No in step S121), the processing unit 440 repeats the processing from step S113.

FIG. 48 is a flowchart showing the flow of a selected keyword determination process executed as a subroutine of the conversation assist application by the conversation assist device 100.

48, in step S201, the processing unit 440 receives from the input unit 410 the character input instruction received by the input unit 410.

In step S203, the processing unit 440 searches the predicted keyword list 428 for a predicted word corresponding to the character input acquired in step S201.

In step S205, the processing unit 440 displays the predicted word searched in step S203 on the display unit 462 as a keyword candidate.

Here, the details of step S205 will be described with reference to FIG. FIG. 49 is a flowchart showing the flow of keyword candidate display processing executed by the conversation assist device 100 as a subroutine of the conversation assist application.

Referring to FIG. 49, in step S <b> 301, processing unit 440 receives a keyword candidate selection instruction received from input unit 410 from input unit 410. Specifically, the selection instruction is an instruction in response to pressing of the upper key 304a, the lower key 304b, or the like.

In step S303, the processing unit 440 searches for an example sentence related to the predicted word selected by the selection instruction received in step S301. Specifically, the processing unit 440 acquires the template number 510 associated with the predicted word selected based on the index data 424. Then, the processing unit 440 searches the template database 422 for the template 500 having the acquired template number 510.

In step S305, the processing unit 440 displays, on the display unit 462, an example sentence preview in which keyword candidates and example sentences related to the predicted word selected based on the instruction received in step S301 are arranged. More specifically, the processing unit 440 displays the representative word sentence (Japanese) 520 of the template 500 searched in step S303 in the example sentence preview. Note that the processing unit 440 may display an example sentence in which the variable part is replaced with the selected predicted word in the example sentence preview.

After the processing of step S305, the processing unit 440 returns the processing to be executed to the calling source processing of this keyword candidate display processing.

Referring to FIG. 48 again, in step S207, processing unit 440 determines whether input unit 410 has accepted a keyword determination instruction. Specifically, processing unit 440 determines whether search / decision key 306 has been pressed.

When it is determined that the input unit 410 has received a keyword determination instruction (when determined Yes in step S207), the processing unit 440 determines the keyword selected at the time of receiving the determination instruction as a selected keyword. Then, the processing unit 440 returns the process to be executed to the caller process of the selected keyword determination process.

On the other hand, when it is determined that the input unit 410 has not received the keyword determination instruction (when it is determined No in step S207), the processing unit 440 repeats the processing from step S201.

FIG. 50 is a flowchart showing a flow of example sentence candidate display processing executed by the conversation assist device 100 as a subroutine of the conversation assist application.

Referring to FIG. 50, in step S401, processing unit 440 searches for an example sentence related to the selected keyword determined in step S103.

In step S403, the processing unit 440 acquires from the input unit 410 an instruction to select example sentences accepted by the input unit 410. When there is no selection instruction, the processing unit 440 considers that there is an instruction to select a default example sentence (for example, the first example sentence among the search results).

In step S405, the processing unit 440 searches the template database 422 for a translation of the example sentence designated by the selection instruction received in step S403.

In step S407, the processing unit 440 displays the search result in step S401 as an example sentence candidate. Further, the processing unit 440 displays the translated text searched in step S405 as a translated text preview.

FIG. 51 is a flowchart showing the history example sentence management process executed by the conversation assist device 100 as a subroutine of the conversation assist application.

Referring to FIG. 51, in step S5812, processing unit 440 causes input keyword, example sentence, translation, sentence number, variable part number, variable part word (number of words for each variable part, each variable part) for the example sentence being displayed. Is stored in the storage unit 420 as history example sentence information 432.

Next, in step S5821, the processing unit 440 determines whether or not to display a history example sentence is selected based on the instruction received by the input unit 410. If it is determined that the history example sentence display is not selected (No in step S5821), the processing unit 440 returns the process to be executed to the calling source process of the history example sentence management process.

On the other hand, when it is determined that the display of the history example sentence is selected (Yes in step S5821), in step S5822, the processing unit 440 stores the example sentence search result of the history example sentence information 432 from the storage unit 420 as a history. Read as example sentences. In step S5823, the processing unit 440 displays the history example sentence list of the read history example sentence information 432 as shown in the history example sentence screen 5310 of FIG. 46 so that the history example sentence can be selected. To display. The instruction regarding the display of the history example sentence may be realized using a dedicated call button (not shown).

Next, in step S5831, the processing unit 440 determines whether or not a list number has been selected based on the instruction received by the input unit 410.

When it is determined that the list number has been selected (when it is determined Yes at step S5831), at step S5832 the processing unit 440 displays the conversation sentence one sentence display screen of the history example sentence of the selected list number as shown in FIG. As shown in the single sentence display screen, it is displayed on the display unit 462.

Next, in step S5833, the processing unit 440 executes a single conversation text display screen process that is a process on the single conversation text display screen shown in FIG. When the return key 308 or the clear key 310 is pressed in the one conversation sentence display screen process and the process to be executed is returned to this process, the processing unit 440 advances the process to be executed to step S5836.

On the other hand, if it is determined that the list number is not selected (No in step S5831), in step S5835, the processing unit 440 determines whether the return key 308 or the clear key 310 is selected and pressed. to decide. When it is determined that neither the return key 308 nor the clear key 310 has been selected and pressed (when it is determined No at step S5835), the processing unit 440 returns the process to be executed to the process at step S5831.

On the other hand, when it is determined that either the return key 308 or the clear key 310 is selected and pressed (when it is determined Yes at step S5835), the processing unit 440 advances the process to be performed to the process of step S5836.

In step S5836, the processing unit 440 displays the conversation assist initial screen 1020 shown in FIG. In step S5837, the processing unit 440 executes conversation assist initial screen processing that is processing on the initial screen 1020.

Thus, by exchanging the variable part based on the history example sentence information 432, various similar example sentences can be easily obtained and displayed. For example, if the number of example sentences in the example sentence search result stored in the history example sentence information 432 is 1000, the average number of variable parts included in each example sentence is 3, and the average number of candidates for each variable part is 100, then 1000 × 100 An example sentence of × 100 × 100 = 1,000,000,000 can be obtained, and convenience can be improved.

FIG. 52 is a flowchart showing the flow of a registered example sentence management process executed by the conversation assist device 100 as a subroutine of the conversation assist application.

Referring to FIG. 52, in step S5912, processing unit 440 causes input keyword, example sentence, translation, sentence number, variable part number, variable part word (number of words for each variable part, each variable part) for the example sentence being displayed. Are stored in the storage unit 420 as the registered example sentence information 430.

Next, in step S5921, based on the instruction received by input unit 410, processing unit 440 determines whether or not the “conversation assist” application has been selected on the word book list screen of all applications of conversation assist device 100. Judging. If it is determined that “conversation assist” has not been selected (NO in step S5921), processing unit 440 returns the process to be executed to the caller process of this registered example sentence management process.

On the other hand, if it is determined that the “conversation assist” application has been selected (YES in step S5921), in step S5922, the processing unit 440 stores the example sentence search result of the registered example sentence information 430 from the storage unit 420. Read as. In step S 5923, the processing unit 440 displays the list of registered example sentences of the read registered example sentence information 430 as shown in the word book screen 5210 of FIG. 45 so that the registered example sentences can be selected. To display.

Next, in step S5931, the processing unit 440 determines whether a list number has been selected based on the instruction received by the input unit 410.

When it is determined that the list number has been selected (when it is determined Yes in step S5931), in step S5932, the processing unit 440 displays the one display screen for the registered example sentence of the selected list number as one display in FIG. As shown on the screen, it is displayed on the display unit 462.

Next, in step S5933, the processing unit 440 executes a single conversation text display screen process that is a process on the single conversation text display screen shown in FIG. When the return key 308 or the clear key 310 is pressed in the one-conversation-sentence display screen process and the process to be executed is returned to this process, the processing unit 440 displays the original list of registered example sentences in FIG. As shown in the 45 word book screen 5210, it is redisplayed on the display unit 462.

On the other hand, when it is determined that the list number is not selected (when No is determined in step S5931), and after step S5934, in step S5935, the processing unit 440 selects the return key 308 or the clear key 310. It is determined whether or not it has been pressed. If it is determined that neither the return key 308 nor the clear key 310 has been selected and pressed (when it is determined NO in step S5935), the processing unit 440 returns the process to be executed to the process of step S5931.

On the other hand, when it is determined that either the return key 308 or the clear key 310 is selected and pressed (when determined Yes in step S5935), the processing unit 440 advances the process to be performed to the process of step S5936.

In step S 5936, the processing unit 440 displays the conversation assist initial screen 1020 shown in FIG. 20 on the display unit 462. In step S 5937, processing unit 440 executes conversation assistance initial screen processing that is processing on initial screen 1020.

Thus, since the variable part can be replaced based on the registered example sentence information 430, similar example sentences can be easily displayed. Moreover, since the variable part is replaced, various example sentences can be displayed. For example, if the number of example sentence search results stored in the registered example sentence information 430 is 1000, the average number of variable parts included in each example sentence is 3, and the average number of candidates per variable part is 100, then 1000 × 100 An example sentence of × 100 × 100 = 1,000,000,000 can be obtained, and convenience can be improved.

The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

100 conversation assist device, 110 input key, 120 handwriting pad, 130 display, 132 display controller, 140 speaker, 142 voice controller, 172 conversation assist program, 180 hard disk, 210 first housing, 220 second housing, 230 Stylus pen, 302 character key, 304 direction key, 304a up key, 304b down key, 304c right key, 304d left key, 306 enter key, 308 return key, 310 clear key, 312 retreat key, 314 conversation assist key, 316 switching Key, 318 voice key, 320 menu key, 322 previous page key, 324 next page key, 410 input unit, 420 storage unit, 422 template database, 423 dictionary, 42 Index data, 425 keyword / category data, 426 category template data, 428 predicted keyword list, 430 registered example sentence information, 432 history example sentence information, 440 processing section, 442 keyword selection section, 443 prediction conversion section, 444 example sentence selection section, 446 Translated sentence output section, 448 registered example sentence management section, 449 history example sentence management section, 460 output section, 462 display section, 464 voice output section, 500 template, 510 template number, 520 representative word sentence, 580 keyword, 612 Japanese input, 614 Japanese notation, 616 category number, 700 category list.

Claims

An information processing apparatus (100),
An input unit (410);
A display (462);
A storage unit (422) for storing a template database (422) having a plurality of templates that associate a first example sentence described in a first language with a second example sentence describing the first example sentence in a second language 420),
The first example sentence has at least one first variable unit capable of replacing each of the plurality of first candidate words in the first language,
The second example sentence includes at least one second candidate word that can replace each of the second candidate words in which the plurality of first candidate words of the corresponding first example sentence are described in the second language. Having a variable part,
A processing unit (440) for controlling the operation of the information processing apparatus;
The processor is
A keyword selection unit (442) for designating a first keyword from a plurality of predetermined keywords based on a first instruction from the input unit;
Extracting at least one first example sentence having the first variable part capable of replacing the first keyword as the first candidate word, and extracting the first variable of the extracted first example sentence The first example sentence in which the first keyword is entered in a part is displayed on the display in a selectable manner, and the first example sentence displayed on the display based on a second instruction from the input part An example sentence selecting unit (444) for selecting one of the first example sentences from:
Based on the template, the second example sentence corresponding to the first example sentence selected is transferred to the second variable part corresponding to the first variable part containing the first keyword. An information processing apparatus comprising: a translation output unit (446) for inputting the second candidate word in which the keyword is described in the second language and displaying the second candidate word on the display.
The storage unit further stores index data that associates the keyword with the template having the first example sentence that can replace the candidate word that matches the keyword,
The information processing apparatus according to claim 1, wherein the example sentence selecting unit extracts the first example sentence based on the first keyword and the index data.
The index data is
Keyword category data in which the keyword is associated with the category;
Category template data associating the category with the template related to the first example sentence having the first variable unit capable of replacing the candidate word that matches the keyword corresponding to the category;
The example sentence selecting unit acquires the category corresponding to the first keyword based on the keyword category data, and based on the category template data, the example sentence selecting unit is included in the template corresponding to the acquired category. The information processing apparatus according to claim 2, wherein one example sentence is extracted.
The input unit accepts input of characters,
The keyword selection unit displays the keyword corresponding to the input character on the display in a selectable manner, and designates the first keyword from the displayed keywords based on the first instruction. The information processing apparatus according to claim 1.
The storage unit further stores a predicted keyword list that associates the input character with the keyword that partially or entirely matches the input character,
5. The information according to claim 4, wherein the keyword selection unit extracts the keyword corresponding to the input character based on the predicted keyword list, and displays the extracted keyword on the display. Processing equipment.
The keyword selection unit extracts a first example sentence having the first variable unit capable of replacing the first keyword, and is selected by the first variable unit of the extracted first example sentence. The information processing apparatus according to claim 4, wherein the first example sentence including the keyword and the keyword corresponding to the input character are displayed on the display.
The keyword selection unit specifies a second keyword from among a plurality of predetermined keywords based on a third instruction from the input unit,
The example sentence selecting unit extracts the first example sentence having the first variable unit capable of exchanging at least one of the first and second keywords, and the first example sentence of the extracted first example sentence is extracted. The information processing apparatus according to claim 1, wherein the first example sentence in which the first keyword is entered in one variable part is displayed on the display in a selectable manner.
The keyword selection unit specifies the second keyword from among a plurality of predetermined keywords based on a third instruction from the input unit,
The example sentence selection unit extracts the first example sentence having the first variable unit capable of replacing at least one of the first and second keywords, and selects both the first and second keywords. The first example sentence that can be exchanged and the first example sentence that can exchange one of the first and second keywords are distinguished and displayed on the display. Information processing device.
A first unit having an input unit, a display, a processing unit, and at least one first variable unit described in a first language and capable of replacing each of a plurality of first candidate words in the first language. And a second candidate word in which a plurality of the first candidate words in the corresponding first example sentence are described in the second language, respectively. An information processing method executed by an information processing apparatus including a storage unit that stores a template database having a plurality of templates that associate a second example sentence having at least one second variable unit that can be replaced with each other,
The processing unit accepting selection of a first keyword from a plurality of predetermined keywords via the input unit;
The processing unit extracting at least one first example sentence having the first variable unit capable of replacing the first keyword as the first candidate word;
The processing unit displaying the first example sentence in which the first keyword is entered in the first variable part of the extracted first example sentence on the display in a selectable manner;
Receiving a selection of one first example sentence from at least one first example sentence displayed on the display via the input unit;
The second variable corresponding to the first variable unit in which the first keyword is entered, the second example sentence corresponding to the first example sentence selected based on the template. And placing the second candidate word in which the first keyword is described in the second language and displaying the first keyword on the display.
The storage unit further stores index data that associates the keyword with the template having the first example sentence that can replace the candidate word that matches the keyword,
The information processing method according to claim 9, wherein the step of extracting the first example sentence includes a step of extracting the first example sentence based on the first keyword and the index data.
The index data is
Keyword category data in which the keywords and categories are associated;
Category template data associating the category with the template relating to the first example sentence having the first variable unit capable of replacing the candidate word that matches the keyword corresponding to the category;
Extracting the first example sentence based on the index data comprises:
Obtaining the category corresponding to the first keyword based on the keyword category data;
The information processing method according to claim 10, further comprising: extracting the first example sentence included in the template corresponding to the acquired category based on the category template data.
The processing unit accepting input of characters via the input unit;
The information processing method according to claim 9, further comprising the step of displaying the keyword corresponding to the inputted character on the display in a selectable manner.
The storage unit further stores a predicted keyword list that associates the input character with the keyword that partially or entirely matches the input character,
The step of selectively displaying the keyword on the display includes a step of extracting the keyword corresponding to the inputted character based on the predicted keyword list and displaying the extracted keyword on the display. The information processing method according to claim 12.
The step of selectively displaying the first example sentence on the display is input with the first example sentence in which the first keyword is entered in the first variable part of the extracted first example sentence. The information processing method according to claim 12, further comprising: displaying the keyword corresponding to the character on the display.
The processing unit further includes a step of receiving selection of a second keyword from a plurality of predetermined keywords via the input unit,
The step of extracting the first example sentence includes the step of extracting at least one first example sentence having the first variable part capable of replacing at least one of the first and second keywords. The information processing method according to claim 9.
The processing unit further includes a step of receiving selection of a second keyword from a plurality of predetermined keywords via the input unit,
Extracting the first example sentence includes extracting the first example sentence having the first variable unit capable of replacing at least one of the first and second keywords;
The step of selectively displaying the first example sentence on the display includes either the first example sentence that can replace both the first and second keywords, and the first and second keywords. The information processing method according to claim 9, comprising a step of distinguishing and displaying the first example sentence that can be replaced on the display.