JPS5920133B2 - Input device in information processing system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present application relates to an input device in an information processing system, and is configured to facilitate monitoring of input information and to easily modify the input information.

In recent years, the information industry, including news agencies, has become able to process kanji information from multiple angles and at a high level due to new developments in computer processing technology and input/output equipment related to kanji information processing systems. Kanji information processing has been widely implemented, but a major problem with this kanji information processing is information input.

As an input device for the kanji information processing method, the usual method is to create a perforation table using a kanji keyboard and input this perforation table into a computer, or to input signals directly from the kanji keyboard to the computer. ing. In Kanji information processing, since the target language is Japanese, the number of character types handled is several thousand, and it is necessary to select and operate these characters on the keyboard of the Kanji TV keyboard. Therefore, in order to facilitate the selection operation of such a large number of character types, pen-touch type and the like are now being commercialized instead of the keyboard key touch type. Recently, in order to reduce the number of keys, kanji have been changed to ``sound'' by combining two characters such as kana, alphanumeric characters, numbers, etc.
, ``Kun'' readings, or selecting kanji by appropriate association have also been put into practical use. According to this two-character combination method, the number of keys can be reduced to around 100 characters. . However, in any of the above cases, key operations not only require considerable skill, but also require a dedicated operator. It is also the cause of the delay. Therefore, as a solution to the information input problem, we used the Alphabet keyboard, which is widely used in Europe and America so that anyone can easily use a word processor, to computerize information articles in Japanese using Roman letters. It is conceivable to input this romanization into a processing system, convert it into kanji and kana sentences, and output the kanji and kana sentences.

According to this method, the number of keys is alpha, numeric,
The total number of symbols can be reduced to about a few dozen, making it easy for anyone to
Information such as articles, documents, etc. can be entered into a word processing system.

The problem here is the method of converting Roman letters into Kanji and Kana characters, but this method has already been developed by the applicant, and the applicant is stationed in countries around the world, including Europe and America. News information from correspondents is sent via international leased lines and telex to the international standard ITA-2 (InternatiOnal) Tel.
egraphAlphabet- 2) Enter the code,
This has been put into practical use by converting it into Kanji-kana sentences using the Romaji/Kanji-Kana conversion method developed above. The keyboard of the input device currently in use by the applicant is an ordinary alpha keyboard with a printer attached.
This method outputs input information in Roman letters to a monitor printer by operating keyboard keys, and also outputs a perforated tape in ITA-2 5-unit code, and sends this perforated tape to a transmitter.

The present application relates to this input device, in which keys are pressed using a normal alphanumeric keyboard, but the monitor output is mainly based on flat name characters, making it easier to read, making it easier to detect misspellings, etc., and making it easier to correct them. It is.

FIG. 1 shows a block diagram of the human-powered device of the present invention, in which 1 is a keyboard equipped with an alphanumeric key and a function key for standard Latin telex, 2 is a keyboard buffer section, 3 is an input buffer section, and 4 is a text editing section. , 5
1 is a Roman display buffer section, 6 is a pattern generator, 7 is a Roman display that can display one line or the required number of clauses, 8 is a word conversion section, 9 is an output buffer section, 10 is a printer buffer section, 1
1 is a pattern generator, 12 is a flat name display 1 that can display one line or the required number of phrases on the display.
3 is a monitor printer 14 is a cassette buffer section 15 is a micro cassette 16 is a communication control section 17 is a current conversion section 1
8 is an acoustic coupler section 19 between the keyboard 2 and the input buffer section 3, between the input buffer section 3 and the word conversion section 8, and between the word conversion section 8 and the Roman alphabet display buffer section 5.
Between the text editing section 4 and the word conversion section 8 Between the cassette buffer section 14 and the input buffer section 3 Between the word conversion section 8 and the cassette buffer section 14 Between the word conversion section 8 and the output buffer section 9 Output Input/output connected between the buffer section 9 and the printer buffer section 10, between the output buffer section 9 and the cassette buffer section 14, between the output buffer section 9 and the communication control section 16, and to the bypass circuit of the word conversion section 8, respectively. This is an operation section for operating the control sections a to k as appropriate.

p in FIG. 2 indicates a cursor. As with standard telex, when you use the keyboard 1 to press the alphanumeric keys while looking at a Japanese manuscript, the keyboard buff knob 2 inputs Roman letters one character at a time using the ITA-2 code. Then, phrases are assembled into the input buffer section 3 by inputting one character at a time, and, for example, a group of phrases for one line is stored.

On the other hand, based on the information from the input buffer section 3, the Roman letters are displayed one by one on the Roman letter display 7 via the Roman letter display buffer section 5 and the pattern generator 6 at the same timing as the keystrokes. and keyboard 1
With this operation, you have finished typing keys for one phrase. Next, when you press a key with a function code such as a space code, return code, or line feed sign, the word conversion section 8 converts the Roman alphabet into katakana and a plain name mixed with symbols (hereinafter referred to as A phrase of the name subject in JIS code is sent to the output buffer section 9, and the information is further displayed on the plain name display 12 via the printer buffer section 10 and pattern generator 11. At the same time, it is also printed on the monitor printer 13. In Figure 2, the display surface of the Roman alphabet display 7 and the flat name display 12 are shared, one line of Roman characters is displayed in the lower row of this display surface, and one line of the flat name corresponding to the Roman characters is displayed in the upper row of the Roman characters/flat board, which will be described later. This example shows an example in which the beginnings of the phrases are aligned and displayed using the name correspondence display device. After completing the information corresponding to one line consisting of a plurality of phrases by pressing the key in this way, when the carriage return key at the end of the line is pressed, the one line is recorded in the microcassette 15 in Roman letters,
By repeating these operations, one message is sent to printer 1.
For 3, mainly flat name, and micro cassette 15
are recorded in Roman letters, and information corresponding to one telegram is stored in the text editing section 4 in the form of ITA-2 codes. Therefore, according to the above, since the Roman characters and Kana characters are displayed in correspondence on the display surfaces 7 and 12 for the Roman characters and plain name, input information is very easy to read, and typographical errors and misspellings can be avoided. can also be easily detected.

If a typographical error or a typographical error is detected from the input information displayed on the display, it is corrected according to the following procedure. That is, if a typo or omission is detected in the phrase or line currently being pressed that is displayed on the display, use the cursor operation to indicate the position of the character with the cursor p, or check the typo or omission on the monitor print 13. If detected, call the line containing the character from the text editing section by specifying the line number,
While displaying on the Roman alphabet display, use the cursor to indicate the position of the character with the cursor p, and in this state, modify the character by operating the modification keys provided on the keyboard 1. Then, completed messages are stored in the text editing section 4 and the microcassette 15, respectively.

FIG. 3 shows a cursor operation panel 20 provided on the keyboard 1 and used for correction, which includes a line return key 21 for calling the previous line, a line feed key 22 for calling the next line, Backspace key 23 to move back the previous character by one character, Forward space key 24 to advance the next character by one character, Diagonal upper left key 25 to call up the character diagonally upper left, and Diagonal upper left key 25 to call the character diagonally upper right. It consists of a diagonal upper right key 26 for calling the character on the diagonal lower left, a diagonal lower left key 27 for calling the character on the diagonal lower left, and a diagonal lower right key 28 for calling the character on the diagonal lower right. can be moved to the position of If you wish to send a telegram completed in this way online, for example, using the microcassette 15P) in which the message is recorded in the Roman alphabet,
Enter the romanized telegram from cover section 3 to word conversion section 8.
is bypassed and directly guided to the output buffer section 9. When transmitting mainly using plain name, the message information is guided to the output buffer section 9 through the word conversion section 8, and then sent to the communication control section 16. In the communication control unit 16,
At the communication speed specified by the switch, the current converter 17
The signal is transmitted as a direct current signal or as an acoustic signal by the acoustic coupler part 18.

In this case, in the former case, the flat name display 12 and monitor print 13 are displayed in Roman letters, and in the latter case, the flat name display 12 and monitor print 13 are displayed mainly in the flat name. Next, the word conversion unit 8 which converts Roman characters into plain names will be described in detail below. FIG. 4 is a block diagram of the word conversion unit 8.
0, a clause identification section 40 consisting of a clause end identification circuit 41, a paragraph identification circuit 42, and a page identification circuit 43, which has a function of identifying clause breaks from the arrangement of alphabetic characters, a character type verification circuit 51, and a particle determination circuit 40. Circuit 52, "n" test circuit 53, long sound test circuit 54, and consonant test circuit 5
5, and a Romaji/kana conversion section 60, which includes a Romaji/kana conversion circuit 61 and a 50-on table storage circuit 62.

When an input signal is input directly from the input buffer section 3 or from the text editing section 4 in Roman letters in ITA-2 code via the word buffer section 30 to the end-of-section identification circuit 41, the circuit 41 detects one or two alphabetic characters. A passage is identified by the function code (referring to a space code, return code, line feed code, etc.) that immediately follows the above sequence.

For example, when the Roman alphabet input is "SAKURAOMIR[Jl", a space code is inserted between "A" and "0" and "0" and "M", so USAKURAl and "0" are inserted.
and [MIRU] are each identified as a clause. Next, when the paragraph identification circuit 42 detects a sequence of three or more space codes at the beginning of a line or a character string "BARA" at the end of a line, it adds a paragraph code immediately after them for post-processing. For example, Romaji input ends with (EMS consultation European currency...).

The meeting of roses was adjourned. (End), insert a paragraph mark immediately before "O" at the beginning of the third line of the above example sentence, "D" at the beginning of the fifth line, and immediately after [PARA]. In addition, the page identification circuit 43 uses characters indicating the end of the page in Roman alphabet input, such as [;:ENDI or 1NNNN] in the above sentence example.
When it is detected, a page identification code is inserted immediately after it, and immediately before the symbol indicating the start of the page, for example, in the example sentence above, a page identification code is inserted for post-processing. In section 40, after identifying the clauses, bar graphs and pages, and attaching paragraph codes and page identification codes to the paragraphs and pages, respectively,
The information signal is guided to the next-stage bunsetsu estimating section 50, and the phrasal estimating section 50
In the 0 character type saving circuit 51, the following identification is performed.
In other words, in general, when composing Japanese sentences using plain names,
Considering that it is effective to display foreign place names or people's names in katakana, and to display abbreviations of organization names in roman letters, it is effective for deciphering sentences. For example, "LX" is added as a character type identification code immediately after the phrase, and "VX", for example, is added as a character type identification code immediately after the Roman phrase to be displayed in Roman characters. Then, in the character type saving circuit 51, "LX
For example, phrases with the character type identification code [KANADA
When "LX" is detected, a katakana shift code is inserted in place of "LX", and UV, MSVXl
(European Transit System) is detected, an alpha-abbet shift code is inserted in place of [VX], and the above [1, The signal passes through the circuit 51 without being processed. That is, in the character type verification circuit 51, the flat name and
It is distinguished by three types of output: Katakana and Alphabet.
Next, the particle verification circuit 52 performs the following conversion on the character string at the end of the clause. In other words, "WA" (wa) is "HA"
For example, (is) ISJIHO)^AJ is “NIHONH
Al (Japan) (ko, [E], “WE”, “YE” (e) to “HE” (to), for example, ISJIHONYE” to R
IHONHE” (to Japan) (ko, “O”, [WO” (?
) to [QO] (for example, 1NIH0NW0” to 1N
IH0NQ0U (Japan) (ko, ^~'' 0M01 (mo) [QOMOl (mo) for example 1NIH0NW0M
01 to 1SJIH0NQ0M0” (Japan too), further 1YEWA11W~, “WEWAU to 1-1E”
For example, let's say [NIHONYEWAL as "NI
HON}IEHAl (for Japan). In addition, in the "n" test circuit 53, when "N" is not immediately followed by a vowel or "Y" in a phrase, in order to make it "n", the "N" is changed to "N▼", for example, ISJIHONNO.
Convert `` to NIHONVNOl (Japan).

Further, the long sound verification circuit 54 performs the following conversion on the long sound within the phrase. In other words, in the katakana shift code clause,
For example, change “AA” to “A-” and change 1GAANA1 to [GA-
For example, change “II” to “I-” for “NA” (Ghana).
"ITARII" to "ITARI-" (Italian one), "
For example, change [RUUMANIAL to “R
U {ANIAL (Romania), "EE" is [E-]
For example, [SUEEDENl becomes "SUE-DENl (Sweden)", "00" becomes "O-", for example, 100S[J
TORIA”℃-SUTORIAL (Austria)
In addition, change "00" in the plate name to "0U", for example, 1S0.
Convert "0RI" (soori) to ISOURU (soori).

The conversion is the same when two or more of the same vowels follow,
For example, if "0000" is processed as a flat name, "0U0U
”, or in the case of katakana processing, it is converted to [0-0-”. Furthermore, in the consonant verification circuit 55, when two or more of the same consonants occur in succession in a phrase, or when a "T" is placed before a consonant, "X" is inserted immediately before the same consonant.
Insert "TU" (09). For example (3)0KKY01
is converted to [TOXTUKYOl(TOKI2). As described above, each of the verification circuits 51 to 55 detects the particle, "
After the necessary conversion processing is performed on the long sounds and consonants, the data is sent to the next stage, the Romaji/kana conversion section 60, and is input to the Romaji/kana conversion circuit 61. The contents of the alphabetical spelling table, symbol, and function conversion table of the alphabetical alphabet shown below are implemented in the alphabetical alphabet memory circuit 62. The Romaji signal inputted from the bunsetsu verification unit 50 to the Romaji/kana conversion circuit 61 is converted character by character to flat name, katakana, alphabet, and numeral according to the above-mentioned alphabetical alphabet spelling table, symbol, and function conversion table. and symbols are converted into JIS codes.

That is, Roman characters with katakana shift codes and alphanumeric shift codes are converted to katakana and European characters, respectively.
Other Roman characters are converted to plain names.

Here, for Roman spellings that do not have an alphanumeric shift code added to them and are not included in the alphabetical alphabet, the entire phrase is converted to JIS alphanumeric characters, and a paragraph identification code is added. , three space codes are inserted after the return code and line feed code, and the required number of return codes and line feed codes are added to the page identification code in this embodiment.
FIG. 5 shows a block diagram of a text editing section that corrects typos and omissions, and consists of a phrase correction circuit 71, a line filling circuit 72, a text filling circuit 73, and a text buffer circuit 74. The Roman letter message in the input buffer section 3 has a first route passing through each of the circuits 71, 72, 73 and 74, and a second route leading directly to the text buffer circuit 74.

That is, in the second route, when a Roman character signal is input from the keyboard 1 to the microcassette 15 to the input buffer section 3, the input signal is directly input from the input buffer section 3 to the text buffer circuit 74, and all of one message is processed. In this case, if no line is specified from the keyboard 1, the newest line will always be displayed on the Roman display 7, and if a line number is specified, the specified line will be displayed on the Roman display 7. 7 is displayed. Next, the first route is when modifying characters, call up an arbitrary line for one telegram in the text buffer circuit 74 to the Roman display 7, and while operating the cursor on the display 7, write a phrase as described below. It performs modification (deletion, correction, addition) operations. First, regarding deletion, operate the keys on the cursor operation panel shown in Figure 3 to make the cursor p correspond to the position to be corrected, and then press the all mark key, thereby deleting the required number of characters one by one. , the deleted characters are processed by the line filling circuit 72, and further, the entire text from the next line onward is processed by the text filling circuit 73, and the newly corrected line is as follows: It is displayed on the Roman numeral display T each time the key is pressed. Next, corrections are made by pressing the character key to be corrected instead of pressing the all mark key after performing a force operation, and insertion is performed by operating the insert function button and pressing the character key. The required number of characters are inserted at the position indicated by the cursor p using the substitute key. Next, we will explain a display device for displaying Roman characters and flat names for aligning the beginnings of phrases of one line of Roman characters and the corresponding one line of flat names displayed on the common display surface of the Roman alphabet display 7 and the flat name display 12. Then, as mentioned above, the Roman letters are displayed one by one on the Roman alphabet display 7 at the same timing as the keys are pressed, and the Roman letters are divided into phrases by typing function codes such as space codes. In addition, on the plain name display 12, it is displayed in the state (upper part of FIG. 2) of the plain name that has been converted for each phrase by the word conversion unit 8. In the above, as shown in FIG. It is determined whether the desired character pattern is an alphabetic character pattern (including numbers and symbols), a plain name character pattern, or a katakana character pattern, and this determination information is sent to the matrix circuit 2.
Send to 1.

On the other hand, each time a Roman character signal is sent from the input buffer section 3 to the text buffer circuit 74, a word beginning position register 23 is provided which stores the position of the first (initial) character of the word currently being input via the counter circuit 22. I am. This register 23 is set to zero by a carriage return code, and when there is a space code among the characters on the currently input line, it sets the value of the column from the beginning of the line where that code is located. It has a function. For example, in Figure 2, “2
．． 25'' is 1, the next space code becomes 6, and the value is set so that the next space code of ``PAASENTOL
1, the circuit 21 determines in which position on the flat name display 12 the character pattern is to be displayed based on the character pattern signal and the signal from the word beginning position register 23, and displays the determined display on the display. At the corresponding position on 12, one bunsetsu's worth of clauses consisting mainly of flat names are displayed.

By doing this, as shown in FIG. 2, the beginnings of the roman alphabetical characters and the triangular characters of the flat name are displayed so that they match vertically, making it possible to achieve an extremely easy-to-read display.
The above describes the case where the Roma ¥ input by typing on the keyboard 1 including the Romata key is outputted in Roman letters or as a plain name mainly through the communication system monument section 16.
It is also possible to transmit using the microcassette 15 in which a completed electric fork is recorded. Alternatively, it is also possible to send the plain text from the input buffer section 3 via the word conversion section 8 and the output buffer section 9, and if necessary, the contents of the message stored in the text editing section 41 or the microcassette can be transmitted. It is also possible to perform operations such as calling up and monitoring the contents of the electric fork stored in 15 on the Roman alphabet display 7 or the plain name display 12 as appropriate.
This can be achieved by controlling switching using the operating section 19.

As described above, according to the present application, a Roman character input is converted into a flat name-based phrase in units of clauses, and a Roman character collocation and a phrase mainly based on a flat name are displayed on the same display screen, vertically and with the beginnings of the words matching. As a result, it is extremely easy to decipher, and errors in keystrokes can be checked quickly and accurately.In addition, by using the device of the present application, the conversion efficiency of the device related to the Romaji/Kanji/Kana conversion method previously developed by the applicant is improved. can be significantly improved.

Moreover, compared to the conventional kanji information processing input method that uses a kanji keyboard, it is easier to operate and anyone can use it, and it is also effective in reducing the size of the input device, making it popular not only for newspaper companies but also for Japanese word processors. This is a useful invention that has general-purpose notes as an input device. Furthermore, the present invention can of course be applied to hiragana typewriters and hiragana communication, and can be expected to have a wide range of applications.

[Brief explanation of drawings]

The drawings show an embodiment of the present application, and FIG. 1 is a block diagram of the input device, FIG. 2 is a partial diagram of the display unit, FIG. 3 is a block diagram of the cursor operation panel, and FIG. 4 is a word conversion unit. Figure 5 is the block diagram of the text editing department.
FIG. 6 is a block diagram of a display device that supports Roman characters and flat name names. In the figure, 1 is the keyboard, 2 is the keyboard buffer section, 3 is the input buffer section, 4 is the text editing section, 5 is the Roman alphabet display buffer section, 6 is the pattern generator, 7
1 is a Roman alphabet display, 8 is a word conversion section, 9 is an output buffer section, 10 is a printer buffer section, 11 is a pattern generator, 12 is a flat name display, 13 is a printer, 14 is a cassette buffer section, 15 is a micro cassette, 16 19 is a communication control unit, 19 is an operation unit, a to k are input/output control units, 21 to 28 are cursor keys, 30
40 is a word buffer section, 40 is a clause identification section, 41 is a clause end identification circuit, 42 is a paragraph identification circuit, 43 is a page identification circuit, 50 is a disjunction verification section, 51 is a character type verification circuit, 52
53 is a particle test circuit, 53 is a ``n'' test circuit, 54 is a long sound test circuit, 55 is a consonant test circuit, 60 is a Romaji-kana conversion section, 61 is a Romaji-kana conversion circuit, 62 is a 50-syllabary memory circuit, and 71 is a Also, clause correction circuit, 72 is line filling circuit, 73
74 is a text filling circuit, and 74 is a text buffer circuit.

Claims (1)

[Scope of Claims] 1. A keyboard equipped with alphabet keys, an input buffer unit that assembles Roman characters input in response to keystroke operations from the keyboard into phrases and stores the required number of phrases, and a keyboard that stores the required number of phrases; A Roman character display section that displays Roman characters in sequence each time, a word conversion section that converts a passage of Roman character information input from the input buffer section into hiragana-based information, and an output buffer that converts the hiragana-based information converted by this word conversion section. a hiragana display section that is displayed through the section;
and a recording section that stores one electronic message in Roman characters via an input buffer section, and the Roman character information recorded in the recording section is transferred directly via the input buffer section or through a word conversion section to the output buffer section where it is converted into hiragana-based information. The configuration is such that the Roman character display section and the hiragana display section share a display surface, and each character code of the hiragana-based clause converted by the word conversion section is sent to the output buffer. The discrimination signal from this circuit is sent to the matrix circuit, while the Roman character signal accumulated in the input buffer section is sent to the shift discrimination circuit which discriminates the character pattern through the input buffer section. An output signal from a word-initial position register that stores the position of the initial letter of the word being written is sent to the matrix circuit, and a Roman character clause and a hiragana-based clause are displayed on the display screen so that the beginnings of the words match vertically. An input device for an information processing method characterized by: 2. The word conversion unit includes a word buffer unit, a clause identification unit that identifies clauses, paragraphs, and pages from the alphabetical arrangement, and adds paragraph identification codes and page identification codes at required locations; There is a phrase verification unit that detects character type identification codes to identify three types of hiragana, katakana, and European characters, and also examines particles, ``n'', long sounds, and consonants and performs replacement processing, and a Japanese syllabary table and 2. The input device according to claim 1, further comprising a Roman-ji-kana conversion section that stores a symbol conversion table and converts Roman characters into hiragana characters while collating the symbol conversion table with input Roman characters. 3. A keyboard equipped with alphabet keys, an input buffer section that assembles the Roman characters input by keystroke operations from the keyboard into phrases and stores the required number of phrases, and sequentially displays the Roman characters each time the keystrokes are performed via the input buffer section. A Roman character display section and a word conversion section that converts a passage of Roman information from the input buffer section into hiragana-based information, and a word conversion section that displays the hiragana-based information converted by this word conversion section via the output buffer section. Hiragana display section,
It consists of a recording section that records one message in Roman letters via an input buffer section, and a text editing section that stores one message in ITA-2 code via the input buffer section, and displays information in the text editing section in Roman letters. , the Romaji information recorded in the recording section can be inputted selectively to the storage section and the word conversion section, and the Romaji information recorded in the recording section can be converted into hiragana-based information from the output buffer section directly via the input buffer section or through the word conversion section. The configuration is such that the Roman character display section and the hiragana display section share a display surface, and each character code of the hiragana-based clause converted by the word conversion section is sent to the output buffer. The signal is sent to the shift discrimination circuit that discriminates the character pattern through the
A discrimination signal from this circuit is sent to the matrix circuit, and a word initial position register stores the position of the initial letter of the currently input word via a counter circuit every time the Roman alphabet signal stored in the input buffer section is sent. An input device for an information processing system, characterized in that an output signal of the above is sent to the matrix circuit to display on the display screen a phrase in Roman characters and a phrase mainly in hiragana so that the beginnings of the words are the same at the top and bottom. 4. The word conversion unit includes a word buffer unit, a clause identification unit that identifies clauses, paragraphs, and pages from the alphabetical arrangement, and adds paragraph identification codes and page identification codes at required locations; There is a phrase verification section that detects character type identification codes and identifies three types of characters: hirakaki, katakana, and European characters, and also tests particles, ``n'', long sounds, and consonants, and performs replacement processing. 4. The input device according to claim 3, further comprising a Roman-ji-kana converter that stores a table and a symbol conversion table, and converts Roman characters into Hiragana characters while collating the table and the Roman character input. 5. The text editing department includes a phrase correction circuit that corrects deletions, corrections, and insertions in phrases, a line filling circuit that fills the lines by the number of characters deleted by the circuit, and a text filling circuit that fills the entire text. 4. The input device according to claim 3, comprising a circuit and a text buffer section for storing the corrected information of one message.