JPH0833883B2 - Input device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of use] According to the present invention, a character string formed by a single character such as a Roman character input from an input means such as a keyboard is valid in view of previously stored language information. The present invention relates to an input device capable of determining whether or not.

[Prior Art] Conventionally, in this type of input device, a character string such as a Latin character input from an input unit such as a keyboard is stored in a storage unit such as a RAM, and this character string is checked for each word at the time of checking. It has a configuration in which it is read out in sections and checked for spelling errors and the like while referring to a dictionary prepared in advance. in this case,
When a spelling error or the like is found, the incorrect word is displayed on the display in a special manner, for example, in reverse video, blinking, or with a specific symbol added, to draw the user's attention.

[Problems to be Solved by the Invention] Although such an input device is extremely useful for detecting an input error, etc., its usability may not be sufficient depending on the mode of inputting and checking a character string. It was considered, and further improvement was desired. That is, when a manuscript exists prior to the input of the character string, the user must check the display and the manuscript alternately displaying the input and stored character string before checking. However, there is a problem that work efficiency is reduced. Also, due to the difference in brightness between the display and the original, the difference in focal length, etc.
There was also a problem that it is likely to cause eye fatigue. Further, as an input error, in addition to the case where the word is not in the dictionary, a word that is in the dictionary but is different from the intended input sound may be mistakenly input. For example, when only the vowel is mistaken and "SIN" or "SON" is input instead of "SIN". Furthermore, there are cases where the user forgets to enter the word itself.

Therefore, the present invention has been made for the purpose of solving such a problem, and provides an input device capable of checking an input character string conveniently.

Structure of the Invention [Means for Solving Problems] In order to achieve the above object, the present invention has the following structures as means for solving the problems. That is, as illustrated in FIG. 1, a character string storage means M2 for storing a character string of monophonic characters such as Roman characters input from the input means M1 such as a keyboard.
And reading means for reading out the stored character string in word units
M3, dictionary storage means M4 storing at least the correct spelling of words, linguistic information including the grammatical attributes and sentence patterns of the words, and the words read by the reading means M3 are stored in the dictionary in advance. An input device provided with at least a judgment means M5 for judging whether or not the spelling is correct, and for judging whether or not the text is valid by comparing with the language information stored in the means M4. When the read word is determined to be an accurate spelling by the determination means M5, the word is pronounced as it is, and it is determined that the read word is not an accurate spelling. In this case, the spelling is pronounced in units of single-tone characters that make up the word, and even if the word has a more accurate spelling, it is determined whether the word is the correct spelling when it is determined that the word is not legal. What is the pronunciation processing based on crab? In ready, an input device, characterized in that only the words corresponding with pronouncing voice output unit M6 and the gist thereof.

[Operation] In the input device of the present invention configured as described above, input means
The reading means M3 reads a character string of monophonic characters such as Roman letters input from the M1 and stored in the character string storing means M2 word by word, and the judging means M5 reads this character string from the dictionary storing means M4.
Compared with the language information including the correct spelling of the word and the grammatical attributes and sentence patterns of the word stored in advance, at least the determination of whether the spelling is correct and whether the sentence is legal Make a decision.

Then, when it is determined that the read word is the correct spelling, the word is pronounced as it is, and when it is determined that the word is not the correct spelling, the unit of the single-note character that constitutes the word. Pronounce spelling in. In other words, when a so-called spelling error is found, this is not pronounced as a word, but is pronounced in roman letters when a single letter character unit that constitutes the word, that is, a roman letter constitutes a single tone character.

As a result, if you enter "SKN" or "SJN" where you should have typed "SIN", the word "SKN" or "SJN" is not in the dictionary.・ Because it is pronounced in Roman letters as “N” or “SJN”, you can notice the mistake. Also, if you enter "SUN" or "SON" where you should originally enter "SIN",
Since the word "SUN" or "SON" is in the dictionary itself, it cannot be pronounced in Roman letters, but the original pronunciation of the word is made, so the input operator thinks it should be pronounced "thin". However, because it is pronounced as "san," it feels strange, and you can find a typo.

In addition, even if the word is correctly spelled, if it is determined that it is not legal in text, only the corresponding word is distinguishable from the pronunciation process based on whether it is correct spelling. Pronounce. For example, after performing the above-mentioned so-called spell check, if there is a word that is not textually valid, it is pronounced. As a result, not only the so-called spelling mistakes described above, but also any word that is not valid in terms of text when viewed from the grammatical attributes and sentence patterns of the word is pronounced. For example, if you have forgotten to enter a verb and have an incorrect sentence pattern (for example, part of speech that is not in the basic sentence pattern), you can also check it.

Example In order to further clarify the configuration of the present invention described above,
Next, a preferred embodiment of the present invention will be described. FIG. 2 is a schematic perspective view of an electronic typewriter as one embodiment of the present invention, and FIG. 3 is a block diagram showing the same electric system.

As shown in both figures, the electronic typewriter 1 takes in the character string input from the keyboard 2 into the electronic control unit 3, stores it, and displays it on the liquid crystal panel 4.
The stored character string is input to the sheet 8 attached to the platen 6 at the same time as the input or in response to a predetermined key input.
In addition, the thermal transfer method is used for printing. In this embodiment, the electronic typewriter 1 mainly handles English sentences, and the keyboard 2 also has an arrangement for English sentences.

When the electronic typewriter 1 performs printing, the electronic control unit 3 drives the stepping motor 18 to control the position of the printing carriage 16 in the row direction to control the printing carriage 16.
The print information is output to the print head 19 mounted on the 16. The printing carriage 16 is slidably supported by a guide shaft 20 provided in parallel with the platen 6, and is provided with a stepping motor.
The rotation of 18 drives the platen 6 in the axial direction via the toothed belt 22. The print head 19 has a plurality of heating elements (not shown), and melts the pigment of the thermal transfer ribbon 24 mounted on the printing carriage 16 to thermally transfer and print. The print information is output to the print head 19 moved along the platen 6 by a flexible printed board (not shown).

A speaker 27 is provided on the side of the electronic typewriter 1.
Is attached to the corresponding part of the case 28,
A plurality of through holes 29 are provided so that the voice output of the speaker 27 can be clearly heard.

The electronic control unit 3, as shown in FIG.
1. Control R that stores control programs and data in advance
OM32, RAM33 for reading and writing, word checking that stores information for checking the words that make up the input character string
It is configured as a logical operation circuit from the ROM 34 and the like. CPU31
In accordance with the procedure (program) stored in the control ROM 32, while performing the processing of temporarily reading and writing the necessary data in the RAM 33, the next input / output processing is performed. That is, (1) processing such as performing key input from the keyboard 2 via the keyboard input / output circuit 35, (2) driving the stepping motor 18 via the print driving circuit 36, and a printing carriage on which a print head 19 is mounted. 16
Process of sending print information to the print head 19 to print characters while controlling the feed of the print head. (3) When printing of one line is completed, the stepping motor 18
Is driven to return the printing carriage 16 to the home position (left end), the line feed motor (not shown) is driven to rotate the platen 6, and the paper 8 is fed by one line. (4) Input from the keyboard 2 Displaying the input characters and other necessary information on the liquid crystal panel 4 through the panel drive circuit 37. (5) Checking the input character string and outputting the data corresponding to the words forming the character string by voice. The processing is such that it is output to the circuit 38 and is output as sound from the speaker 27 using the sound data ROM 39 and the amplifier 40. The CPU 31 sends necessary data, etc. via the bus 41 to the R
OM32, RAM33, keyboard input / output circuit 35, print drive circuit 36,
By interacting with the panel drive circuit 60, the audio output circuit 38, etc., control for realizing the above processing is performed.

Of the above processes, a process for checking an input character string will be described with reference to the flowchart of FIG. The user operates the keyboard 2 first,
A character string (English sentence in this embodiment) is input. The input character string is stored in the RAM 33 as an input sentence. After the input is completed, the key 2c (second
4), the input check routine shown in FIG. 4 is executed from step 100.

First, the sentence is extracted one by one from the beginning of the input sentence stored in the RAM 33 (step 100). In English, one sentence is separated by a period ".", But from the balance with grammatical check, comma ",", colon ":" or semicolon ";" may be treated as one sentence. Then, the sentence is decomposed into words based on spaces and delimiters (step 110), and the value of the variable n indicating the position of the word in the sentence is initialized to 1 (step 120).

After the above preparation, the words in one sentence are sequentially compared with the spelling of the word as one of the language information stored in the word checking ROM 34 (step 130).
If a matching word is found, the word is pronounced (step 150); otherwise, the word is spelled one letter at a time (step 160). Such pronunciation is performed by the CPU 31 extracting a syllable forming a word or a character into a method of regular speech synthesis and outputting a code corresponding to this syllable to the speech output circuit 38. When the voice output circuit 38 receives this code,
The data corresponding to the code is read from the pronunciation data ROM 39, and the voice corresponding to the word or character is synthesized.

Here, if no matching word is found,
Words one letter at a time, eg "D" for "DOG"
The pronunciation is divided into sections such as “O” and “G” (this is called spelling pronunciation), and the sentence including the word is displayed on the liquid crystal panel 4.
Then, the correction process (step 170) is executed. The correction process means a process in which the user uses the keyboard 2 to correct the word or inputs a word (proper noun, coined word, abbreviation, etc.) that is not in the dictionary as additional data. It should be noted that the correction processing may be collectively performed after all the words included in the sentence have been checked.

After the processing of step 150 or step 170 described above is completed, the value of the variable n is incremented by 1 (step
180) until the checking of the words in the sentence is completed (step 190), the above processing (steps 130 to 190) is repeated.

When the spelling check of the word of one sentence is completed, the process proceeds to step 200 and subsequent steps. First, the sentence that has been read (step 100) is divided into basic sentence patterns. If the basic sentence pattern is a compound sentence, it is divided into single sentences, and if it contains a relative pronoun, it is divided into the smallest sentence patterns. Subsequently, the part-of-speech information of the words included in the basic sentence pattern is searched with reference to the data in the word checking dictionary ROM 34, and is compared with the pattern of the part-of-speech arrangement allowed for the possible basic sentence pattern (step 210). . As a result of the comparison, it is judged whether or not the same pattern is found (step 220), and if found, the above comparison is repeated for each of the basic sentence patterns divided in step 200 until the end of the sentence ( Step 210
Or step 230).

On the other hand, if the same pattern is not found in any of the basic sentence patterns, words with different parts of speech are pronounced compared to the most similar sentence pattern (step 24
0), similar to step 170 described above, the sentence is displayed on the liquid crystal panel 4, and the correction process is performed (step 250).
Perform

When the word correction is completed, the process returns to step 200, and the process is performed again from the division into basic sentence patterns. This is because the recognition of the sentence pattern is different even if the punctuation mark or the like is mistaken. When the sentence pattern check is completed to the end of the sentence (step 230),
Further, it is judged whether or not the checking of all the input sentences stored in the RAM 33 is completed (step 260), and the above processes (step 100 to step 260) are repeated one by one until the completion.

When checking the input sentence, the electronic typewriter 1 of the present embodiment configured as described above reads out the input character example by voice in word units. Moreover, if there is an error in spelling, it is spelled out, and if there is an error in the sentence pattern, a word that seems to be incorrect is pronounced, so that the error in the input sentence can be easily recognized. That is, in this embodiment, the spelling and sentence pattern are checked as the language information of the word.

Therefore, according to the electronic typewriter 1 of the present embodiment, it is possible to easily, surely and quickly check the input character example. Moreover, in spelling errors, spelling is pronounced character by character, so it is possible to easily find the location of an error even in a long word or the like. As a result of these, it is needless to say that even if a document exists in advance, the check can be performed by paying attention to only the document, and the work efficiency can be dramatically improved.

Although one embodiment of the present invention has been described above, the present invention is not limited to such an embodiment. For example, a configuration as a word processor or a configuration for outputting a specific voice when an incorrect word is found, or an error As a matter of course, the present invention can be implemented in various modes without departing from the scope of the present invention, such as a configuration in which the most similar word is also output for correction when the found word is found.

EFFECTS OF THE INVENTION As described in detail above, the input device of the present invention pronounces a word as it is when it is determined to be valid, and composes the word when it is determined to be not valid. With the so-called spell check function that pronounces spelling in single-note characters, and even with correctly spelled words,
By appropriately demonstrating the sentence check function of pronouncing only the corresponding word in a distinguishable state from the pronunciation process based on whether the spelling is correct or not when it is determined that the sentence is not legal , The excellent effect that the input character string can be checked extremely easily, surely and quickly. Therefore, the work efficiency of the check can be made extremely high.

[Brief description of drawings]

FIG. 1 is a block diagram illustrating the basic configuration of the present invention, FIG. 2 is a perspective view showing the outline of an electronic typewriter as an embodiment of the present invention, and FIG. 3 is a block diagram showing the same electrical configuration. FIG. 4 is a flowchart showing the processing performed by the electronic typewriter when checking the input sentence. 1 ... Electronic typewriter 2 ... Keyboard, 3 ... Electronic control unit 4 ... Liquid crystal panel, 8 ... Paper 19 ... Print head, 27 ... Speaker

Front page continued (72) Inventor Yasuhiro Hayashi 9-35 Hotta-dori, Mizuho-ku, Nagoya, Aichi Brother Industries, Ltd. (72) Inventor Makoto Suzuki 9-35 Hotta-dori, Mizuho-ku, Aichi, Nagoya Brother Industries, Ltd. In-house (72) Inventor Satoshi Furukawa 9-35 Hotta-dori, Mizuho-ku, Nagoya-shi, Aichi Brother Industries, Ltd. (56) Reference JP-A-57-69470 (JP, A)

Claims (1)

[Claims] 1. A character string storage means for storing a character string composed of monophonic characters such as Roman letters input by an input means such as a keyboard, and a reading means for reading out the stored character string word by word. Dictionary storage means for storing accurate spelling, linguistic information including grammatical attributes and sentence patterns of words, and word information read by the reading means in advance for linguistic information stored in the dictionary storage means. In comparison, the input device includes at least a determination unit for determining whether the spelling is correct and a determination unit for determining whether the text is valid. When it is determined that the read word is the correct spelling, the word is pronounced as it is, and when it is determined that the read word is not the correct spelling, the monophonic sentence that composes the word. Even if a word is spelled in units and is more accurately spelled, it can be distinguished from the pronunciation process based on whether or not the word is the correct spelling if it is determined to be textually incorrect. An input device, comprising: a voice output means for pronouncing only a corresponding word in a state.