JP4195134B2 - Japanese input system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a Japanese input system such as a word processor that inputs Japanese using a keyboard, and more particularly to a Japanese input system that inputs Japanese using Roman characters.
[0002]
[Prior art]
The method of inputting Japanese using a keyboard can be roughly divided into a romaji input method and a kana input method. The Romaji input method has an advantage that the key position is easy to remember because there are fewer types of input keys than the kana input method. However, first, in order to input the kana of the second row, it is necessary to input two consonant Roman letters and u. In order to input a stuttering, it is necessary to input three consonant Roman letters, y, and a vowel Roman letter. In addition, in order to input “tei”, “to ぅ”, and “di”, it is necessary to input three, for example, thi, two, and dhi. For the particle “he”, it is necessary to input two of he. Also, for the particle “wo”, it is necessary to input two wo. As a result, in order to input “U”, for example, it is necessary to input a total of three Roman letters in order to distinguish “O” from spelling such as uxo. In addition, in order to distinguish between “n” and the kana of the N-line, it is necessary to overlap n to input “n”. As a result, when inputting a kana in the N line after “n”, it is necessary to input a total of three n. Second, ti corresponds to “Ch”, tu corresponds to “tsu”, di corresponds to “ぢ”, and du corresponds to “zu”. Two types of pronunciation, “Tsu”, correspond. Similarly, two types of pronunciation “da, de, do” and “ぢ, z” correspond to the same romaji d. Similarly, the particle “ha” and the non-particle “ha” are pronounced differently, but both are spelled ha. That is, two pronunciations correspond to the same romaji. Third, “o” and the particle “ha” have the same pronunciation, but the former is spelled o and the latter is spelled wo, so two types of Roman letters correspond to the same pronunciation. Fourth, in order to obtain the confirmation of the long sound symbol “-”, it is necessary to input the key “-”.
[0003]
[Problems to be solved by the invention]
In the conventional technique, as described above, firstly, the number of Roman characters input for kana is large. Second, since two types of pronunciation correspond to the same Romaji, it is complicated and difficult to master. Third, since two types of Roman letters correspond to the same pronunciation, it is complicated and difficult to master. Fourth, the "-" key that must be entered to get the confirmation of the long clef "-" is far away from the alphabetic key. Even when the input is performed with a blind touch, it is necessary to keep an eye on the screen and check the "-" key with the eyes unless the user is very proficient.
[0004]
[Means for Solving the Problems]
In order to solve the first problem described above, when inputting the kana of the Udan, it is possible to input u without omitting u or input only consonants. However, this will cause a problem that the consonant that overlaps the consonant at the time of input of the consonant sound is erroneously converted to the top of the consonant, so to avoid this, When inputting a sound, instead of overlapping consonants, a sound key unique to the sound is set, and when the sound key is input, the sound is converted to “tsu”. In addition, if the consonant and y are overlapped at the time of the input of the stuttering as in the conventional case, there is a problem that the consonant is erroneously converted as a stepped kana, so that the stuttering is avoided. When inputting, the consonant and y are not overlapped, but the consonant and five kinds of stuttering keys set as keys for the stuttering are overlapped. The roaring key alone corresponds to “nya”, “i”, “yu”, “e”, “yo”, so that the number of Roman characters required for inputting the roaring sound is reduced by one. It is also possible to input a kana with a small stutter by using this key alone. Similarly, when inputting “kua” or the like, if the consonant and w are overlapped as in the conventional case, a problem arises that the consonant is erroneously converted as a kana in the U-dan. To avoid this, when entering “kwa” or the like, instead of overlapping consonants and w, instead of combining w and vowels, use a key with two corresponding stuttering keys superimposed. Use. That is, if the same stuttering key is duplicated after the consonant, it is converted to a kana such as “kua”. In addition, when inputting a vowel after “n”, exceptionally, a sounding key is input after n, and both are collectively converted to “n”. By doing so, it is not necessary to overlap n in order to distinguish between “n” and the kana of the Na line, and when inputting the kana of the na line following “n”, it is not necessary to overlap 3 n. It becomes like this. In order to solve the above second problem, in the present invention, “tati toteto”, “tachitsuetsutsu”, and “daidou dedo” are respectively set to t rows. , C and d. As a result, “ti” is associated with ti, “to ぅ” is associated with tu, “di” is associated with di, and “do” is associated with du. Also, “chi” is associated with ci, and “tsu” is associated with c. Instead, “ぢ”, which conventionally corresponds to di, has the same pronunciation as “ji”. Therefore, once “ji” is input, it is converted to “ぢ” by the kana-kanji conversion means. Like that. Similarly, “zu”, which “du” corresponds to in the past, has the same pronunciation as “zu”, so once “zu” is input, it is converted into “zu” by conversion means such as kana / kanji. Like that. In addition, when inputting the particles, “ha”, “he”, each of them is regarded as a kind of kanji, and the readings are input as wa and e, respectively, according to pronunciation, and the roman character kana conversion means. Are converted into “wa” and “e”, respectively, and then converted into “ha” and “to” respectively by the kana-kanji conversion means. Of course, in order to avoid confusion, as before, even if ha is input, it is converted to an auxiliary tone “ha”. Similarly, as before, even if he is entered, it is converted to the particle “to”. In order to solve the third problem, the particle “ha” is associated with o as pronounced. Treating “as” as if it were a kind of Kanji, input “o”, convert it to “O” by the above-mentioned Roman-kana kana conversion means, and then convert “to” by the Kana-Kanji conversion means Like that. Of course, in order to avoid confusion with the prior art, even if the input is “wo” and is converted into “U 一旦” by the Romaji-kana conversion means as before, the particle “ To be converted to. In order to solve the above fourth problem, in the kana-kanji conversion unit, when converting to a word including the long kana "-", the long sound "-" Instead of the key, even if a kana representing the pronunciation of a long sound, that is, a kana of a vowel immediately before the long sound is input, a word including the long sound “-” can be selected from the kana-kanji conversion dictionary.
[0005]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 2 is a block diagram of an embodiment of the present invention. 1 is a keyboard, 2 is a Roman character kana conversion means, 3 is a kana / kanji conversion means, 4 is a kana / kanji conversion dictionary, 5 is a selection means, and 6 is a display device. The romaji kana conversion means 2 receives a romaji input from the keyboard 1, converts a portion that can be converted into a kana according to a certain rule, and displays a character string mixed with kana and romaji on the display device 6. Put a break in the kana part at an appropriate place. When the conversion key on the keyboard 1 is pressed, the delimited part of the kana in the romaji kana conversion means 2 is transferred to the kana-kanji conversion means 3, and the kana-kanji conversion means 3 converts the kana-kanji conversion etc. With reference to the dictionary 4, candidate character strings composed of kanji, kana or other characters as candidates are passed to the selection means 5. The selection unit 5 displays candidates on the display device 6. Then, the selection unit 5 receives which character string is selected from the keyboard 1 and displays the finally selected character string on the display device 6.
[0006]
The keyboard 1 is a device for inputting a sentence in which the kanji, kana, romaji and symbols are mixed in romaji. The keyboard can be replaced by an electronic character input device using a pen.
[0007]
15 types of kanji, k, g, s, z, c, t, d, n, h, b, p, m, y, r, and w or f or v added to them Is a consonant key, a, i, u, e, o are vowel keys, a key other than a vowel key other than a vowel key and a key other than a stuttering key, which is specially assigned in an area where blind touch can be done in alphabet or non-alphabet One kind of key is referred to as a prompt key. Five types of keys that are not consonant keys but other than vowel keys and other than prompting keys and that are in an area where an alphabet or non-alphabet can be blind-touched and are specially assigned are called stuttering keys. A key for converting the displayed kana into a character string including kanji, kana or roman as required is called a conversion key.
[0008]
FIG. 1 is a flowchart of conversion control by the Roman character kana conversion means 2. The Romaji or the like referred to here is a symbol corresponding to a Romaji key or a key other than Romaji used for Romaji input. The vowels, consonants, prompting sounds, and roaring sounds shown here represent the Roman letters represented by the vowel keys, consonant keys, prompting keys, and roaring keys in the Roman character kana conversion means 2, respectively. If there is no Roman character that has not been converted to Kana, and a new Roman character is input, that is, if the first Roman character that has not been converted (S1) occurs, that Roman character is a vowel or a sound key. In the case (S1 vowel or prompting sound), the first romaji or the like that has not been converted, that is, the input vowel or prompting sound is converted into a kana based on the single romaji conversion table (FIG. 3) (S2). Along with this, there are no Roman characters that have not been converted, and the process returns to S1. If the first Romaji etc. that has not been converted is a stutter (S1 stutter), then if the conversion key (S3) is pressed (S3 pressed), the first Romaji etc. that has not been converted is A certain roaring sound is converted into a kana based on a conversion table for single Roman characters (FIG. 3) (S2). Along with this, there are no Roman characters that have not been converted, and the process returns to S1. If the conversion key (S3) is not pressed (S3 is not pressed), then when a Roman character or the like is input and there are two unconverted Roman characters, the second unconverted Roman character (S4) ) Is a vowel, a consonant or a consonant, or a stuttering that is different from the first Romaji, etc. (whether it is a vowel, a consonant or a consonant of S4, or a stuttering that is different from the first Romaji, etc.) Based on the conversion table (FIG. 3), it is converted into a kana (S2). The second romaji etc. is not converted and remains romaji etc. Along with this, the second Roman character that has not been converted becomes the first Roman character, and the process returns to S1. If the second roman character (S4) that has not been converted has the same stuttering sound as the first roman character (same sound as the first roman character in S4), the second roman character that has not been converted and the second roman character, etc. Romaji and the like are combined and converted into one kana based on the duplicate stuttering conversion table (FIG. 4) (S5). Along with this, there are no Roman characters that have not been converted, and the process returns to S1. When the first unconverted Roman letter (S1) is a consonant (S1 consonant), the next conversion key (S6) is pressed (S6 is pressed), the first unconverted Roman letter, etc. Is converted into a kana based on the conversion table (FIG. 3) such as a single Roman character (S2). Along with this, there are no Roman characters that have not been converted, and the process returns to S1. If the conversion key (S6) is not pressed (S6 is not pressed), then when a Roman character or the like is input and the second unconverted Roman character or the like (S7) occurs, the second unconverted Roman character or the like When (S7) is a vowel or a sound (vowel or sound of S7), only the first unconverted Roman letter and the like are combined, and a combination conversion table with consonants (FIGS. 5, 6, or 5, Based on FIG. 6 and FIG. FIG. 5 shows that when the first consonant Romaji is k, g, s, z, t, c, d, or n, the second or second and third Romaji are followed by vowels, prompts or stuttering. Indicates which kana is converted to a kana when it is input and combined with a consonant. Along with this, there are no Roman characters that have not been converted, and the process returns to S1. When the second unconverted Roman alphabet (S7) is a consonant (consonant of S7), only the first unconverted Roman alphabet is converted into a kana based on the single Roman alphabet conversion table (FIG. 3). (S2). Along with this, the second Roman character that has not been converted becomes the first Roman character, and the process returns to S1. When the second unconverted Romaji (S7) is a stutter (sound of S7) and then the conversion key (S9) is pressed (S9 is pressed), the first and second unconverted characters Only romaji etc. are combined and converted into kana based on the combination conversion table (FIG. 5, FIG. 6 or FIG. 5, FIG. 7, FIG. 7) with consonant romaji etc. (S8). Along with this, there are no Roman characters that have not been converted, and the process returns to S1. If the conversion key (S9) is not pressed (S9 is not pressed), then when a Roman character or the like is input and a third Roman character or the like (S10) that is not converted is generated, a third Roman character that is not converted or the like When (S10) is a vowel, a prompt or a consonant, or a stutter different from the second Roman letter (a vowel or a prompt or consonant of S10, or a stutter different from the second Romaji, etc.), the unconverted head and 2 Only the second Roman letter or the like is combined and converted to kana based on the consonant 4 combination conversion table (FIG. 5, FIG. 6 or FIG. 5, FIG. 6, FIG. 7) (S8). Along with this, the third Roman alphabet that has not been converted becomes the first Roman alphabet, etc., and the process returns to S1. If the third unconverted romaji (S10) is the same stuttering as the second romaji (the same stuttering as the second romaji in S10), all the unconverted romaji etc. are combined and consonant Are converted into kana based on the combination conversion table (FIG. 5, FIG. 6 or FIG. 5, FIG. 6, FIG. 7) (S11). Along with this, there are no Roman characters that have not been converted, and the process returns to S1.
[0009]
In the romaji kana conversion means 2, when inputting the particles “wa”, “to”, and “wa”, if the respective readings wa, e, o are inputted, the romaji kana conversion means 2 respectively , “Wa”, “e”, and “o”, and further converted into “ha”, “to”, and “to” by the kana-kanji conversion means 3, respectively.
[0010]
In the kana-kanji conversion means 3, even when the particle “wo” is input, even if “wo” is input as usual and converted into “u ぉ” by the kana-kanji conversion means 2, the kana-kanji conversion etc. By means 3, “to” is converted.
[0011]
In the kana-kanji conversion means 3, when converting to a word including the kana “−” which is a long sound, the kana corresponding to the long sound instead of the key of the long sound “−”, If the kana of the vowel immediately before the long sound is input, a word including the long sound “-” can be selected from the kana-kanji conversion dictionary 4. Of course, the same result can be obtained even if the key of the long sound “-” is inputted as usual.
[0012]
【Example】
The Romaji-kana conversion means 2 is effective for words consisting of Kanji readings that contain a lot of stuttering, that is, Kanji. For example, for “compress”, only seven roman letters “a, x, s, q, k, s, r” are input and then the conversion key is pressed. If you use a Japanese input system, you must input eleven Roman letters such as “a, s, s, y, u, k, u, s, u, r, u”. It is also effective when you enter a word that comes with a kana in the Na line after “n”. For example, for “new appointment”, you can enter six roman letters “s, i, n, n, i, n” and then press the conversion key. When done in the system, you have to enter as many as seven Roman letters, “s, i, n, n, n, i, n”. Next, in order to explain FIG. 1 as an example, for example, when inputting “Fanfare”, FIG. 7 shows a process in which Roman characters are gradually converted to kana each time one key is pressed. Next, FIG. 8 shows a process in which a Roman character is gradually converted into a kana each time a key is pressed when “sharp” is input. Next, in the case of inputting “start construction”, FIG. 8 shows a process in which Roman characters are gradually converted into kana each time one key is pressed.
[0013]
【The invention's effect】
First, in the present invention, it is not necessary to input u in principle when inputting kana in Udan, which occupies a considerable part of the Japanese language, so it is expected that the input speed will be much faster than before. . In the conventional method, in order to input a kana such as a roaring sound, a prompt sound, “n” or “kua”, the consonant key is combined with the consonant key, so u is simply omitted from the Udan kana. It is impossible to distinguish between the two. However, in the present invention, it is possible to save u by omitting the use of keys other than consonant keys for sounds such as prompting sounds, roaring sounds and “kwa”, and setting a dedicated key. In addition, the number of keys required to input the stuttering sound and the prompt sound itself can be eliminated, which greatly contributes to the improvement of the input speed. Specifically, in the present invention, when a vowel is input after “n”, a sounding key is sandwiched between n and the vowel. In this case, even in the conventional general technique, since n must be combined after n, the number of keys that must be input is the same. However, when inputting the kana of the Na line after “n”, the number of n that must be input is three in the past, but in the present invention, one n is sufficient for “n”. Since it can be distinguished even if n is not interposed between them, the effect of two appears. Further, even when a sounding key is inserted in the input according to the present invention, it may be said that there is almost no difference between the spelling and the actual pronunciation, and is consistent with the gist of the present invention that the pronunciation and the spelling are matched as much as possible. Secondly, in the present invention, two types of pronunciation do not correspond to one Roman letter, and two types of Roman letters do not correspond to one pronunciation. In other words, since the spelling of the Roman characters and the pronunciation of the kana to be converted correspond one-to-one, they are simple, clear and consistent, and are quickly learned. As a result, for “tei”, “to ぅ”, “dei”, and “how”, the vowels can be entered directly after t or d. The number of keys required for this is reduced from three to two, and a secondary effect of increasing the input speed is also produced. In addition, since it is only necessary to input one Roman character for “he” and “wo”, there is also a secondary effect that the input speed is increased. Third, in the present invention, even when a word including a long sound is input, it is not necessary to press the key of the long sound “-” at a different position from the blind touch area on the keyboard. However, the effect of being able to input long sounds smoothly appears.
[Brief description of the drawings]
FIG. 1 shows an example of a flowchart of a Roman character kana conversion system according to the present invention. FIG. 2 shows an example of a block diagram of a Japanese input system according to the present invention. Example of Romaji single conversion table included in the Romaji-kana conversion system of the present invention when l, v, q, j, f are set as the stuttering key for each step and x is set as the prompting key. 4. For example, when the l, v, q, j, and f are set as the roaring keys for the A-stage, I-stage, U-stage, D-stage, and O-stage, respectively, they are included in the low-character kana conversion system of the present invention. FIG. 5 shows, for example, l, v, q, j, and f as sounding keys for stage A, stage I, stage U, stage D, and stage O, and x as a sounding key. Consonant k included in the romaji kana conversion system of the present invention. FIG. 6 is an example of a combination conversion table of g, s, z, t, c, d, or n. FIG. 6 shows, for example, l, v, An example of a combination conversion table of consonants h, b, p, m, y, r, or w included in the romaji kana conversion system of the present invention when x is set as a sounding key when q, j, and f are set. FIG. 7 is a combination conversion table of consonant f or v included in the Romanized kana conversion system of the present invention, in which f or v functions as a consonant key when a key other than f or v is used as a stuttering key. An example of FIG. 8 For example, when l is set as the stuttering key for the first stage, when inputting the word “fanfare” using the Romaji-kana conversion system and the Romaji-kanji conversion system of the present invention, The screen display Fig. 9 shows how the screen display changes each time a Roman key is pressed when the word "advanced" is input using the Romaji-kana conversion system and Romaji-kanji conversion system of the present invention. [FIG. 10] For example, when l is set as the stuttering key for the first stage and x is set as the prompting key, “Start construction” is performed using the Romaji-kana conversion system and the Romaji-kanji conversion system of the present invention. Figure showing how the screen display changes each time you press the Roman key when you enter the word
1 Keyboard. 2 Romaji kana conversion means. 3 Kana kanji and other conversion means. 4 Kana-kanji conversion dictionary. 5 Selection means. 6 Display device.

Claims (2)

Romaji kana conversion means that converts the displayed romaji to the corresponding kana in accordance with the specific rules described in the specific romaji kana conversion table, and the display kana when the key for converting kana to kanji is pressed Conversion means for displaying several kana / kanji / romaji / symbol-mixed character string candidates, and selecting means for selecting a desired character string from the displayed character string candidates ,
The Romaji-kana conversion means includes five types of vowel keys a, i, u, e, and o that can input vowels, and k, g, s, z, c, t, d, n, and n that can input consonants. 15 kinds of consonant keys h, b, p, m, y, r, w, 5 kinds of stuttering keys l, v, q, j, f that can input stuttering, the vowel keys, and the consonant keys And a key other than the stuttering key and capable of inputting a sound, and a conversion key,
A specific kana can be input with the vowel key, a specific kana can be input with the consonant key and the conversion key, and the kana `` nya '', `` i '', You can input “yu”, “e”, “yo”, input the kana “t”, which is a sound by using the sound key, and press the roar key twice in succession, Enter “i”, “ぅ”, “e”, “ぉ”, and omit the “u” of the vowel key when entering the kana of the U-dan, and input with the consonant key and the conversion key Do
In inputting the kana “tei”, the consonant key “t” and the vowel key “i” can be used, and in inputting the kana “to ぅ”, the consonant key “t” Use the consonant key “t” and the vowel key “u” to enter the kana “Tua” using the consonant key “c” and the vowel key “a”. When inputting the kana "chi", the consonant key "c" and the vowel key "i" can be used, and when inputting the kana "tsu", the consonant key "c" or the consonant The key "c" and the vowel key "u" can be used to input the kana "tsue" using the consonant key "c" and the vowel key "e". When inputting the kana "Tsuge", the consonant key "c" The vowel key “o” can be used to input the kana “de”, and the consonant key “d” and the vowel key “i” can be used to input the kana “do”. When inputting, the consonant key "d" or the consonant key "d" and the vowel key "u" can be used. When inputting the kana "ji", the consonant key "z" The vowel key "i" can be used to input the kana "z" using the consonant key "z" or the consonant key "z" and the vowel key "u". ,
The kana-kanji conversion means can convert the kana “to” and the kana “o” to each other, can convert the kana “ha” and the kana “wa” to each other, and the kana “to” and the kana “E” can be converted to each other, kana “ji” and kana “ぢ” can be converted to each other, kana “z” and kana “zu” can be converted to each other, A Japanese input system that can convert the kana of the vowel immediately before the long sound, which is a kana representing pronunciation, to the long sound "-".   2. The Japanese input system according to claim 1, wherein the kana “n” can be input by using “n” of the consonant key and one of the consonant keys or the prompt key.

Images