JPS6068423A - Input device of japanese sentence - Google Patents

Abstract

PURPOSE:To obtain a Japanese sentence written in KANJI (Chinese character) admixed therewith at a high speed by retrieving a table only to a word or phrases fed in KANA (Japanese syllabary) code, storing the corresponding words, etc. to a homonym table and producing plural sentences including the words stored in the homonym table. CONSTITUTION:A KANA/KANJI converting part 6 extracts the words and KANJI corresponding to a KANA part out of a dictionary of an external memory 4 via a retrieving circuit 5 together with corresponding homonyms and holds temporarily them at a homonym table 10. Then a sentence including KANJI is produced from the words and KANA characters as well as the alphameric characters and symbols which are previously separated and stored to a conversion text table 9 for display. When a homonym selection key is pressed, the homonyms of the word or phrase at the head of the conversion text are combined with the homonyms of the word or phrases which is designated by a cursor and displayed at a homonym display part 17. Then a desired word or phrases is selected out of the homonyms and written to a page memory 11 via a writing circuit 13.

Description

[Detailed Description of the Invention] Industrial Field of Application The present invention relates to a conventional structure and problems associated with a device for inputting Japanese sentences with a touch of kanji into a 1/W submeter q machine or other text processing device. In recent years, the demand for inputting kanji into computers and other text processing devices has increased, and many input devices and input methods are already known, such as kanji keyboards, kanji tablets, Japanese type, etc. All-character arrangement methods are known. K, who operates this type of input device, is two-dimensional (
There are 8 characters that memorize all the positions of all the characters arranged in a row, which requires a lot of skill.

In addition to the full character arrangement method, there is also a method for inputting kanji using Kanakin 11 board as a method for scanning Japanese 8T4 sentences containing just one kanji. One of them is line input/Koni.

This requires a great deal of skill, as each kanji character must be encoded into two to q kana characters and the code must be memorized.

In addition to the method using a kana keyboard, there is a method in which the homonyms corresponding to the kana are obtained by reading the onkun, and the homonyms are sequentially selected based on usage order information of the homonyms. This method is
In order to manage usage order information, the device becomes complicated, and it is necessary to specify the corresponding range of reading kana for every kanji at the time of input1.
There are so many options that it takes time to choose between them. In addition, as a method that does not specify kanji intervals, for example, there is the bunsetsu-seki-seki input method or the solid placement method.The disadvantage of this method is that there are words or kanji whose pronunciations are separated in different ways. be. Therefore, since there are homophones for each word, it takes time and effort to select a desired word or kanji.

OBJECT OF THE INVENTION The present invention aims to provide a kana/kanji/kanji/bunjin-blade device which can quickly obtain desired kanji in a kana/kanji/kanji/bunjin-powered system and which can be easily operated even with an accent. This is the purpose.

Structure of the Invention The present invention searches a table for the pronunciation of a word or kanji input in a kana code, and temporarily stores some words, phrases, and their homophones corresponding to the kana code in a homophone table. Do these words and idioms coexist? Create a kanji conversion sentence for the included phase number and store it in the conversion table. If the number of kanji interchange sentences that you have created is a predetermined number (this number depends on the number of sentences that can be displayed simultaneously in the homophone display area), for example, 8 or less, all of them will be homophones. If there are 8 or more, display the homophones in the homophone display section for character strings in which the number of combinations of homophones does not exceed 8 from the beginning of the character string, and display the homophones in the homophone display section. Select one of the character strings displayed on the screen using the selection key. In this example, if you select some character strings from the beginning of a sentence, repeat the same operation for the next character string to select all 9 desired kanji combinations from among the many homophone combinations. It is a choice.

In this case, a predetermined number of selection keys, ie, eight in the above example, are required, but the number keys on the keyboard can be used for this purpose.

DESCRIPTION OF EMBODIMENTS Next, an embodiment of the present invention will be explained with reference to all the drawings. FIG. 1 is a block diagram of one embodiment of the invention. Each component will be explained below.

(1) is a keyboard device, including kana keys, numeric/alphabet keys, separator keys, homophone selection keys, and homophone designation keys (
(Can also be used as a numeric key), OK key, etc. The kana key is used to input the reading of Japanese sentences in kana. The numeric/alphabetic keys are used to enter combinations of numeric or alphabetic characters in the input text. You can also input the pronunciation of Japanese sentences in Roman letters using the alphabet keys.

The delimiter key is used to specify all the delimiters between words, idioms, and clauses when inputting the pronunciation of Japanese sentences in kana.

The homophone selection key is used to specify the position of the desired kanji when it is not displayed in a text that has been converted to kana-kanji. The homophone corresponding to that kanji is 0.
It is used to display on the homophone display section of the display device (8) from the homophone table of 0.

The homophone designation key is used to specify a desired homophone from among those displayed in the homophone display section (8), and is generally used as a numeric key.

The OK key is used when it is necessary to select all the same sound candy.

(21 is an input buffer, which stores the pronunciation of the Japanese sentence input from the keyboard device (1) 111).

(3) is the number of detections, which detects the kana part in the reading (kana string) of the Japanese sentence in the input buffer (21) and instructs the kana g + < diacritic to the kana-kanji converter (6) described later. The information is attached and output, and the U process of kana-kanji conversion is performed.

(4) is an external storage device, which stores all the words used for kana-kanji conversion and the kanji characters corresponding to each word, as well as table c (hereinafter referred to as a dictionary) containing grammatical information about the words. Furthermore, it is also possible to accommodate all original forms (fonts) of character patterns for a character pattern generator (7) to be described later. A magnetic dinuk, a magnetic drum, or the like is used for this external storage device.

(5 is a search circuit that searches the external storage device (4) for dictionary pages necessary for kana-kanji conversion.

(6) is a kana-kanji conversion unit that searches the dictionary in the external storage device (4) using the search circuit (5) and selects some kana parts corresponding to the kana string in the input buffer (2). Homophones of words with different reading lengths, idioms and sone-sore words? obtain. Furthermore, for each word, the reading length of the word that corresponds to the kana sequence that follows it and that can be correctly connected to the previous word using grammatical or other information in the dictionary. Search the dictionary again for several different words, phrases, and homophones of each word. In this way, for further kana strings,
Obtain words that can be connected at the end of a bunsetsu delimiter or kana part.

In this way, several word strings are created for the input t'L kana string. Here, one single edict sequence is interpreted as one sentence.

These interpreted sentences are combined with characters other than the kana part and output to the conversion text table (9) of the Vc edge description together with pointers of homophones of words in the interpreted sentences. Here, the pointer is the storage position of the homophone in the homophone table αd, which will be described later. Further, the kana-kanji conversion unit (6) outputs the homophones of each word in the interpretation table according to the homophone table (1 (i K pointer).

Next, I will give an example and explain it, including the structure of the dictionary mentioned earlier. As an input example, let's say "Kanrishasenyouno". If we search the dictionary for the word that matches the word with the longest reading length at the beginning of this input kana string, ``Kanri'' is obtained. The page structure of the dictionary at this time is shown in FIG.

Furthermore, from this page, you can also get ``Chikara'' and ``Riki/'', but in the case of ``Chikara'', there are no words starting with the next kana string ``NJ, so there are two candidates.

Search for kana strings following "Kanri" and "Kan".

If you search from the fourth sound "shi j" as a series of "kanri", the dictionary page is shown in Figure 5, and the candidates are "sha" and "shasen" (there are no words starting with the lowercase letter "ya") .

If you search a dictionary for the kana sequence that follows ``sarani'' and ``shasen,'' you will get ``you'' and ``yo.'' Continue [Yo J
Searching for the word M for the kana string that follows K yields ``noko (case particle), and this case particle is used as a clause delimiter, so one interpretation is completed here.Similarly, the kana string that follows vc ``yo'' is obtained. One more interpretation can be obtained by searching for the ninth sound "tsu".

``Kanri J ``Shasen''``You''``No''・-m-
Series A “Kanri” “Shasen” “Yo” “Tsu” “No”・-
・・・-Series B Kakita [1 from the 6th note “se” in the kana sequence following the candidate for Nya 1
0'', the following five interpretations can be obtained.

“Kanri” “Mantis shrimp” “Senyou” “)”・−・・
-・Series C "Kanri""Sha""Sen J "You""No"・
−・・−・Series D [Kan January ``Sori ``Sen''``Yo''``Tsu'' ``)
"・-・・-・Series E Similarly, as a continuation of "1 Kan", the first sound of the kana string "When searching from January, only "Rishi" (interest) is found, and this word is followed by < r, j (lower case). ), there are no words in the dictionary jμ, so it cannot be interpreted as a ``kan'' series based on the last sentence of the clause.

Therefore, this series will be deleted. As a result, we obtain five interpretations that have homophones for each word. FIG. 6 is an explanatory diagram of this.

(7) is a character Bakun generator, and the display device (8) is described later.
) generates a character pattern to display. In general, Ikon 1st grade is used, and the character pattern is +JfJ outside) 96ki 悌,
Device (4)K memorized.

(8) is a display device, and the front bC character/cturn generator (
7) Display the character pattern generated from. C.R.T.

A device such as a plasma is used.

(9) is a conversion text table, which temporarily stores the interpreted sentence outputted from the kana-kanji conversion unit (6) together with a pointer pointing to the position of the homophone in the homophone Chee Lumid.

Oa is a homophone table, which temporarily stores the homophones outputted by the kana-kanji converter (6).

(1) I is a page memory that sequentially adds and stores kana, kanji, and sweet sentences whose homophones have been determined, and can display part or all of them on the display device (8). .

The gloss is a homophone selection section that assembles the homophones in the head of the homophone table and displays them on the homophone @ display section.

To explain this operation using the above example, the homophone table αO includes interpreted sentences as shown in FIG. Since the number of sentences that can be displayed simultaneously on the homophone display section is limited, it is difficult to display all five series having homophones in each series on the homophone display section at the same time.

In this invention, the text is divided and displayed in all stages. R11.In the first stage, regarding the ``kanri'' from the beginning of the interpreted sentence,
Since these are common to each series, as shown in FIG. 7 (al), 1 official, 2 supervisor, and 5 management are designated by the homophone in the homophone display section VC display L7 and by the operator.

In the second step, for the above-mentioned "Kan" J VCk7e <next homophone, the combination of words and phrases is displayed again in the homophone selection section. In this case, as shown in Figure 7(b), there are "Shasen" and "Sha", and the total is 7.
There are groups. Therefore, in the homophone display section, "1 diagonal line, 2 diagonal line,
"3 lanes, 4 lanes, 5 lanes, 6 lanes, 7 lanes" is displayed, and the operator specifies the homophone.

In the fifth stage, as for the homophones following the above, as shown in Figure 6, "senyou" and "sen" are the same, and the total is 4 ffi! ', Ru. Therefore, “1 exclusive use, 2 public use,
4 selections” (Figure 7 (C)). In this way, while displaying homophones in all stages within the range that can be displayed on the display section (8), the desired kanji is selected using the homophone selection section and the Japanese sentence is input.

This circuit is a circuit for adding the determined sentences of the same sound island to the page memo IJ Qli one after another.

α41 is a controller that controls signal exchange between the devices. A microcomputer or the like is used for this purpose. Other circuits (
Write circuit α4), search circuit (5), detection circuit (3)
etc.) can also serve as a function.

Inputting a Japanese sentence using the apparatus of the present invention having the above configuration is performed in the following manner. FIG. 2 is a flowchart showing all the work steps performed by the operator when inputting Japanese text using the apparatus of the present invention.

Step 1 The Sunzu operator uses the keyboard (1) to input Japanese sentences in kana according to their pronunciation.Furthermore, a separator key may be inserted at the segment or word break!ll.

When inputting a Japanese sentence of one line or less, check that the inputted kana sentence has been converted into 9 Kanji sentences and displayed. It is OK as long as it is the desired kanji sweet sentence.
Press the key to move to step 0, and if the kanji is not the desired kanji, move to step 2.

Step 2 Move the cursor to the beginning of the character string you want to change, press the homophone selection key, and display the desired word or phrase in the homophone display section.Homophone specification key (also used as number keys)
Specified by. Returning to the above operation H0, all desired words and kanji are determined for the reading of the input sentence. Once confirmed, move on to the next step.

Step) In step 3, if there are any more sentences to be input, the process moves to step 1, and steps 1 and 2 are repeated until all sentences have been input, and this operation ends.

Next, the configuration of the display device (8) will be explained with reference to FIG. In the same figure, the input display section 05 is a section that displays the inputted kana sentence. 0()1 is the document display section,
In the part that displays the kanji mashirimata whose homophones etc. have been determined in a single step, and the converted text that has been converted from the inputted kana daizen kana kanji, words or kanji that have homophones will blink or be reversed.
Alternatively, the brightness is emphasized to distinguish it from words and characters that do not have homophones, and from kanji combinations that have already been established in '14. The homophone display section Q71 is a section that displays the homophone of a designated word or phrase among the words that have a homophone of the document display section rabbit. Note that while inputting kana, homophones are not selected, and kana manual input is not performed while selecting homophones, so input display section Q51 and homophone display section 071 are located at the same location on the display section. You can also use

Next, the operation of the input device shown in FIG.
:Explain according to the operation flowchart in Figure 2.

The apparatus of the present invention performs the next step when the operator inputs the pronunciation of a Japanese sentence in kana using the keyboard (1).

(1) Step 1. The kana sentence input from the keyboard CI) is temporarily stored in the input buffer (2). Detection circuit (3
)teeth.

This manual buffer (21) separates the delimiter marks, alphabetical characters, numbers, symbols, etc. from the kana part and passes the kana part to the kana-kanji converter (6). Together with the kana-kanji converter, the search circuit (5) It is taken out from the dictionary in the storage device (4) and saved as a homophone table Odπ-station.

In addition, we first divided the words and kana characters into alphanumeric and kanji characters, and then created a sentence with kanji mixed in with the conversion text table (9
! The information is stored in the display device (8) K and displayed on the display device (8)K.

After the operator inputs one section of paperback, if the word or kanji of the converted kanji-mixed sentence is different from the desired sentence, press the homophone selection key to select the next step. Move to. When the desired kanji character is displayed, press the OK key all the way to move to the next sentence.

(11) Step 2. When the Dojin selection key is pressed, up to 8 combinations are created using the homophone of the first word or phrase of the conversion text or the homophone of the word or phrase designated by the cursor, and the homophone of the word or phrase specified by the cursor is created. It is displayed on the display section Q71. The operator selects all desired words or phrases from among these homophones and presses the homophone designation key (which also serves as the number keys 1 to 1). The controller Q4) is
A command is given to the conversion text table (9) to delete the word corresponding to the selected word.Furthermore, according to a command from the controller α, the homophone selection unit Cl21 sends the selected word to the writing circuit α3. The writing circuit aa writes this word into the page memory all. Through this series of operations, the selected homophone is replaced with the designated word on the display of the display device (8).

Next, the homophone selection unit @ selects the homophone table (1 (if there is a homophone in
Then, the operation of step 2 described above can be repeated with the contents of the homophone table aO being empty Kf:r. Once in the air, Nutep returns to IK and moves on to input the next sentence.

As mentioned above, the invention is to input one line or one clause of Japanese text in kana sentences, and use the full text legal system information to correctly connect each word in the context of the kanji word. ``4-This is to convert into sweet sentences, and if there are many combinations of homophones of the sentence input in kana sentences, and the number of combinations exceeds the predetermined number that can be displayed simultaneously on the display screen, this predetermined number is Display a combination of several character strings at the beginning of a sentence on the display device, select the desired character string from among them, repeat this operation all the way, and divide all the long sentences in stages. and display it.
Since it selects homophones, long sentences can be converted into kana characters at the same time, and even long sentences can be quickly converted into kanji characters with simple operations.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the configuration of an embodiment of the present invention, Fig. 2 is a flowchart illustrating the working procedure of the device shown in Fig. 1, and Fig. 3 shows an example of the screen configuration of the display device of the present invention. Figures 4 and 5 are diagrams showing the page structure of the dictionary, Figure 6 is a diagram showing the rejection of word strings consisting of homophone kanji corresponding to kana jinriki and kone, and °70 is a homophone. It is a figure which shows the display state of the homophone in a word display part. (1) --- Keyboard device, (21 --- Input buffer,
(3) --- Kana part detection circuit, (4) --- External storage device, (5, --- Search circuit, (6) --- Kana-kanji converter, (7) --- Character Nocturne Generator, (8)...Display device, (9)...Conversion text table, 0σ...Homophone table, a]]...Page memory, @...Homophone selection section, ■... -・＠Input circuit, 0 bleed--controller, shout...input display section, Ou...document display section, α7I...homophone display section. Name of agent: Patent attorney Etsuji Yoshizaki Figure 3 Figure 4 Figure 5 Figure 6

Claims (1)

[Claims] An input section for inputting Japanese sentences in full kana code, a table containing words and kanji corresponding to each word along with grammatical information corresponding to their reading, and a reading of the word or kanji input in kana code. Search the table and output some corresponding words, idioms, and their homonyms to a table of all homonyms, as well as multiple kanji sweets that include words, idioms, and their homonyms. The kana-kanji converter outputs the ji sentence to the conversion text table, and when the number of homophone combinations of the kanji-mixed sentence corresponding to the Japanese sentence that has been scanned is within a predetermined number, all of the is displayed, and for poems that exceed the above-mentioned number of combinations or a predetermined number, display a string of combinations of homophones within a range of a predetermined number ′(!-) from the first word of these kanji sweet sentences. 1. A Japanese sentence input device comprising: a homophone display section for displaying a homonym; and a selection means for selecting one of the character strings that are not displayed on the homonym display section.