JPH0245213B2 - - Google Patents

Description

[Detailed description of the invention]

[Object of the Invention] (Field of Industrial Application) This invention relates to a Japanese typewriter with high-speed printing performance. (Prior Art) Various types of Japanese typewriters have been known. The most common Japanese typewriter has, for example, a 4.2 mm square lead pillar with mirror characters such as Chinese characters formed in an uneven shape on one end, and a large number of mirror characters are arranged on a two-dimensional plane. Characters to be printed are deepened from this plane and are struck onto the printing paper or the like by a mechanical mechanism. At this time, legible printing can be achieved by inserting carbon paper, ribbon, etc. between the lead piece and the printing paper. Printing base paper can also be made by printing directly onto wax-coated paper. As a variation of this, there is also a structure in which lead pieces are placed on a cylindrical surface and the selection is printed by rotating the cylinder and moving it left and right. (Problems to be Solved by the Invention) However, such Japanese typewriters have the following problems. (b) The operator must memorize the position of the characters. (b) Only frequently used characters are called “internal characters” and are placed in an easily accessible manner.
Other characters, called "external characters", can only be accessed by manually loading them into a plurality of reserved positions provided in a two-dimensional plane. (c) Printing speed is slow because it is based on mechanical operation. (d) Providing multiple types of fonts is difficult in terms of performance and cost, and the font size (number of points) is also one size for each model. (e) The operator must memorize the arrangement of letters and undergo training to handle them, and unlike English typewriters, anyone cannot freely use them. (f) The operating arm does not have a so-called home position, so it is not possible to perform blind typing while looking at the original. This is one of the reasons why printing speed decreases.
It is one. On the other hand, there is a so-called kanji input device, which has a function similar to a conventional Japanese typewriter and handles kanji. These are one terminal device of an information processing system centered on a computer (such as a minicomputer), and are classified into the following five types depending on the method of inputting kanji, etc. The conventional systems of this type will be explained in detail below according to this classification. (1) All character arrangement method By arranging the required number of input characters such as kanji, symbols, etc., the operator finds out what is necessary from among them and specifies its location, thereby creating character codes. It is something that generates. This is basically based on the same principle as the old Japanese typewriter mentioned earlier; the former prints directly with a lead piece, whereas the latter generates the code and stores it in a computer. The only difference is that Depending on the mechanism by which a desired character is specified and a code is generated based on the specified character, it can be further classified into the following three types. (1-1) Keyboard method This is the so-called Kantele method and is the most widely used method. 2000 to 3000 characters are arranged on the keyboard. Usually 10 to 15 characters are assigned to around 200 keys, and by pressing this one key, a character is written on the keyboard.
A group of 10-15 characters is selected. 1 more
There are 10 to 15 keys called shift keys to select one character on the keyboard.
Finally, one character can be specified and a code corresponding to it can be generated. (1-2) Tablet method All characters are placed on a two-dimensional plane, and one character is designated using a device that can generate electrical signals, such as a pen. This surface, called a tablet, detects electrical signals emitted by the pen, reads out the position on a two-dimensional plane, and generates a code corresponding to this as a code for a designated character. In some devices, the signal source is located on the tablet side, and the detection function is located on the pen side. Also,
Some use ultrasonic waves instead of electrical signals. (1-3) Improved Japanese typewriter It has the same structure as the Japanese typewriter mentioned above, but has a bar code, uneven code, etc. on the lead piece.
You can detect this when lifting it with the handle, or use the position of the handle to
Some systems find coordinate axes and generate codes based on this by referring to a code table. Furthermore,
There is also a method in which a code for the characters is printed at the same time as the characters, and this code is read again with a separate code reader after the paper is completed. (2) Kana input method This method involves operating keys with only around 100 kana and symbols, and inputting kanji in kana according to their reading. In this case, it can be classified into the following three types depending on the processing method for the same and different characters. (2-1) Combination of kana type and computer All sentences are keyed in in the form of kana characters according to the pronunciation, and based on the context, software installed in the computer judges the text and converts it into a sentence mixed with kanji. (2-2) Select one character from homophones by giving additional information other than the pronunciation at the same time. Other readings, radicals, etc. are used for additional information. (2-3) Two-stroke system: One kanji character is associated with two kana or alphanumeric characters, the operator memorizes this correspondence, and when looking at a character, uses these two corresponding characters. How to key in. (3) Input method for component elements This is a method for specifying characters by dividing the text into two or three parts vertically and horizontally, and by pressing the key indicating the component multiple times. (4) Numerical code method A method in which a several-digit number is assigned to a kanji character, and these numbers are used to enter the text using the numeric keypad. In general, it is used to input characters that are not used very often, and there is no device that can handle all kanji using just this. (5) Handwriting input A method in which characters are handwritten on a tablet, which is recognized and a code is generated. Various kanji input devices other than this Japanese typewriter also have the following problems. The reason for the problem below is due to one of the following items, and is indicated using the code number below. A: It is necessary to memorize the positions of letters, etc. on a two-dimensionally arranged keyboard or tablet. B: Because it is mechanically operated, the operation speed is slow and the noise is loud. C There is no home position, and when selecting letters, it is necessary to follow the surface on which the letters are placed with your eyes. Therefore, it is not possible to type blindly while looking at the manuscript. D: The arm movements required to select characters are extensive, making it impossible to work continuously at high speed for long periods of time. E Must memorize special pronunciations or combinations of alphanumeric characters that correspond to characters. Therefore, instead of reading the original as it is, it is necessary to read the original, remember the corresponding combination, and perform key operations based on this. F Accommodating a large number of characters increases the size of the device, making it impractical for office use. G: It is impossible to perform 100% complete kana to kanji conversion, and in order to correct errors, it is essential to use a combination of numerical code methods. The code corresponding to the letter H must be drawn from a code table (a book that spans several hundred pages). Therefore, the working speed is low. I. It cannot be used as an extremely expensive office Japanese typewriter. (1) Full character arrangement method The common points from (1-1) to (1-3) are that it is necessary to memorize the arrangement of several thousand characters, that there is no home position, and that the operator's arm If you create a device that requires large movements and accommodates a large number of characters, it will be extremely large, making it impractical for office use. In addition, mechanically operated devices have a low operating speed, are difficult for the operator to use for long periods of time, and are loud. (2) Kana input method Kana-Kanji conversion using a computer
It is impossible to achieve 100% perfection, the use of numerical code methods cannot be avoided, and it is also extremely expensive, making it impossible to make a cheap one for office use. In the two-stroke style, students must memorize two kana alphanumeric characters that correspond to kanji. This includes the reading of the kanji (Iwayama for rocky mountain), the meaning of the kanji (Mama) for mother, the shape of the kanji (Hamu for public), and the idiom of the kanji (Himi for secret). 4
The decision was made based on two methods. The number of kanji that can be handled by the representative device, Lineup, is limited to 1,987 characters. The input speed is said to be close to that of an English typewriter, but this combination is also essential to memorize and is not common. Furthermore, the biggest drawback is that instead of reading the manuscript as it is, you follow the text with your eyes, and based on this, you mentally work out the input character symbols you have memorized, and then manually. You will have to key in. This series of tasks goes against the natural functioning of the human brain, is not common, and is not a desirable working environment. Mental fatigue increases and it cannot be used as a Japanese typewriter for office use. The additional information method is also similar. (3) Full character arrangement methods that use constituent elements have similar drawbacks. (4) The numerical code system requires reference to the 100-page code table in the main format, and has drawbacks such as slow operation speed and a high incidence of errors due to misreading. (5) The handwritten input method is most desirable if used at the stage of writing the manuscript. However, at present, input of several hundred characters is analyzed and recognized using a large computer, so it cannot be called a general-purpose device for office use. The above results are summarized in Table 1.

[Structure of the invention]

(Means for Solving the Problems) According to the present invention, there is a keyboard that inputs kana by distinguishing between kanji and non-kanji characters according to the pronunciation of a document, and supports input of kana readings of kanji input from this keyboard. a search unit that outputs the kanji code to be searched; and a search unit that is distinguished for each page to store the kanji code output from the search unit and the code from the keyboard; etc. for each page according to a predetermined format, a pattern memory that outputs pattern information such as characters corresponding to the code output from the page memory, and the pattern output from the pattern memory. The display unit includes a display unit that displays information and a printing device that prints the pattern information output from the pattern memory, and the search unit outputs a plurality of homonyms exceeding a predetermined number in response to a kanji reading kana input. provided is a Japanese typewriter, comprising means for displaying desired characters in multiple times on the display device so that desired characters can be selected, and storing only the kanji code selected by the keyboard in the page memory. be done. (Operation) In the present invention, when an operator inputs kanji and non-kanji characters according to the reading of the document he wants to input, the kanji part is converted into a kanji code by the search unit and stored in the page memory. Ru. In this case, when a plurality of homonyms are output from the search unit, they are displayed on the display device in multiple parts, and the one selected by the keyboard is stored in the page memory. (Embodiment) The Japanese typewriter of the present invention is composed of six elements as shown in FIG. 2, and the basis of this configuration is based on considerations regarding the procedures to be taken by a Japanese typewriter operator. That is, the procedure after the operator obtains the manuscript can be classified into the three operations shown in FIG. In Figure 1, preliminary reading of the manuscript 1
Then, we check the words and phrases of the received manuscript, find and correct typographical errors, and perform work to obtain the necessary information prior to typing, including determining the print format. Next, in type-in 2, the user operates the keyboard while reading the manuscript. In this case, the most natural and quickest operation method is to blindly type on the keyboard, which has the home position while the eyes follow the original, and this is most suitable for continuing to operate the keyboard for a long time. Finally, a mistake is corrected.
In confirmation 3, the typing results are reviewed after completion, and errors are discovered and corrected. The above three types of operations are always performed no matter what type of typewriter is used, and the same applies to English typewriters as well. The Japanese typewriter consists of six main elements. As shown in FIG. 2, there are a keyboard 11, a search function 12, a page memory 13, a pattern memory 14, a display section 15, and a selection function 16. The function of each is as shown in FIG. These are controlled by minicomputers, microcomputers, etc. The keyboard 11 is used by an operator to input information and change information that has already been input. This keyboard has at least the following four types of keys. The first is a kana character key that specifies a kana character. By pressing this key, a code assigned to each kana character is generated.
The second key is for specifying the distinction between kanji and non-kanji input. The third key is an input mode/display mode switching key. When the input mode is specified using this key, character codes and control signals generated on the keyboard are supplied to the outside. Number 4 is a selection key. This key is used to select one of a plurality of homonyms displayed on the display by pressing the key corresponding to the position on the display. The second element is the search function 12. The search function inputs the reading of a kanji character in kana, and generates the corresponding homophone character code. The third element is page memory 13. This memory stores codes given from the keyboard 11, format information indicating positions for placing characters, etc., homophone and allograph codes output from the search function, and the like. Mainly uses large-capacity memories such as CCD shift registers. The fourth element is pattern memory 14.
This is a fixed memory that receives a code such as a character as input and generates a two-dimensional pattern such as a character. The fifth element is the display section 15. This display section displays two-dimensional character patterns obtained from the pattern memory 14 according to the format information and character codes stored in the page memory 13 when the display mode is specified using the keyboard described above. It is to be displayed. For example, six homonyms are displayed in descending order of frequency. When a desired kanji is specified, control is performed so that only that one character is displayed. The sixth element is a selection function 16 for selecting one character from a plurality of homophones. When one kanji is specified according to its arrangement etc. using the selection key (d) on the keyboard 11, only the corresponding one code in the page memory is left and the others are removed, and the input is performed according to the original. is confirmed. By using the above six elements, during the type-in operation 2 shown in FIG. 1, the person following the document visually can type-in by blind typing at the home position. When a certain amount of input, for example half a page, is completed, the input mode is shifted to the display mode. Here, homophone and allograph processing is performed together with typographical error correction and confirmation of the finished product. This flow will be explained in detail using FIG. FIG. 3 shows a flowchart of the work procedure of a Japanese text type operator. The operator of the Japanese typewriter first receives a manuscript 21. The received manuscript is pre-readed 22. In the preliminary reading 22, preliminary preparations are performed, such as checking the reading of characters that are illegible, marking out characters that are not clear, and understanding the printing format. Next, type in 23 using the kana key (a) on the keyboard 11.
Execute. However, kanji are designated in kana according to their reading in the text. In this way, the operator's brain functions naturally, just like reading a book. Therefore, high-speed operation can be performed over a long period of time. However, since it is a kanji,
It is necessary to provide it at the same time as input information for kanji.
Then, press the key (b) for specifying the distinction between kanji and non-kanji input to specify that the input information is kanji. For example, if the keyboard 11 is manufactured so that key (b) is operated with the thumb, this operation will be similar to the shift key on an English typewriter.
Therefore, in terms of key operability, it can be said that there is almost no physical or mental burden on the operator. At this time, the operator follows the document with his eyes and operates the keyboard 11 at the home position with his hands. This is exactly the same as the operability of an English typewriter. Next, in step 24, when you have completed typing in a few lines to one page by blind typing while reading the manuscript, press the switch key (c) to switch from input mode to display mode.
operate. For the data input up to this point, the format to be printed and the code of the character to be printed at the position of the format are stored in the page memory 13. Therefore, in the display mode, the input data is input by the display section 15. The text is displayed in a format that follows the layout of the text. Based on this display, the type operator performs checks 25 such as selecting homophones and correcting erroneous characters. The work procedure up to this point will be explained using the example sentence "Tomorrow's weather will be sunny." First, the procedure for inputting this sentence from the keyboard 11 is performed by the following steps. Step1 Press the kanji specified key (b). Step2 “Asu” or “Ashita” and kana key (a)
key in using . Here, the output of the search unit 12 generates a code corresponding to <tomorrow>. Step3 Press the key to move from kanji to kana. (Cancel the kanji designation key (b).) Step 4 Press the "no" key. Here, the kana code ``no'' appears. Step5 Press the kanji designation key (b). Step6 Operate "Tenki" and Kana key (a). Search function 12 includes the following kanji that can be read as ``tenki'': ``weather'', ``tenki'', ``turning point'', ``outcome'', and ``transfer''.
It contains 5 characters, each of which generates a code for the kanji that makes up each character. Step7 Press the key to switch to kana. Step8 Press the kana key “ha”. Here, the code corresponding to the kana ``wa'' appears. Step9 Press the kanji designation key (b). Step10 Use the kana key (a) to key in "Hare". This code is generated because the search function 12 contains only "Haru" as a kanji character corresponding to the reading "Hare". Step11 To indicate that you want to return to kana, press the key that indicates the separation between kanji and non-kanji input. Step12 When you key in "ru" with the kana key, a code corresponding to the kana "ru" will be generated. When operating the keys as described above, the operator does not need to pay attention to the selection of homonyms or take his eyes off the original when it comes to kanji. However, all you have to do is read the manuscript as you normally would, and use your hands blindly to operate the keyboard. If the input mode is now switched to the display mode by operating the changeover key (c), a display as shown in FIG. 4a will be obtained on the display surface. This corresponds to the kana reading of the search function 12 described in Steps 2, 6, and 10 above. In the display mode, the user selects the correct one as input data from among the homonyms shown on the display screen, and also rereads the input text. In other words, in Figure 4a, at ``tenki,'' one of the five idioms is selected, and after this is finished, one line is checked. Where you need to select kanji, follow the flow of the text. This is in line with the order in which the text is reread. Use a numeric keypad, etc. to change the order of the flow of the sentences displayed over two or more lines, as shown in Figure 4a, where "Weather → Transcription" is written over five lines. Specify the numbers for the top and bottom positions. At this time, corresponding numbers may be displayed simultaneously as shown in FIG. 4a. In Figure 4a, specify (1). As a result, the surface of FIG. 4a changes as shown in FIG. 4c, and as a result, only the "weather" code remains in the page memory 13. and,
Next, pressing a number on the numeric keypad performs the same function as above at the point where the next selection should be made. That is, this selection operation is controlled so that it is performed sequentially only where two or more homophones exist. Homophones can also be numbered and displayed in order of frequency of use. Assuming that a maximum of 6 characters can be displayed once, if there are 8 homonyms, 2 characters will be displayed the second time. The numbers at this time are 1 and 2, and the numbers indicate the order in the display. In this way, the operator can select characters very easily. For example, Figure 5 shows a flowchart for selecting the next candidate for a homophone. A case is shown in which seven candidates are read out and the predetermined number (N) of homophone and allograph words to be displayed is set to five. In the display state shown in Fig. 4b, when the next candidate key is pressed to request display of the next candidate, the pointer P currently fetching and displaying the homophone is incremented by a predetermined number (N). 51). In this example, the value of P is changed from 1 to 6,
The next displayed homonyms start from the 6th one.
If the added value P exceeds the number M of homophones and allographs, the value of P is changed so that the first homophone and allograph is displayed (steps 52 and 53). By performing these operations in steps 51 to 53, the extraction position of the next candidate homophone is determined. Next, the homophone extraction pointer P added to the homophone extraction work pointer m used for display work is entered. At the same time, a counter n that stores the number of homophones displayed on the screen is cleared (step 54). The m-th homophone from the homophone and allograph storage area is extracted and displayed at the position where the first line b currently displaying the homophone and the number n of homophones displayed on the screen are added (step 55). In FIG. 4a, the value of b is in the first row. The homophone extraction work pointer m and the homophone counter n displayed on the screen are incremented by 1 (step 56). When the homophone counter displayed on the screen reaches a predetermined number (N), the display of the homophone is stopped and the process waits for selection or input of the next candidate key (step 57). If the predetermined number (N) is not reached, check to see if there are any more homophones, and if there are other homophones, proceed to step 55.
The return display is repeated (step 58). At this time,
Delete the previously displayed homophone from the screen and display another homophone. If there are no other homophones, the display from the line following the last homophone to n lines is erased (step 59). In the above operations, the operator performs completely different labor at completely different times. Moreover, the switching can be freely selected depending on the operator's own mental and physical fatigue level. Moreover, in all operations, the hands are in the home position and the key can be entered without looking at the fingertips. Therefore, compared to the conventional system in which the eye moves around a wide plane for each word, it is much easier to work with. Furthermore, compared to the conventional method of inputting kana, since the text is keyed in as it is read, the level of mental fatigue is much lower, and it is obvious that no special training is required. If the text is long, the user switches from the display mode to the input mode 26 as shown in FIG. 3, type-in again using the kana key, and repeats the same operation to complete the entire task. In this way, by performing different types of tasks alternately, they mutually help each other recover from fatigue. When all the inputs are confirmed in this way, the print instruction 27 is executed. This causes the printer to issue a print command according to the contents of the page memory 13 and print at high speed. In this case, the contents of the page memory 13 can be determined for one page, for example, and the print command can be generated already at the time of switching 26 from the display mode to the input mode. By the way, the capacity of the page memory 13 and the display section 15
Considering the display ability of , and the ease of selecting the displayed homonyms, it is not appropriate in terms of performance or cost to store all homonyms in the page memory 13 or to display them on the display unit 15. . Therefore, the distribution of homophones is used to limit the kanji that can be input using the kana key. Table 2 shows the results of investigating the number of homonyms for 5,541 frequently used kanji. The number of kanji that can be uniquely determined for one reading is 1082, which is about 20% of the total number of kanji and about 55% of the total reading. Similarly, the number of 12 kanji for one reading is 11, and the ratio to the reading is 0.5%. There are 1 to 12 homonym kanji for one reading, but the total is equivalent to 97.5% of the total. Therefore, 1 to 12 homonyms are input using the kana key (a), and all the corresponding codes from the search section 12 are stored in the page memory. As is clear from Table 2, 25% of all reading materials have 13 or more homonyms, or 50 characters. This will be treated separately. This method will be described later. Table 3 shows the number of different idioms for the same reading for 29,135 idioms. There are 18,292 ways to read 2-kanji compound words. i.e. homonyms
The sum of up to 12 kanji makes up 99.9% of the total reading, which is higher than the rate for single kanji. Those with 13 or more characters are treated the same as single kanji characters, and will be explained later. Tables 4 and 5 are 3 kanji compound words and 4 kanji compound words, respectively.
This is the result of investigating 5505 and 2130 readings of kanji compound words, respectively. The maximum number of different idioms for the same reading is 4 and 3, respectively. Therefore

【table】

【table】

【table】

【table】

〔Effect of the invention〕

According to the present invention, since homophones exceeding a predetermined number are displayed in a plurality of times, an empty area is created on the screen and the lines before and after the same can be displayed together. Therefore, the user can refer to the sentences before and after the relevant line to select an appropriate homonym, which has great practical advantages.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the procedure of a type operator, Fig. 2 is a block diagram explaining the configuration of this invention, Fig. 3 is a flow chart of the work procedure when using the Japanese typewriter of this invention, and Fig. 4 a, b, and c are diagrams showing sentences displayed on the display unit, FIG. 5 is a flowchart explaining an embodiment of the present invention, and FIG. 6 is a configuration diagram of a Japanese typewriter according to an embodiment of the present invention. FIG. 7 is a diagram showing the configuration of another embodiment of the invention. 11...keyboard, 12...search section, 13...
...Page memory, 14...Pattern memory, 15
...display section, 16... selection section, 17... controller, 18... printing section.

Claims (1)

[Claims] 1. Provided with keys such as a kana input key, a key for specifying distinction between kanji and non-kanji input, and a switch key for switching between input mode and display mode for selecting homonyms, etc., by operating each key. a keyboard that generates a corresponding code; a search unit that outputs a kanji code of a homophone and allograph corresponding to the reading kana input of a kanji input from this keyboard; a page memory that is divided into pages for storing codes and stores the output codes of the search section and the codes output from the keyboard, etc., page by page according to a predetermined format; and a pattern corresponding to codes such as characters. a pattern memory that outputs information; a display section that displays pattern information output from the pattern memory according to a code stored in the page memory in response to switching to the display mode; and a display section that prints the pattern information. a printing device; a selection section for selecting any one of the homophones displayed by the display section in response to an instruction from the keyboard; If multiple homonyms exceeding .
A Japanese typewriter characterized by comprising: means for displaying homophones in a plurality of times together with the line before the corresponding line on the display section; and a controller for controlling these.