JP3274652B2 - Translation equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates, when translating a sentence <br/> the first language to a second language, relating to the translation apparatus for performing translations using the example sentence registered in advance.

[0002]

2. Description of the Related Art At the present time when the number of opportunities to go abroad for sightseeing and the like has increased, many book examples of conversations that collect many conversational expressions have been published so that conversations in foreign countries are not troubled. Also,
A conversation translation apparatus has been proposed which digitizes these conversation example collections, selects a conversation example sentence by key operation or the like, and displays a translation of the selected conversation example sentence. For example, the invention of JP-A-6-325072 corresponds to this. Also,
A conversational sentence translator that automatically selects an optimal example sentence for conversation by freely inputting a sentence to be translated and displays the translated sentence has been proposed (Japanese Patent Application Laid-Open No. 9-062681). In this method, a large number of conversational example sentences and phrases characterizing each example sentence grouped for each synonym are stored. The former is called a fixed phrase, and the latter is called a semantic feature. In this method, semantic features are extracted from an input sentence, and a fixed sentence having a pattern close to a combination of the extracted features is searched. After selecting a fixed sentence for an input sentence, in order to select another fixed sentence, it was necessary to input the sentence again and search for the fixed sentence.

[0003] Further, for a conversational sentence translator that analyzes an input sentence, automatically selects an optimal example sentence for conversation, and displays the translated sentence, the previous conversation scene is stored. After that, a method has been proposed for improving the accuracy in retrieving a conversation sentence (Japanese Patent Laid-Open No. 7-143833). Also, register the conversation example sentence once searched,
From the second time, conversational sentence translations that can display necessary conversational sentences by simply selecting from a list of sentences registered once have already been commercialized.

[0004]

[Problem to be Solved by the Invention] There is a case where it is desired to say a plurality of sentences continuously in one scene . In such a case, in the conventional method, it was necessary to input a sentence again. As an example, the user first inputs a sentence, "Would you like to give me a little more?", Extract semantic features from this sentence, and match the input sentence with each fixed sentence to say, "It's a little cheaper." The sentence `` Nuka '' was selected and its English translation `` Could you makeit a little cheape
r? "is displayed. The user can use this statement
When negotiating with the clerk, but the other party gives a negative reply, the user may want to return a reply "Please show another" or "I will do it again". In order to search for a translation of such a sentence, in the conventional method, it was necessary to input these sentences from scratch. In the present invention, once a sentence is input, a sentence that may be used before or after the sentence is displayed according to a request from the user.

[0005] When the (challenge claim 2) a statement is entered, when attempting to display a statement that may be used in the front and rear, sea the statement belongs
There is a case in which emissions can not be determined to one. As an example, type the sentence "you pay by credit card", from the bilingual database "I would like to pay by credit card, but" statement that is selected, the statement that this "I want to pay by credit card, but" is, Food and Shopping 2
Suppose you belonged to a street scene . And the meal sea
The word "Let's split the bill" is included in the shopping scene , but it is not included in the shopping scene . Conversely, in the shopping scene , "Please pay in installments"
Is included. In such a case, is the genus to the scene of the selected sentence "credit card payment I want, but in", depending on whether restaurants or shopping,
The sentence used in the scene is also different. In the present invention,
In such a case, by selecting a scene to the user, displaying only statements belonging to that scene.

(Problem of Claim 3) When a sentence once searched is registered and used later, it is simply desired to use it from a registered list.
With the method of searching for sentences , sentences registered for different purposes may be mixed. As an example, the sentence "How much is a single room per night" that the user has registered for a conversation at a hotel and the sentence "Is this the only product available" registered for a shopping conversation are displayed together There are cases. In the present invention, when a sentence is registered, the user inputs a sentence group, and only the sentences registered in the same group are displayed.

[0007]

According to a first aspect of the present invention, there is provided an input means for inputting a free sentence in a first language which is a language to be translated, and a meaning representing a meaning of a component of the sentence in the first language. a string according to the first language is an element of the feature and the semantic feature is a semantic feature dictionary is registered in correspondence with each other, with reference to the semantic feature dictionary, free text input from said input means Analyzing means for extracting the semantic features from the first sentence ; an example sentence in the first language;
And translation of the example sentence in the language, a semantic feature string is a set of semantic features related to the example sentence, the scene classifying the sentences to the class of more statements be used in the same scene, associated with each other and bilingual database, the scene finger
And scene input means for inputting to the constant, by searching the semantic features of the translation the free text from the semantic feature string of each sentence registered in the database has been extracted by the analysis means, based on the search result A semantic search unit for selecting an example sentence having the same semantic feature; a scene to which the example sentence searched by the semantic search unit belongs; and a scene registered in the bilingual database according to a scene input from the scene input unit . and scene retrieval means for selecting a sentence belonging from sentences in the scene, the example sentence selected in the selected example sentence <br/> or scene retrieving means by the sense search means is the selected
Was in the translation apparatus and a display means for displaying a translation of the example sentence or selected example sentence, said input in a first language selected from free text and the sentences to obtain a translation in the second language After confirming the selection sentence , the user
Display of example sentences belonging to the previous scene by scene input means
Is specified, the scene search means
Was entered just before and confirmed as the selection sentence
Example sentences belonging to the same scene as the example sentences and / or their translations
The above problem is solved by enabling selection and display, thereby reducing the user's trouble of inputting and easily obtaining translations used before and after the translation.

[0008] invention of claim 2 is the invention of claim 1, similar the scene retrieving means, is inputted immediately before the
This scene that is finalized sentence as a selection statement belong there are multiple
Is detected , the scene input means allows the user to select a desired scene , and then displays an example sentence belonging to the scene and / or a translated sentence thereof .
The above object is achieved by reducing the user's trouble of inputting and easily obtaining a translation used in a selected scene .

[0009] The invention of claim 3 is the translation of claim 1 or 2
The apparatus further includes a bilingual registration unit that classifies and registers an example sentence displayed on the display unit and a translation thereof by the user specifying an arbitrary classification name, and a plurality of example sentences registered by the bilingual registration unit. comprises adding a registration translation selection means for selecting sentences specifying the category name for the genus of sentences desired, wishes the sentences belonging to the classification, in advance, the user responsibility
The above problem is solved by quickly obtaining a bilingual example sentence and its translation corresponding to the purpose and usage scene from the registered example sentences that have been classified and registered.

According to a fourth aspect of the present invention, there is provided a computer-readable storage medium storing a program having any one of the first to third functions.

[0011]

BRIEF DESCRIPTION OF THE DRAWINGS FIG .
A conversational sentence translator for translating a conversational sentence using such a translation device.
Will be described in detail with reference to an example . In the following example, Japanese is used as the first language and English is used as the second language. Obviously, the language is not limited to these languages. FIG. 2 shows the appearance of a conversational sentence translator in one embodiment of the translator according to the present invention. A large number of Japanese and English sentence examples are stored in pairs, and these can be displayed on the liquid crystal display 14. In addition, this device uses handwritten character recognition using a dedicated pen 15 as an interface. A display other than liquid crystal such as a CRT can be used, and other means such as voice or keyboard input can be used as input means.

FIG. 1 shows an outline of a circuit configuration of such a conversational sentence translator. The conversation translator has a CPU (Central Processing Unit) 1 for performing a central function of various controls. The CPU 1 is connected to each of the ROM 6, the working memory 5, the display driver 3, and the input recognition unit 4 through a bus 13 such as a data bus. RO
M6 is a large-capacity read-only memory that stores programs for performing various controls of the conversational sentence translator and a large number of phrases and the like. The working memory 5 is a random access memory for temporarily storing data required for executing the program. The display driver 3 is a driver for displaying a bilingual sentence or the like on the liquid crystal display 2. The input recognition unit 4 is a handwriting input board 1 shown in FIG.
6 recognizes the input data from the dedicated pen 15.

The ROM 6 stores a bilingual database 7 and a semantic feature dictionary 8 in addition to programs for performing various controls of the translation apparatus. For example, the bilingual database 7 stores addresses 0 to 50,000, and the semantic feature dictionary 8 stores addresses 50001 to 70000. The programs stored in the ROM 6 for performing various controls of the translation device are divided into an analysis unit 9, a search unit 10, a registration unit 11, and a control unit 12, depending on the nature of the operation executed by the CPU 1. The analysis unit 9 analyzes the input data recognized by the input recognition unit 4 . The search unit 10 searches the bilingual database 7 and the semantic feature dictionary 8 stored in the ROM 6. The registration unit 11 stores the displayed bilingual sentence in the working memory 5.
And the stored translation is displayed. The control unit 12 performs control other than the above such as control of screen display.

An embodiment of the data structure of the bilingual database 7 and the semantic feature dictionary 8 will be described below. These are common in claims 1 to 4. The bilingual database 7 defines a number of useful sentences that are desired to be translated and conveyed to local people when traveling abroad in the following form. Example 1 Please check in. Check in, please. Check-in request Hotel Example 2 Can I check-in now? Can I check in now?

Here, the first line of the sentence example is a Japanese sentence to be displayed.
(That is, the fixed phrase in the translated language, that is, the example sentence) , the second line is the translation, and the English sentence displayed as the translation result together with the Japanese sentence (that is, the translation of the example sentence in the translation language) ,
The third line is a set of semantic features associated with the bilingual sentence. This is called a semantic feature sequence. The fourth line is a conversation scene indicating a scene (scene) to which the sentence belongs. The semantic feature is a character string that becomes a keyword when searching for a sentence example, and is often assigned one to one segment of each Japanese sentence. In "Please check in" in sentence example 1, "Check in" and "Please" are the semantic features of this sentence. To be precise, these are the names of semantic features, and what character strings are defined as these semantic features are described in the semantic feature dictionary 8. The semantic feature dictionary 8 indicates the meaning of the components of the sentence in the language to be translated.
The name of to semantic feature, according to the translation language, which is a component of the Japanese character string (the semantic feature to be extracted as the semantic features
Character string) is defined. For example, "check in"
The semantic feature "please" is defined in the semantic feature dictionary 8 as follows.

[Semantic Feature Dictionary 8] Check-in Check-in Please thank you please thank you
Is this possible? This means that the Japanese character string on the right is defined as the semantic feature of the name on the left. For example, the character string "please" is defined as the semantic feature of the name "please". Hereinafter, the operation of this conversational sentence translator will be described. The following is a description of the processing common to the claims. Embodiments of the individual claims are described below.

In the following, the translated sentence, ie, the free sentence , input by the user is referred to as an input sentence, and the Japanese sentence , ie, the example sentence stored in the conversational sentence translator is referred to as a fixed sentence. When a pair of a fixed sentence and its translated English sentence is referred to, it is called a bilingual sentence. The outline of the basic operation of this conversational sentence translator is that a user inputs a translatable sentence, detects a semantic feature from the sentence, detects a fixed sentence from the semantic feature, and displays a bilingual sentence. . Also,
The user can display the translated sentence also by selecting a list of or select the desired scene That conversation scene, previously registered statement (that is, the translation and boilerplate). The operation of the conversation sentence translator will be described with reference to the flowchart of FIG. The description will be given in a form showing details of each step of the flowchart of FIG. The process of translating the sentence "Please check in" will be taken as an example (hereinafter, this sentence is referred to as input sentence 1).

<Step A1> The control unit 12 displays the “search method selection screen” of FIG. 4 on the liquid crystal display 14 through the display driver 3. <Step A2> The user selects “input Japanese sentence” on the handwriting input board 16 with the dedicated pen 15 on the search method selection screen of FIG. If the user has selected "Japanese sentence", the process proceeds to step A3. Otherwise, the process proceeds to step A13. <Step A3> The control unit 12 displays the “translated sentence input screen” of FIG. 5 on the liquid crystal display 14 through the display driver 3. Thereafter, the process proceeds to step A4. <Step A4> An input method by which the user inputs a free sentence in the translated language
In the translated sentence input screen shown in FIG. 5 that constitutes a column, "Please check-in" is handwritten input to the handwriting input board 16 with the dedicated pen 15. FIG. 6 is a diagram in the middle of the input, and FIG. 7 is a state in which the input is completed. The user sees FIG.
Is pressed, the screen of FIG. 8 is displayed,
This input sentence is recognized by the input recognition unit 4 and stored in the work memory 5. Thereafter, the process proceeds to step A5.

<Step A5> The analysis unit 9 executes step A5.
In step 5, a partial character string is created from the input sentence 1 stored in the working memory 5 and stored in the working memory 5. If there is a character string defined as a semantic feature in the semantic feature dictionary 8 among the partial character strings, the search unit 10 of the CPU 1 stores the character string and the name of the semantic feature in the working memory 5 as a pair. The partial character strings of the input sentence 1 are as follows. These are stored in the working memory 5. Search unit 10
Searches the semantic feature dictionary 8 stored in the ROM 6 using the partial character string of the input sentence 1 stored in the working memory 5 in step A6 as a key.

At this time, the partial character strings are searched in order from the front, and the order is stored, so that the order of appearance in the input sentence of each semantic feature is also stored. Hereinafter, the semantic feature dictionary 8 using the partial character string “request” of the input sentence 1 as a key
The process of searching for will be described. The search unit 10 reads the semantic feature dictionary 8 stored in the ROM 6 line by line from address 50001, line by line, and stores it in the work memory 5. Each row of the semantic feature dictionary 8 describes the name of each semantic feature on the left side, and a character string defined as the semantic feature on the right side.

It is assumed that the semantic feature dictionary 8 has the following lines. [Semantic feature dictionary 8] Please, please, please, since the character string "Please, please" is defined in the semantic feature dictionary 8, the search unit 10 determines the name of the semantic feature, ""And a partial character string" Please "of the input sentence are stored in the working memory 5 as a pair. In the semantic feature dictionary 8, the character string “please” is not defined in other places, so the search of the semantic feature dictionary 8 ends. The above operation is performed for all partial character strings of the input sentence 1. It is assumed that the following semantic features are detected from the input sentence 1. [Partial character string to the corresponding semantic features] "check-in" = "check Quinn,""please" = "thank you" here of = the left side of the is the name of semantic features, each right <br/> string sides means that was detected as meaning the identity of the left side name. Further, it is assumed that each semantic feature detected from the input sentence is detected in the order from top to bottom in the above correspondence table. This order is stored in the working memory 5. This concludes the description of step A5 according to the example.

<Step A6> If there is a character string defined as a semantic feature in the semantic feature dictionary 8 among the partial character strings, the search unit 10 of the CPU 1
The process proceeds to step A7, and if not, the process proceeds to step A17. In this example, since the semantic feature has been searched, the process proceeds to step A7. <Step A7> The search unit 10 providing the meaning search means searches the bilingual database 7 stored in the ROM 6 using the name of the semantic feature stored in the working memory 5 in step A5 as a key. Of the names of semantic features, bilingual database 7
Is stored in the working memory 5 together with the English sentence of the translated sentence, the semantic feature sequence, and the conversation scene. Taking input sentence 1 as an example,
The details of step A7 will be described. First, the search unit 10 searches the bilingual database 7 using one of the semantic feature sets stored in the working memory 5 in step A6 as a key. First, the bilingual database 7 is searched using the semantic feature “check-in” as a key.

Assume that the bilingual database 7 has the following description. [Bilingual translation database 7] Sentence example 1 Please check in. Check in, please. Check-in request Hotel Example 2 Can I check-in now? Can I check in now? Check-in is possible Hotel In sentence example 1 and example 2, "check-in" is included in the semantic feature column.
It is included in the semantic feature. Therefore, the search unit 10 sends the example sentences of the sentence example 1 and the sentence example 2 , that is, the fixed sentence, the translated sentence,
The gender and the scene are stored in the working memory 5. Search unit 1
0 indicates that the above operation is performed for all semantic features stored in the work memory 5. After the search for all semantic features is completed, Sentence Example 1 and Sentence Example 2 are stored in the working memory 5. This concludes the description of step A7 according to the example.

<Step A8> If any of the semantic feature names appear in the bilingual database 7, the process proceeds to step A9. If not, step A
Move to 17. In this example, since the fixed phrase has been searched, the process proceeds to step A9. <Step A9> The control unit 12 switches the liquid crystal display 14 from the screen of FIG. 8 to the bilingual sentence list display screen of FIG. 9 through the display driver 3. And a search unit that provides a meaning search means.
10 and the sentence example 1 and the sentence example 2 stored in the working memory 5 in the Japanese sentence (that is, the fixed sentence) and the English sentence (immediate sentence).
9 ) is displayed on the liquid crystal display 14 via the display driver 3 as a translation. Thereafter, the process proceeds to step A10. <Step A10> When the user selects a bilingual sentence that the user wants to use from the displayed bilingual sentences of sentence example 1 and sentence example 2, the control unit 12 changes the liquid crystal display 14 through the display driver 3 from the bilingual sentence list display screen of FIG. The screen is switched to the bilingual sentence enlarged display screen of FIG. The selected translation is enlarged and displayed on the liquid crystal display 14 through the display driver 3 as shown in FIG. Thereafter, the process proceeds to step A12.

<Step A11> The registration section 11 determines whether or not the user has pressed the registration button 18 in FIG. 10, and if so, proceeds to step A1.
Move to 2, if not pressed, end. <Step A12> The registration unit 11 selects the bilingual sentence of the displayed sentence example 1 as a selected sentence.
It is stored in the working memory 5 as. Then, the process ends.
This concludes the description of the operation of the conversational sentence translator using the input sentence 1 as an example. Next, FIG. 3 when a conversation scene is selected.
Steps in the flowchart will be described. <Step A13> The user selects “select from conversation scene” on the handwriting input board 16 with the dedicated pen 15 on the search method selection screen in FIG. If the user has selected "select from conversation scene", the process proceeds to step A14, and if "selected from the registered sentence", the process proceeds to step A16.

<Step A14> The control unit 12 displays the conversation scene selection screen of FIG. 12 on the liquid crystal display 14 through the display driver 3. Thereafter, the process proceeds to step A15. <Step A15> The user configures a scene input means for inputting a scene . In the conversation scene selection screen of FIG. 12, the icon of the hotel conversation scene is selected on the handwriting input board 16 by the dedicated pen 15. This icon is recognized by the input recognition unit 4, and a conversation scene of a hotel is stored in the work memory 5. Search unit 10 that provides scene search means
Extracts and selects a conversation scene of a hotel from the bilingual database 7. Stored in the bilingual database 7
Since the sentence examples 1 and 2 have a conversation scene of a hotel, the search unit 10 stores the sentence examples 1 and 2 in the work memory 5. <Step A16> registration section 11, in the example of the input sentence 1, the example sentence 1 stored in the working memory 5, i.e. as the selected data
The selected sentence is stored again in the working memory 5. This concludes the description of the operation of the conversational sentence translator using the input sentence 1 as an example. Next, steps in the flowchart of FIG. 3 which were not traced above will be described. If semantic feature or fixed phrase at step A6 or A 8 not found, the control unit 12 determines that the non translated, to that effect display. That process is step A17 below.

<Step A17> The control unit 12 switches the liquid crystal display 14 from the screen of FIG. 8 to the non-translatable screen of FIG. 11 through the display driver 3. The above is the common operation of the conversational sentence translator according to the present invention. An embodiment will be described below for each claim. In each embodiment, as in the above-described basic embodiment, a process of displaying a bilingual sentence for a certain input sentence will be described as an example. Hereinafter, embodiments of the conversation sentence translator according to claims 1 and 4 will be described. The process of translating an input sentence by the conversational sentence translating apparatus according to claims 1 and 4 is the same as that of the basic embodiment except for step A15. Therefore, only the difference in step A15 will be described in detail here. Since the operation flow is the same as that of the basic embodiment, the operation will be described with reference to the flowchart of FIG. In this claim, step A15 of the basic embodiment is replaced with the following steps B1 to B3 shown in FIG . Now, the following sentence examples 1 to 5 are registered in the bilingual database, and steps A3 to
It is assumed that the user inputs “Can you give me a little more” in A4, and that sentence example 3 is searched in steps A5 to A7.

Example 1 Please check in. Check in, please. Check-in request Hotel Example 2 Can you check-in now? Can I check in now? Check-in available hotel Example 3: Could you make it a little cheaper? More cheap or shopping Example 4: Please show another Could I have a look at some Please see another thing Shopping example 5: I will come again later.

[0029] <Step B1> Here, in a conversation scene selection screen users 12, "straight-through dedicated pen 15 on the handwriting input board 16
It is possible to select the icon of the conversation scene of the "previous scene", that is, the "previous scene" . This icon is recognized by the input recognition unit 4 and stored in the working memory 5.
A conversation scene "previous scene" is stored. If the selected scene is "the immediately preceding scene", go to step B2.
If not, the process proceeds to step B3. <Step B2> The search unit 10 which provides scene search means selects a conversation scene from the bilingual database 7 immediately before and registers it as a selected sentence.
Extract a scene that matches the recorded scene. In this case, since the conversation scene immediately before is the shopping of sentence example 3, sentence example 3, sentence example 4, and sentence example 5 are selected. Thus,
From a free sentence input in the first language that is the translated language, its second
After obtaining the bilingual sentence including the translation in the language as the selected sentence,
Belongs to the same scene as the selection sentence registered just before
Extracted from the bilingual database 7 and displayed.
Makes it easier to translate before and after the translation
Can be obtained.

<Step B3> The search unit 10 extracts from the bilingual database 7 those that match the pressed conversation scene. If the conversation in FIG.
Assuming that the conversation scene of “hotel” is selected on the site selection screen, the search unit 10
The conversation scene is extracted from hotels. Since sentence examples 1 and 2 have a conversation scene of a hotel, the search unit 10
Stores sentence example 1 and sentence example 2 in the working memory 5. Less than,
An embodiment of the conversational sentence translator according to claims 2 and 4 will be described. The process of translating an input sentence by the conversational sentence translating apparatus according to the second and fourth aspects is the same as that of the basic embodiment except for step A15. Therefore, here, step A
Only the 15 differences will be described in detail. Since the operation flow is the same as that of the basic embodiment, the operation will be described with reference to the flowchart of FIG. In the present invention, step A15 of the basic embodiment is replaced with steps C1 to C1 shown in FIG.
Replace with 5. Now, sentences 1 to 5 are stored in the bilingual database.
In addition to the above, sentence example 6, sentence example 7, and sentence example 8 illustrated below are registered. In steps A3 to A4, the user inputs "pay with a credit card", and in steps A5 to A7, sentence example 6 is searched. Suppose it was done.

Example 6 I want to pay with a credit card. I'd like to pay by credit card. I want to pay by credit card. Meal Shopping Ref. 7 Let's split the cost. Let's split the cost. Meal Ref. 8: I would like to pay in installments. I'd likt to pay in installments.

<Step C1> Here, the user enters the dedicated pen 1 on the handwriting input board 16 on the conversation scene selection screen shown in FIG.
5, it is possible to select the icon of the conversation scene of "the immediately preceding scene", that is, "the immediately preceding scene" .
This icon is recognized by the input recognition unit 4, and a conversation scene called “immediately before scene ” is stored in the work memory 5. If the selected scene is the "previous scene ", the process proceeds to step C2; otherwise, the process proceeds to step C3. <Step C2> When the conversation scene of the immediately preceding sentence is determined in one way, the search unit 10 providing the scene search means proceeds to step C4, otherwise proceeds to step C5. In this case, it means test
Searched by the search unit 10 as a search means, and selected immediately before
The scene of sentence example 6 registered as the selected sentence is a shopping or a meal, and the process goes to step C5 because it is not determined in any way. <Step C3> On the conversation scene selection screen in FIG.
If any other than the “previous scene” is selected, the search unit 10 extracts from the bilingual database 7 those that match the conversation scene selected by the user . If the conversation scene of “hotel” is selected, the search unit 10 extracts, from the bilingual database 7, a conversation scene whose hotel is a hotel. Since Sentence Example 1 and Sentence Example 2 have a conversation scene of a hotel, the search unit 10 stores Sentence Example 1 and Sentence Example 2 in the work memory 5. <Step C4> The search unit 10 extracts, from the bilingual database 7, a conversation scene that matches the scene selected immediately before. Later
As shown in step C5, in this case, since shopping is selected as the immediately preceding conversation scene, sentence example 3, sentence example 4, and sentence example 5 are selected. Then, search just before
If there are multiple scenes to which the sample sentence belongs,
After selecting the desired scene, the sentences belonging to that scene
By displaying, the translation used in the selected scene
Translated sentences can be easily obtained.

<Step C5> The control section 12 takes out shopping and meals from the working memory 5 as the immediately preceding conversation scene, and sets the scene input means.
Only the menus of these two scenes are enabled on the provided conversation scene selection screen of FIG. The user selects an icon of a shopping conversation scene on the handwriting input board 16 with the dedicated pen 15 on the conversation scene selection screen of FIG. This icon is recognized by the input recognition unit 4, and a conversation scene “shopping” is stored in the work memory 5. After that, it moves to step C4. Hereinafter, embodiments of the conversation sentence translator according to claims 3 and 4 will be described. The process of translating an input sentence by the conversational sentence translating apparatus according to claim 3 and 4 comprises steps A12 and A16.
It is the same as the basic embodiment except for. Therefore, only the differences between step A12 and step A16 will be described in detail here. Since the operation flow is the same as that of the basic embodiment, the operation will be described with reference to the flowchart of FIG. In this claim, step A12 of the basic embodiment is replaced with steps D1 and D2 as follows. In addition, step A16 of the basic embodiment is replaced with the following steps D3 to D5 shown in FIG .

Now, sentence examples 1 to 8 are stored in the bilingual database.
In addition to, paragraph 9 illustrated below, and example sentences 10 is registered, in step A3~A4 user inputs a "do single room how much", step A5~A
In step 7, sentence example 9 is searched, and steps A9 to A11 are searched.
Suppose that the user presses the registration button 18. Example 9 How much is a single room per night? How much per night is a single room? Single room How much per night Hotel Example 10 Is this the only product? Is this everything you have?

<Step D1> The registration unit 11 selects the bilingual sentence of the displayed sentence example 9 as a selected sentence.
It is stored in the working memory 5 as. Step D1 is the same as step A12. <Step D2> control unit 12, the liquid crystal display 14 from the translation enlarged display screen in FIG. 10 through the display driver 3, sentences and
A translation registration means for classifying and registering the translation
To the conversation scene name input screen of FIG. Users in the talking scene name input screen of Fig. 13, a conversation by a dedicated pen 15 on the handwriting input board 16 scene name That Noboru
Enter the classification name for the example sentence to be recorded and its translation .
The input conversation scene name, that is, the classification name is recognized by the input recognition unit 4 and stored in the work memory 5 .
The example sentence and its translation (in the above example,
Is registered as sentence example 9) . <Step D3> Here, on the search method selection screen of FIG.
When “select” is selected, the control unit 12 sets the display driver 3
The liquid crystal display 14 is switched from the search method selection screen of FIG. 4 to the conversation scene selection screen of FIG.

<Step D4> On the conversation scene selection screen shown in FIG. 14, which constitutes the registered bilingual selection means, the user previously entered the handwritten input board 16 with the dedicated pen 15 as the classification name.
To select the icon of "at Le. This icon is recognized by the input recognition unit 4 and stored in the working memory 5 as “ Hotel ”
At the conversation scene, that is, a classification name . The search unit 10 reads the conversation scene, that is, the classification from the bilingual database 7.
The one whose name is " at a hotel " is extracted. <Step D5> The registration unit 11 replaces the sentence example 9 stored in the working memory 5 with
It is stored again in the working memory 5 as the selected data, that is, as a selected sentence . This is the same as step A16. Thus, the example sentences and their translations are classified and registered.
Then, from the registered multiple example sentences,
Class, and select the desired example sentence from the example sentences belonging to the classification.
By selecting, the user can arbitrarily classify and register in advance
From the example sentence, a translation example according to the purpose and use scene
Sentences and their translations can be obtained quickly.

[0037]

According to the first and fourth aspects of the present invention, after obtaining a translated sentence in the second language from a free sentence input in the first language, an example sentence belonging to the same scene as the sentence input immediately before
By displaying the (fixed text) , it is possible to easily obtain a translation used before and after the translation.

[0038] In the invention according to claims 2 and 4, if the search <br/> search by example sentence just before there are multiple scenes belong, were allowed to select the desired scene to the user, display the sentence belongs to that scene By doing so, it is possible to easily obtain a translated sentence used in the selected scene .

[0039] In the invention according to claim 3 and 4, and registers to classify the sentence and its translation, a plurality of example sentences or <br/> we registered, specifying a classification belongs sentences desired, the classification By selecting the desired example sentence from the example sentences belonging to
From the example sentences that the user has arbitrarily classified and registered, it is possible to quickly obtain a bilingual example sentence corresponding to the purpose and usage scene and a translation thereof.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a translation apparatus according to an embodiment of the present invention;
It is a block diagram of a speech translator .

FIG. 2 is a diagram illustrating an example of a translation apparatus according to an embodiment of the present invention;
FIG. 3 is an external view of a speech translator .

FIG. 3 is a flowchart of an embodiment of a translation process according to the present invention.

FIG. 4 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 5 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 6 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 7 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 8 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 9 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 10 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 11 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 12 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 13 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 14 is a flowchart of an embodiment of a translation process according to the present invention.

FIG. 15 is a flowchart of an embodiment of a translation process according to the present invention.

FIG. 16 is a flowchart of an embodiment of a translation process according to the present invention.

FIG. 17 is a flowchart of an embodiment of a translation process according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... CPU, 2 ... Liquid crystal display, 3 ... Display driver, 4 ... Input recognition part, 5 ... Working memory, 6 ... ROM,
7 ... bilingual database, 8 ... semantic feature dictionary, 9 ... analysis unit, 10 ... search unit, 11 ... registration unit, 12 ... control unit, 13
... bus, 14 ... liquid crystal display, 15 ... dedicated pen, 1
6 ... handwriting recognition board, 17 ... interpreter icon, 18 ... registration button.

Claims (4)

(57) [Claims] 1. A free language in a first language which is a language to be translated.
Input means for sentence is input, a character string according to the semantic feature and the first language is an element of the semantic feature representing the meaning of the components of the sentence by the first language has been registered in correspondence with each other A semantic feature dictionary, analysis means for extracting the semantic feature from the free sentence input from the input means with reference to the semantic feature dictionary, and an example sentence in the first language;
A translation of the example sentence in a second language that is a translation language, a semantic feature sequence that is a set of semantic features related to the example sentence ,
A scene classifying sentences to the class of more statements be used in the same scene, a translation database associated with each other, the scene input manually to enter to specify the scene
Stage and, from said semantic feature string of each sentence that is registered in the bilingual database of the free sentence extracted by the analysis means
The semantic features search and a-out based on the search results, the same meaning
And means searching means for selecting an example sentence having the feature scene and said meaning retrieved retrieved in unit sentence belongs, the Sea
By input from down input means scene, from said each example sentence registered in the bilingual database and scene retrieval means for selecting a sentence belonging to the scene, are selected in the example sentence or scene retrieval means selected by said semantic retrieving means Was
In the translation apparatus, comprising: a sentence, and display means for displaying a translation of the selected example sentence or selected example sentence, it is selected from the free text input in the first language that
Obtain an example sentence and a translated sentence in the second language and confirm it as a selected sentence
After, the previous scene by the user the scene input means
If you specify the display of example sentences belonging to
In the scene search means, the selection input immediately before
An example sentence belonging to the same scene as the example sentence confirmed as an alternative sentence and
/ Or translation system, wherein the <br/> this making it possible to display the selected translation thereof. 2. A translation apparatus according to claim 1, similar the scene searching means is input immediately before the selected sentence
Biopsy that confirmed the sentence there are multiple belongs scene as
If issued, by the scene input means, since by selecting the desired scene to the user, an example belonging to the scene
Sentence and / or transliteration which is characterized that you display the translated sentence
Translation device . To 3. A translation apparatus according to claim 1 or 2, further bilingual registration the user registers classifies its translation and example sentence displayed in the display unit by specifying an arbitrary classification name means, a plurality of example sentences or <br/> we registered in the translation registering means, specifying the category name for the genus of sentences desired, a registration translation selection means for selecting sentence desired from sentences belonging to the classification Translation equipment characterized by being additionally provided
Place . 4. A computer-readable storage medium storing a program having the function according to claim 1. Description: