JP2003345794A - Electronic translation device - Google Patents

Abstract

(57) [Summary] The present invention provides an electronic translator capable of having a smooth conversation in a dialogue scene by displaying a reply sentence to a question sentence input by voice by an easy operation. That is the task. SOLUTION: An input section for inputting a question sentence, a main first word group used for the question sentence and a second word group to be used for a sentence of an answer corresponding to the question sentence are stored in pairs. The answer information storage unit and the question sentence input by the input unit are searched for whether or not the first word group is included. If the first word group is included, the first word group is paired with the first sentence. The second
An answer candidate extracting unit that extracts a word group from the answer information storage unit, and an answer creating unit that creates a sentence of an answer to the input question sentence using words in the extracted second word group. It is characterized by the following.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic translator having voice recognition and dictionary search functions.

[0002]

2. Description of the Related Art Conventionally, various information devices having a voice recognition function or electronic devices having a dictionary search and translation function have been developed. For example, there are the following as conventional electronic translation devices and electronic dictionaries. JP-A-4-3976
No. 9 discloses words in the input document as upper and lower relations,
There is disclosed a thesaurus generation device that generates a thesaurus having a single tree structure by dividing the words into groups of meanings such as synonymous relationships or whole partial relationships.

Further, Japanese Patent Laid-Open Publication No. Hei 6-162072 discloses an electronic dictionary that uses an inter-word relation table to search for “opposite relation words” having different viewpoints without changing the meaning relation. Japanese Unexamined Patent Publication No. Hei 8-185411 discloses a method of recognizing one of input words, concepts or meanings using a knowledge base on the concept, the meaning of the words, and the grammar of the words. There is disclosed an electronic dictionary system for searching, in which a plurality of independent dictionaries are to be integrated and used.

[0004]

However, these electronic dictionaries and the like are useful for creating texts and translating texts input to a computer or texts obtained through the Internet or the like. , A function that quickly translates and presents a sentence spoken by the other party, a function that generates an answer sentence corresponding to the sentence of the other party, generates antonyms and synonyms that match the expression of sentences used in conversation Since it does not have the function, it cannot be used in a situation where a person who speaks different languages has a direct conversation.

[0005] Further, in the above-mentioned conventional technology, when talking with a foreigner who speaks another language by itself, the words spoken by the other party are smoothly and accurately translated into the native language and understood. You can write your reply smoothly,
It is not possible to translate and translate sentences in their native language into the language of the partner country smoothly and accurately, so in order to communicate including the above work, an interpreter must be in between. It was very difficult for people who could not speak a foreign language to communicate with foreigners.

Therefore, the present invention has been made in view of the above circumstances, and displays a reply sentence to a question sentence input by voice by an easy operation.
An object of the present invention is to provide an electronic translator capable of having a smooth conversation with a person who speaks a language that cannot be understood without using an interpreter.

[0007]

According to the present invention, an input section for inputting a question sentence, a main first word group used in the question sentence, and a first word group to be used for a sentence of an answer corresponding to the question sentence. A search is made as to whether or not the first word group is included in the answer information storage unit storing the two word groups as a pair and the question sentence input by the input unit. An answer candidate extracting unit for extracting a second word group paired with the first word group from the answer information storage unit, and an answer to the input question sentence using the words in the extracted second word group. An electronic translation device comprising an answer creation unit for creating a sentence. According to the present invention, in a dialogue with a person who speaks a different language, it is possible to quickly and accurately present a text of an answer to a question sentence,
Conversation can be smooth.

[0008] Also, a translation unit for translating the input question sentence and a sentence of the answer to the question sentence into a predetermined language, and at least one of the question sentence, the answer sentence and the sentence translated by the translation unit are displayed. The display unit may be further provided.

Further, an antonym dictionary storing a pair of a predetermined word and a word having the opposite meaning of the word, and storing a pair of a predetermined word and a word having a synonymous meaning of the word. A synonym dictionary and at least one related word dictionary stored in association with each other, including predetermined words and words having the opposite and synonymous meanings, and Select any word,
An instruction unit for instructing to search for any of the opposite word, synonym or related word of the selected word, and an opposite word, synonym or related word instructed for the word selected by the instruction unit into the corresponding dictionary. A search unit for referring to and searching, and a change unit for replacing the selected word with an antonym, a synonym or a related word searched by the search unit, and changing a sentence displayed on the display unit in consideration of grammar. May be further provided. According to this, the sentence displayed on the display unit can be easily changed to a sentence having an opposite meaning or a sentence having a similar meaning, and a sentence with more variations can be presented. , Can make the conversation smoother.

If the search unit finds a plurality of opposite words, synonyms or related words by searching, the display unit displays the plurality of opposite words, synonyms or related words in a list, and the change unit However, the word selected by the instruction unit may be replaced with a word specified by a predetermined instruction input among the opposite words, synonyms, or related words displayed in the list. According to this, the sentence intended by the user can be promptly displayed.

Further, a second keyword inputting an arbitrary keyword is performed.
An input unit is further provided, and when there are a plurality of second word groups extracted by the answer candidate extracting unit, the answer creating unit uses the second word group including the keyword input by the second input unit to input an answer. A sentence may be created.

[0012] The information processing apparatus may further include an audio output unit that outputs information displayed on the display unit by audio. According to this, the sentence of the answer can be heard not only by display but also by voice, so that the conversation can be performed more smoothly.

Further, the invention provides a step of inputting a question sentence, and a main first word group used in the question sentence and a second word group to be used in a sentence of an answer corresponding to the question sentence. A step of searching whether or not the input question sentence includes the first word group by using the answer information storage unit stored as a pair;
Extracting a second word group paired with the first word group from the answer information storage unit, and using the words in the extracted second word group to create a sentence of an answer to the input question sentence And a method for creating a sentence in an electronic translation device.

In the present invention, a keyboard, a mouse, a pen, a tablet, and other pointing devices can be used as the input unit and the instruction unit. As the answer information storage unit and each dictionary, in addition to a nonvolatile semiconductor memory such as a ROM, any storage medium such as a rewritable semiconductor memory such as a RAM and a flash memory, a memory card, a hard disk, and a CD-ROM can be used. . Each functional block such as an answer candidate extraction unit, an answer creation unit, a translation unit, a search unit, and a change unit includes a CPU, a ROM,
It can be realized using a microcomputer including a RAM, an I / O controller, a timer, and the like, and the function of each unit can be realized by operating each hardware based on a program stored in a ROM, a RAM, or the like. Since the device of the present invention preferably has portability, it is preferable to use a small display such as an LCD or an EL for the display unit.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS <Configuration of Electronic Translation Apparatus> FIG.
1 shows a schematic configuration diagram of an embodiment of an electronic translation device of the present invention. Here, it is assumed that translation between Japanese and English is performed, and a speech recognition function for recognizing English speech and converting it into character data is provided. However, the language to be translated is
The present invention is not limited to these two languages, and may be any existing language such as German, French, and Chinese.

In FIG. 1, reference numeral 11a denotes a cabinet, 11b denotes a display unit, 11s denotes a microphone, 11r.
A speaker is provided with a large number of keys for setting and inputting various functions and characters, similarly to a conventional electronic dictionary.
For example, reference numeral 11d is a key for selecting an English-Japanese dictionary for converting English to Japanese, reference numeral 11e is a key for selecting a Japanese-English dictionary for converting Japanese to English, 11f is a key for executing speech recognition, and 11q is a key for example sentences. This is a sounding key for sounding from a speaker.

FIG. 2 is a schematic block diagram showing an embodiment of the electronic translation apparatus according to the present invention. As shown in FIG. 2, this device mainly includes a display unit 12a, a key input unit 12
c, voice input unit 12d, speaker 12k (11 in FIG. 1).
r) and a control unit 12b.
Are a display control unit 12e for controlling the display unit 12a, an audio control unit 12j for controlling the speaker 12k, and two input units 1
An input control unit 12g for controlling the devices 2c and 12d, a CPU 12f for controlling the entire device, a ROM 12h for storing programs and data, and a RAM 12i.

As the display section 12a, an EL element or the like can be used in addition to the LCD. As the input key section 12c, a keyboard, a numeric keypad, or the like can be used in addition to the keys shown in FIG. The voice input unit 12d corresponds to the microphone 11s in FIG.

The ROM 12h stores a program for causing the CPU to execute the functions of the apparatus and information necessary for executing functions such as translation and speech recognition (speech recognition table 15b, dictionary 15a as shown in FIG. 3). , A grammar table 15d, and an answer correspondence table 15e) are stored in a readable manner. The dictionary 15a stores information included in so-called ordinary dictionaries, such as meanings, variations, pronunciation symbols, example sentences, and explanations of Japanese and English words.

FIG. 4 shows an embodiment of the grammar table 15d. This grammar table stores so-called grammatical rules including the order of words and idioms in Japanese and English, morphologies, and the like.

<Outline of Method of Using Electronic Translation Apparatus of the Present Invention> Next, an outline of a method of using the electronic translation apparatus of the present invention and an English-to-Japanese translation process in a conversation will be described. A user whose native language is Japanese (Japanese) has the electronic translation device of the present invention,
Consider a conversation with an English-speaking foreigner.

First, the user presses the "English / Japanese" key 11d to put the electronic translator into a standby state for English voice recognition. Next, the microphone 11s is brought closer to the foreigner, and the foreigner utters an English sentence, and the "speech recognition" key 11f is kept depressed during the pronunciation.
For example, if a foreigner utters "what time is it now", this sound is input to the microphone.
The input voice is recognized as each English word by voice recognition.

Next, the recognized English sentence and its Japanese translation are displayed on the display unit 12a (see FIG. 5A). In this display state, the user presses the “answer” key 1
When 1k is pressed, if the displayed sentence is determined to be a question sentence, a candidate for a sentence to be the answer is selected and displayed in English and Japanese (see FIG. 5 (b)). .
At this time, candidates for the answer sentence are selected with reference to an answer correspondence table (15e in FIG. 3) as shown in FIG. 6, and assembled into a sentence matching the grammar using the grammar table in FIG. Can be

FIG. 5B shows "what time is it now".
? An example is shown in which, after an input of a question sentence “”, the display of “It's 10:00 AM (10:00 in the morning)” is displayed as the answer sentence after the above processing. FIG.
In the answer correspondence table 15e, “question” and “answer” are stored as a pair. For example, “what tim”
"e" and "10:00 AM" are stored as a pair. Here, the item content of “question” is composed of a word group including several keywords that are expected to appear in the input question sentence. "Answer" is also made up of a group of words consisting of several words that are central in the answer sentence. For example, in FIG. 6, when “where” and “go” are included in the input question sentence, it means that a sentence including “cinema” should be created as the answer.

FIG. 7 shows the answer of this question "where + go"
This shows a display example in which the correspondence of "inema" is selected. The input sentence “where are you going” shown in FIG.
? ”Includes“ where ”and“ go ”, so the English sentence“ I'm going to the cinema ”containing“ cinema ”as shown in FIG. It is displayed with its Japanese translation.

In FIG. 6, "How"
When a sentence containing “get to” and “get to” is input, the answer means that there are two candidates. In other words, the answer sentence is "get the subway" or
A sentence containing "go straight for" is a candidate. FIG. 8 shows a display example corresponding to this case.

The input sentence "How do I get to th" shown in FIG.
e museum? "And" How "and" get to "
Therefore, referring to the answer correspondence table 15e, two kinds of answer sentences as shown in FIG. 8B are displayed. That is, "Get the subway to th
e museum station. "and" Go straight for
The English sentence "two blocks and turn right."

<Embodiment 1> The processing contents of Embodiment 1 of the electronic translation apparatus of the present invention will be described. FIG. 9 shows an operation flowchart of a part of the recognition processing of the input utterance in the first embodiment of the present invention. Here, it is assumed that the owner of the electronic translation apparatus of the present invention is a Japanese speaker who speaks Japanese, and a conversation partner is a person who speaks English, and performs mutual language translation between Japanese and English.

First, when the user who owns the electronic translation apparatus of the present invention presses the "English-Japanese" key 11d, the apparatus enters a state of waiting for an English voice input. Next, when the "voice recognition" key 11f shown in FIG. 1 is pressed, the apparatus starts voice input. Then, the other party points the microphone 12d and utters an English question sentence (step 21a). At this time, the user keeps pressing the "voice recognition" key 11f while the pronunciation continues. The sound input to the microphone 12d is converted into an analog electric signal, converted from analog to digital, and temporarily stored as digital signal data in the RAM (step 21).
b). Then, the CPU 12f performs voice recognition processing on the digitized question sentence using the voice recognition table 15b, converts it into characters and words (step 21c), and temporarily stores it in the RAM in the form of the spoken language (step 21c). 21d).

Next, translation processing is started for the input English sentence (step 21e). That is, CP
U12f refers to the English-Japanese dictionary 15a, converts English words into Japanese, and creates a Japanese sentence whose meaning can be understood using the grammar table 15d. Then, the CPU 12f
The English sentence of the input question sentence as the recognition result and the Japanese sentence as the translated sentence are displayed on the display unit 12a (step 21).
f). For example, if a question sentence “what time is it now” is uttered, after the above processing, FIG.
Is displayed.

Next, a description will be given of the answer display processing from this state to the display of the sentence of the answer to the question sentence. FIG. 10 shows a flowchart of the answer display process of the first embodiment. First, in a state where the question sentence as shown in FIG. 5A is displayed, the CPU confirms that the user has pressed the "answer" key 11k in FIG. 1 (step 3).
3a) The English sentence of the question sentence stored in the RAM 12i is divided into words (step 33b).

Next, the CPU extracts a key word or idiom from the divided words (step 3).
3c). At this time, a key word or idiom can be extracted using, for example, usage frequency data. The extracted key words or idioms are temporarily stored in the RAM 12i (step 33d). In the example of FIG. 5A, “wha
"t" and "time" are extracted. Next, it is checked whether the extracted key word exists in the "question" item of the answer correspondence table 15e, and if there is a key word, an "answer" portion corresponding to the key word is extracted (step 33).
e). In the example of FIG. 5A, “10:00 AM” is extracted corresponding to “what” + “time”.

Next, the CPU rearranges the extracted word group of the "answer" part by supplementing necessary words so as to be a grammatically meaningful sentence using the grammar table 15d, Is created (step 33f). For example, in the example of FIG. 5A, an English sentence “It's 10:00 AM” is created. The CPU transfers the created text to the RAM
12i (Step 33g), and the display unit 11b
(Step 33h).

Further, referring to the English-Japanese dictionary 15a, a Japanese translation corresponding to the created English sentence is created and displayed on the display unit 11b (step 33i). For example, FIG.
As shown in (b), the English sentence of the "answer" and the Japanese translation are displayed. In this way, even if a question is issued in English or in another language that is incomprehensible to the user, the text of the answer to the question can be sent to the user by simply operating a few dedicated keys. Can be presented.

The operation performed by the user is "voice recognition".
This is a simple operation of pressing the key 11f, the "English-Japanese" key 11d, and the "answer" key 11k.
Since the translation process, the answer search, and the display process are also executed in a short time by the microcomputer centering on the CPU, a smooth conversation can be performed in a dialogue scene with the user who speaks a different language. In the first embodiment, the input sentence is created by recognizing the input speech. However, even if the input sentence is created by key input of characters using a device having no speech recognition processing, The creation and display processing of the answer text as in the embodiment can be used.

<Embodiment 2> Here, an embodiment will be described in which a sentence having the opposite meaning is created and displayed by an easy operation. FIG. 11 is an explanatory diagram of one embodiment of the antonym dictionary of the present invention. The antonym dictionary includes a headword 53a,
And the opposite word 53b corresponding to the headword. As shown in FIG. 11, for the headword "good",
There are four antonyms 53b, but in general there may be multiple antonyms for one headword. In this embodiment, the user selects the "Reverse" key 11 shown in FIG.
By pressing o, the opposite word of the selected word is displayed, and a translation having the opposite meaning is displayed.

FIG. 12 is an explanatory diagram of a display example according to the second embodiment of the present invention. FIG. 12A shows a state in which an input English sentence is speech-recognized and a sentence of an answer to the English sentence is displayed on a display unit. Here, "He is good
at math. "is displayed. In this state, when the user presses the "jump" key 11p in FIG. 1, a cursor is sequentially displayed for each English word. "Jump" key 11
Each time p is pressed, the cursor moves to "He", "is", "good", and FIG. 12A shows a case where the cursor 51c is at "good". In addition, the character 51d of "anti" is displayed at the lower right of the display unit, and this display means that the currently selected word "good" has an opposite word.

In the display state of FIG.
Is pressed, the opposite word of the selected word is searched, and a sentence having a meaning opposite to that of FIG. 12A is displayed. FIG. 12B shows an English sentence and a Japanese translated sentence having opposite meanings. That is, the opposite word “poor” 51e to “good” is searched, and “poor” is translated as “poor” and displayed on the display unit 11b. At this time, as the opposite words to “good”, four are registered in the dictionary as shown in FIG. 11, but the meaning of the displayed sentence was reversed by comparing the meaning with the words before and after. The opposite can be selected as appropriate for the case.

FIG. 13 shows a flowchart of the opposite word search and display process according to the third embodiment of the present invention. If the user presses the "jump" key 11p while the text shown in FIG. 12A is displayed on the display unit (step 5).
2a) A cursor display 51c is displayed on the first word of the display unit. By pressing the "jump" key 11p several times, the cursor is moved to a word having an opposite word (step 52b). At this time, the CPU searches the word displayed by the cursor to see if it is a word registered in the antonym dictionary, and if it is registered, displays the character “anti” in the lower right of the display unit I do. When a word for displaying the opposite word is selected by the cursor display, when the user presses the opposite key 11o (step 52c), the following opposite word search and display processing are executed.

First, the CPU stores words that are candidates for the opposite word of the word currently displayed by the cursor in the opposite word dictionary 15.
Extract from a (step 52d). For example, FIG.
In the case of (a), for the selected word “good”,
Four candidates are extracted from FIG. The extracted words are temporarily stored in the RAM 12i (step 52).
e).

Next, referring to the grammar table, an antonym matching the meaning of the currently displayed sentence is selected (step 52f). Here, it can be determined whether or not the opposite word matches the currently displayed text by using, for example, an idiom of a grammar table, a correlation list between words, or the like. In the case of the text in FIG. 12A, “good” does not mean good, but “good”
Therefore, the word “poor” having a meaning opposite to the meaning of “good” is selected from the candidates in FIG.

Next, referring to the grammar table and the original sentence of FIG. 12A, other words necessary for creating the sentence are extracted from the English-Japanese dictionary (step 52).
g). In the case of FIG. 12A, since other words can be used as they are, there are no newly extracted words or changed words. For example, if it is necessary to change a preposition or a verb by using an opposite word, it is changed in this step 52g if necessary.

Next, the extracted opposite words and other words are rearranged so as to be a grammatically correct sentence, a sentence used as the opposite word is created and temporarily stored in the RAM (step 52h). Then, a sentence using the opposite word is displayed on the display unit (step 52i). Further, for a sentence using the opposite word, a Japanese translation is created with reference to the English-Japanese dictionary and displayed on the display unit (step 52j). For example, a sentence as shown in FIG. 12B is displayed on the display unit. According to this embodiment, in a state where the text of the answer is displayed, the user presses the jump key 11p and the opposite key 11p.
By simply performing an easy operation of pressing o, a sentence having a meaning opposite to the displayed answer sentence can be displayed.

<Embodiment 3> Here, an embodiment will be described in which a sentence using a synonym is created and displayed by an easy operation. FIG. 14 is an explanatory diagram of one embodiment of the synonym dictionary of the present invention. The synonym dictionary includes a headword 62a,
And a synonym 62b corresponding to the headword. For example, in FIG. 14, four synonyms are registered for “often”. FIG. 14 includes a synonym such as “sometimes” for “often” and a word indicating the opposite meaning such as “never”. The words need not overlap with the antonym dictionary. However, from the viewpoint that an example sentence rich in more variations can be created, it is preferable that the dictionary be configured to include not only synonyms but also opposite words. Antonym dictionary consisting of only antonyms,
Instead of separately providing synonym dictionaries consisting of only synonyms, a dictionary including both opposite words and synonyms, that is, a related word dictionary may be used.

In this embodiment, when the user presses the "synonym" key 11n shown in FIG. 1, a sentence using a synonym of the selected word is displayed, and a translation of the sentence using the synonym is displayed. It is characterized by being displayed. FIG. 15 is an explanatory diagram of a display example according to the third embodiment of the present invention. FIG.
(A) shows a state in which the sentence of the answer is displayed.
Here, "I often go to the cinema." Is displayed. In this state, when the user presses the "jump" key 11p in FIG. 1, a cursor 61b is displayed for each English word. FIG. 15A shows a case where the cursor is displayed at “often”. At this time, "ofte
Since “n” is a word in which synonyms are registered, characters of “kin” are displayed at the lower right of the display unit. If an opposite word is also registered, the character "anti" is also displayed.

When the user presses the "synonymous" key 11n in FIG. 1 in the display state of FIG. 15A, a synonym of the word "often" selected from the synonym dictionary is searched, and the synonym is used. The sentence is displayed. FIG. 15B shows a sentence (English and Japanese translations) using two synonyms. That is, a synonym for "often" is "s
ometimes "61f and" rarely "61g are searched,
It is what was displayed. Below the English text, the corresponding Japanese translation is displayed. Where "rarely" is
It is not a synonym of often but an antonym, but as mentioned above, including the antonym and synonym, by registering it as a related word whose meaning is related, it is an easy operation, a sentence with more variations Is preferably displayed.

FIG. 15C shows "have never" and "never" as related words of "often" shown in FIG.
Shows an example sentence displayed using. The synonym search and display processing of the third embodiment is the same as that of the second embodiment shown in FIG.
Is performed in substantially the same manner as the flow of the opposite word search and display processing. That is, the "synonyms" key is used instead of the "opposite" key, and the "synonyms dictionary"
Is used.

According to the third embodiment, a sentence having a similar meaning to the displayed sentence can be displayed simply by the user's easy operation. Also, by relating words that are related in some sense to a word, including synonyms and antonyms,
It may be registered in one related word dictionary. In this case, the character “Seki” may be displayed at the lower right of the display unit in the sense that “Related word” is present in the related word dictionary.

<Embodiment 4> Here, an embodiment will be described in which the dictionary is switched, or the display screen is switched to a search display screen while the display contents shown in FIG. 5 are left. For example, if you want to search for a word while saving the text of the answer, temporarily change the display screen from the text display screen (main screen) to the search screen (sub screen).
It is necessary to switch to In this case, the "switch" in FIG.
The key 111 is used.

When there are a plurality of dictionaries, a "sub" key 11i for switching the dictionary to be used may be provided. For example, the "sub" key is used when switching between an English-Japanese dictionary that is a main dictionary that is normally used and a Japanese-English dictionary that is used as a sub-dictionary. Further, a "return" key 11j for deleting the content displayed on the sub screen and returning to the main screen from the sub screen may be provided. Further, a "search" key 11g used when entering a process such as a word search using the main dictionary may be newly provided.

For example, when the "sub" key 11j is pressed while the display as shown in FIG.
The display is switched to that shown in FIG. 16 while the display contents of (a) are stored in the RAM. That is, a sub-screen for search is displayed so that search processing using the sub-dictionary can be performed.

In the state where the sub-screen shown in FIG. 16 is displayed, for example, by inputting an English word / idiom such as “time”, a Japanese translation can be searched for. Further, during the search on the sub-screen, the user can return to the main screen by pressing the "switch" key 111 and switch to the sub-screen by pressing the "switch" key 111 again. In addition, with the “question” as shown in FIG. 15A displayed, before pressing the “answer” key,
By pressing the "sub" key, "answer" search and display on the sub screen may be performed.

The "switch" key 111 only switches the display screen while maintaining the states of the main screen and the sub-screen, but if the "return" key 11j is pressed while the sub-screen is displayed, All information on the sub screen can be deleted, leaving only the main screen.

When the "sub" key 11i is pressed while the sub screen is displayed, the information on the main screen is erased, and the information on the sub screen is temporarily stored as the information on the main screen. May be continued. At this time, if the “sub” key 11i is further pressed, a new sub screen can be displayed and a new search can be performed. further,
When the "search" key 11g is pressed while either the main screen or the sub screen is displayed, information on both screens may be deleted, and a new search may be performed on the main screen.

<Embodiment 5> Here, FIG.
A description will be given of an embodiment in which all synonym or related word candidates are displayed in a list before displaying a synonym or related word sentence as shown in FIG. FIG. 17 shows an embodiment of a display screen displaying a list of related words of "often" on the display unit. This includes synonyms and opposites as shown in FIG.

In the state where the answer of FIG. 15A is displayed, the user selects “often” and presses the “synonyms” key 11.
When n is pressed, a list display screen shown in FIG. 17 is displayed on the display unit. Thereafter, the user presses the "jump" key 11p to select a desired related word 63a. further,
When the user presses the "search" key 11g, an example sentence using related words as shown in FIG. 18 is displayed. FIG. 18 (a)
When "rarely" is selected, FIG. 18B shows "usua
lly ". By displaying the registered related words in a list, the example sentence of the related word desired by the user can be displayed more efficiently and quickly than when the example sentences using the related words are displayed immediately. This is advantageous in that an efficient search can be performed when the user knows the meaning of the related word to some extent.

FIG. 19 is a flowchart of a related word list display process according to the fifth embodiment of the present invention. FIG. 15 (a)
Assuming that the user presses the "jump" key 11p in a state where the sentence shown in (1) is displayed on the display unit (step 55a), a cursor 61b is displayed on the first word of the display unit. By pressing the "jump" key 11p several times, the cursor is moved to a word having a related word (synonym or opposite word) (step 55b). At this time, the CPU searches the word indicated by the cursor for a word registered in the synonym dictionary, and if registered, displays a character of “class” in the lower right of the display unit. . When a word for displaying a related word is selected by the cursor display, when the user presses the "synonymous" key 11n (step 55c), the following search and display processing is executed.

First, the CPU searches the synonym dictionary 15 for a candidate word for a related word of the word currently displayed by the cursor.
Extract from a (step 55d). For example, FIG.
In the case of (a), eight candidates are extracted for the selected word "often" as shown in FIG. The extracted words are temporarily stored in the RAM 12i (step 55e).

Next, a list of all eight extracted candidates is displayed (step 55f). Here, for example, FIG.
Is displayed. When the user presses the jump key 11p with the display as shown in FIG. 17, a cursor display 63a for selecting a word is displayed (step 55g). At this time, the user presses the jump key 11p several times to display the desired word on the cursor. Then, the user determines the word to be selected and presses the "search" key 11g to execute the following display processing. The CPU uses this “search” key 1
If it is detected that 1 g has been pushed down, step 55h
Proceed to.

Next, referring to the grammar table and the original sentence in FIG. 12A, other words necessary for creating the sentence are extracted from the English-Japanese dictionary (step 55).
k). When "rarely" is selected in FIG. 15A, other words can be used as they are, and there is no newly extracted word or changed word. For example,
At this step where "have never" is selected, "go" is changed to "been".

Next, the extracted related words and other words are rearranged so as to be a grammatically correct sentence, a sentence using the related words is created and temporarily stored in the RAM (step 55i). Then, a sentence using the related word is displayed on the display unit (step 55j). Further, for a sentence using the related word, a Japanese translation is created with reference to the English-Japanese dictionary and displayed on the display unit (step 55k). For example, a sentence as shown in FIG. 18 is displayed on the display unit. According to this embodiment, while the sentence of the answer is displayed, the user can simply press the “jump” key 11p, the “synonymous” key 11n, and the “search” key 11g to display the sentence. A sentence having a similar or opposite meaning to the answer sentence can be displayed.

<Embodiment 6> Here, an embodiment will be described in which a user directly inputs a keyword word in an answer sentence. FIG. 20 is an explanatory diagram of a screen display according to the sixth embodiment of the present invention. FIG. 20 (a) shows a state where the "answer" key 11k is pressed after the question sentence input by voice is displayed. At the bottom, a symbol "[" indicating an area where the user inputs characters is displayed. If the user inputs “subway” using the character keys 11c in this state, the result is as shown in FIG.

Thereafter, if the user presses the "search" key 11g, the process of creating an answer sentence is executed in the same manner as in the first embodiment. However, since "subway" is entered as a keyword, An answer sentence containing this keyword is created. That is, the answer correspondence table 15e in FIG.
Although registered as such, “get the subway” including the keyword “subway” is selected. Therefore, in this case, as shown in FIG. 8B, the two sentences are not displayed on the screen, but as shown in FIG. 21A, the “Get the subway to the museum station” is displayed. Will be displayed.

According to this embodiment, although it is necessary for the user to input a keyword, only the text of the answer including the keyword can be displayed. Therefore, when a plurality of texts of the answer are prepared, It is possible to avoid displaying all the sentences, and it is possible to quickly and accurately present an answer that matches the user's intention. Therefore, a smoother response can be made in a dialogue scene.

A plurality of keywords or conditions may be input in the display state as shown in FIG.
For example, when two words “subway” and “change” are input in the display state of FIG. 20A, the answer correspondence table is searched for each of both words, and as shown in FIG. The sentence of the answer containing these two words is displayed. The condition that the user should input is 2
AND condition (A and B) meaning that both keywords are included, OR condition (AB, A or B) meaning that either one is included, NOT condition (not A) meaning that no keyword is included, etc. is there. Also in this case, only the answer sentence that matches the user's intention can be efficiently presented.

<Other Embodiments> FIG.
Although an example of displaying a list of synonyms of a word has been described, the present invention is not limited to this. When a word indicating the frequency or degree of an event is selected, all words related to the word may be displayed in a list. Good. Also, if the list is displayed in order of frequency or degree, it is convenient for the user to select a word. In particular, since words representing the degree are often used in a dialogue, the selection of words by the list display is considered to be very effective.

For example, in FIG. 17, words representing eight different frequencies from "always" to "never" are displayed in order, so that the user can select a word having a frequency as close as possible to his / her own intention. You will be able to more accurately convey what you want to convey. In addition, when synonyms of “big” are displayed, “huge”, “small”, and “tiny” shown in FIG. 14 may be displayed in a list in this order.

Further, as a word group indicating the degree, "high-low", "fast-slow", "long-short", "heavy-light", "hot-light"
"Cold", "large-small", "good-bad", "thick-thin",
“Dark-light”, “light-dark”, “large-small”, “high-low”,
There are "hard-soft" and the like, and a list may be displayed in the order of these degrees.

Further, in the above-described embodiment, all the sentences are displayed on the display unit 11b. However, the sentence inputted by voice, the translated sentence, the sentence of the answer, other words displayed on the display unit, and the like are displayed. Speaker 11
r may be output as voice. For example, as shown in FIG. 1, a “pronunciation” key 11q is provided, and when a certain sentence or the like is displayed, the user can press the “pronunciation” key 11q.
Pressing q causes the displayed text to be output as voice from the speaker.

According to this, the answer to the question sentence can be notified to the other party without having to show the answer sentence displayed to the other party each time. Further, even if the language spoken by the partner is difficult to pronounce, the partner can hear the answer sentence in the language of the partner. Furthermore, the user can know the pronunciation of the displayed sentence by listening to the voice output by himself, and can learn the pronunciation and accent of the unknown language.

[0071]

According to the present invention, when interacting with a person who speaks a different language, it is possible to present a sentence of an accurate answer to the question of the other party, so that it can be performed quickly and without an interpreter. You can have a smooth conversation.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of an embodiment of an electronic translation apparatus according to the present invention.

FIG. 2 is a block diagram illustrating a configuration of an embodiment of an electronic translation apparatus according to the present invention.

FIG. 3 is an explanatory diagram of one embodiment of each table stored in a ROM of the present invention.

FIG. 4 is an explanatory diagram of one embodiment of a grammar table of the present invention.

FIG. 5 is an explanatory diagram of contents displayed on a display unit according to the present invention.

FIG. 6 is an explanatory diagram of one embodiment of an answer correspondence table according to the present invention;

FIG. 7 is an explanatory diagram of contents displayed on a display unit of the present invention.

FIG. 8 is an explanatory diagram of contents displayed on a display unit of the present invention.

FIG. 9 is an operation flowchart of a speech recognition process of one embodiment of the electronic translation device of the present invention.

FIG. 10 is an operation flowchart of an embodiment of an answer search display process according to the present invention.

FIG. 11 is an explanatory diagram of a specific example of the antonym dictionary of the present invention.

FIG. 12 is an explanatory diagram of one embodiment of generating a sentence using the opposite word of the present invention.

FIG. 13 is an operation flowchart of an opposite word selection process according to the present invention.

FIG. 14 is an explanatory diagram of a specific example of a synonym dictionary according to the present invention.

FIG. 15 is an explanatory diagram of an embodiment for generating an example sentence of a synonym of the present invention.

FIG. 16 is an explanatory diagram of a display example of a sub screen according to the present invention.

FIG. 17 is an explanatory diagram of a list display of related words according to the present invention.

FIG. 18 is an explanatory diagram of one embodiment of generating a sentence using related words according to the present invention.

FIG. 19 is an operation flowchart of a related word list display and selection process according to the present invention.

FIG. 20 is an explanatory diagram of a display example of keyword input according to the present invention.

FIG. 21 is an explanatory diagram of a screen displayed after a keyword is input according to the present invention.

[Explanation of symbols]

11a cabinet 11b Display section 11c character key 11d English-Japanese key 11e Japanese-English key 11f Voice recognition key 11g search key 11i subkey 11j Back key 11k answer key 11l switching key 11n synonymous key 11o opposite key 11p Jump key 11q pronunciation key 11r speaker 11s microphone 12a Display (LCD) 12b microcomputer 12c Input part (key) 12d voice input unit (microphone) 12h ROM 12i RAM 15a Various dictionaries 15b Speech recognition table 15d grammar table 15e Answer correspondence table 31a, 31c, 31e Display example of question text 31b, 31d, 31f Display example of answer sentence 34a Keyword input part 34b Display example before keyword input 34c Keywords 34d Display example after keyword input 35a, 35b Display example of answer 51a Example of cursor display of jump candidates 51b Example of display after antonym search 51c, 51e Cursor 51d antonym icon 53a Headword of antonym dictionary 53b Opposite words 61a Cursor display example of jump candidate 61b, 61f, 61g, 61i, 61j Cursor 61c antonym icon 61d Synonyms Icon 61e Display example after synonym search 61h Display example after antonym search 62a, 62c Synonym dictionary entry words 62b Correspondence word 63a cursor 64a cursor

Claims (7)

[Claims] An input unit for inputting a question sentence is paired with a main first word group used in the question sentence and a second word group to be used in a sentence of an answer corresponding to the question sentence. The stored answer information storage unit and the question sentence input by the input unit are searched for whether or not the first word group is included. When the first word group is included, a search is made with the first word group. The second
An answer candidate extracting unit that extracts a word group from the answer information storage unit, and an answer creating unit that creates a sentence of an answer to the input question sentence using words in the extracted second word group. An electronic translation device characterized by the above-mentioned. 2. A translation unit for translating a sentence of the input question sentence and an answer to the question sentence into a predetermined language,
The electronic translation apparatus according to claim 1, further comprising a display unit that displays at least one of the question sentence, the sentence of the answer, and the sentence translated by the translation unit. 3. An antonym dictionary storing a pair of a predetermined word and a word having a meaning opposite to the predetermined word, and storing a pair of a predetermined word and a word having a synonymous meaning of the word. A synonym dictionary and at least one related word dictionary stored in association with each other, including predetermined words and words having the opposite and synonymous meanings, and An instruction unit for selecting one of the words and searching for an antonym, synonym, or related word of the selected word; and an inversion, synonym, or synonym for the word selected by the instruction unit A search unit for searching for a related word by referring to the corresponding dictionary; and replacing the selected word with an antonym, synonym or related word searched by the search unit, and displaying the selected word on the display unit in consideration of grammar. Change the sentence 3. The electronic translation apparatus according to claim 2, further comprising a change unit. 4. When the search unit finds a plurality of opposite words, synonyms or related words by searching, the display unit displays a list of the plurality of opposite words, synonyms or related words, and displays the change unit. Replacing the word selected by the instruction unit with a word specified by a predetermined instruction input among the opposite words, synonyms or related words displayed in the list. . 5. The method according to claim 1, further comprising a second input unit for inputting an arbitrary keyword, wherein when there are a plurality of second word groups extracted by the answer candidate extracting unit, the answer creating unit is input by the second input unit. Using the second word group including the keyword,
2. The electronic translation device according to claim 1, wherein a sentence of the answer is created. 6. The electronic translation device according to claim 2, further comprising a voice output unit that outputs information displayed on said display unit by voice. 7. A step of inputting a question sentence, and storing a main first word group used in the question sentence and a second word group to be used in a sentence of an answer corresponding to the question sentence. Searching the input question sentence whether or not the first word group is included by using the answer information storage unit. If the first word group is included, the first word group is searched for. A step of extracting a pair of second word groups from the answer information storage unit and a step of using the words in the extracted second word group to create a sentence of an answer to the input question sentence A method for creating a sentence in an electronic translation device, characterized by that: