JPS62293376A - Document processor - Google Patents

Abstract

PURPOSE:To print normal corresponding candidate words despite the presence of unselected homonym words by printing the highest priority or the homonyms selected by a selecting means in case document information includes unfixed homonyms in a print mode. CONSTITUTION:A selection processing program for printing homonyms is stored in a ROM 2. The address of the head character of a line buffer is stored in a pointer PTR of a RAM 3 and a variable I is cleared. Then the control byte indicated by the pointer is checked and a homonym pointer is decided. Thus a homonym pointer (a) indicated by a pointer +1 is extracted. Based on the value of the pointer (a), the address of the homonym data stored in a homonym buffer is obtained. Then a number (b) of the homonym serving as a pointer +2 is extracted and then a code (c) of the b-th character in the character code string of a 1st candidate is extracted out of the obtained homonym data. Then COH is stored in the byte indicated by a pointer with the code (c) stored in the following two bytes respectively. The homonym pointer data is stored in the line buffer is replaced with the corresponding KANJI codes of the 1st candidate.

Description

Detailed Description of the Invention 3. Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a document processing device that inputs document information through kana-kanji conversion.

[Prior Art] There are many methods for inputting kanji in document processing devices such as Japanese word processors, such as full character array keyboard input and multi-stage shift input, but recently the kana (romanji) kanji conversion method has become mainstream. It is becoming. In this method, by inputting the pronunciation of a kanji character in kana or romaji characters from the keyboard, a word dictionary is searched for words that have kana (romaji) characters in their pronunciation, and the words written in kana are replaced with words written in kanji. However, if there are two or more words with the same reading (homonyms), one of the plural candidate kanji (first candidate) is displayed on the display with low brightness or inverted. If you want to display other candidate kanji, press the "Previous candidate" or "Next candidate" key, and a list of homophones will be displayed at the bottom of the display. Correspondingly, the list of homophones in the document will also be displayed. It's about to change.

For example, as shown in Figure 2 (A), when you enter the reading ``Daisankai'' and press the conversion key, the first candidate ``3rd'' is displayed instead of the kana notation -, as shown in 20. Is displayed. If you press the "Next candidate" key here, the input pronunciation 22 and a list of homophones 23 are displayed at the bottom of the display, and the display 21 of the homophones in the document is the second candidate "Third time" 24. Changes to When a desired Chinese character is selected in this way and the selection key is pressed, the homophoned Chinese character is selected, the low brightness or inverted display is replaced by a normal display, and the display is stored in the document memory as document information.

There are two types of kana-kanji conversion methods: "successive conversion method" and "batch conversion method." In the former case, when a homophone occurs, the next input cannot be made unless the homophone is selected.
In the latter case, the next input can be performed while leaving the homophone, so
In the case of the batch conversion method, some unselected homophones may remain in the created document.

When attempting to print such a document, with conventional devices of this type, unselected homophones are printed with spaces or cannot be printed at all. For example, if 30 to 33 are not selected in Figure 3, 30 to 33 will be selected as shown in 34 when printing.
The part corresponding to 33 was printed with spaces.

[Problems to be Solved by the Invention] The present invention has been made in view of the drawbacks of the conventional examples described above, and is a document processing method that allows normal printing even when homophones are left unselected in a document. The purpose is to provide equipment.

Another object of the present invention is to provide a document processing device that can print homophone candidate words even if the candidate words are selected and left unconfirmed in a document.

[Means for Solving the Problems] In order to achieve the above object, the document processing device of the present invention has the following configuration.

That is, an input means for inputting document information, a conversion means for performing kana-kanji conversion by referring to a dictionary section based on reading information input from the input means, a storage means for storing the document information, and a storage means for storing the document information. A document processing device comprising a printing means for printing out document information, a determining means for determining a priority order of homophones corresponding to the reading information, and a determination means for extracting the homophones from the dictionary section according to the priority order. extracting means, determining means for selecting and determining one of the homophones, and means for printing the homophone with the highest priority when there is an undetermined homophone in the document information at the time of printing. Be prepared.

Furthermore, the document processing device of the present invention includes an input means for inputting document information, a conversion means for performing kana-kanji conversion by referring to a dictionary section based on the reading information input from the input means, and a document processing device, comprising: a printing means for printing out a homonym; a display means for extracting and displaying a homophone corresponding to the pronunciation information; a selection means for selecting one of the homonym; and storage means for determining and storing the homophones.

[Operation] In the above configuration, if an undetermined homophone exists in the document information at the time of printing, an operation is performed to print the homophone with the highest priority.

Further, when an undetermined homophone exists in the document information at the time of printing, an operation is performed to print the homophone selected by the selection means.

[Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a hardware block diagram of a document processing apparatus according to an embodiment of the present invention.

In the figure, CPU 1 is, for example, a microprocessor,
It performs calculations and logical judgments according to the control program shown in the flowcharts from FIG. 12 onwards in ROM2, and
Control signals and various data are transferred to devices, memories, etc. connected to 0. Of the buses 10, CB is a control signal bus, DB is a data bus, and AB
is the address bus. The ROM 2 is a read-only memory that stores programs for control procedures of the CPU 1, fixed data, and the like. The RAM 3 is a random access memory and includes various work areas such as buffers and counters, which will be described later. DISK 4 is an auxiliary storage device such as a floppy disk, and is used to store word dictionaries and input documents.

A keyboard (KB) 5 is used for inputting document data and various operation instructions. 6 is document memory (DBUF
), which temporarily stores input documents and homophone data. The CRT 7 is a display device such as a cathode ray tube, and is used to display input document data, homophones, and the like.

CO2 is a character generator that stores dot patterns corresponding to character codes. 9 is a printer (P
RT) is used to print documents entered.

[Description of Document Memory (FIGS. 4 to 11)] FIG. 4 is a diagram showing the structure of the document memory DBUF6.

In the figure, 40 is a document management data area, and the address and format of document data and homophone buffer (number of lines on one page, 1
(number of characters in a line, etc.). 41 is a document data area, the first address of which is DSAD, and the last address is DE.
Stores the character string of document data input in AD. The document data area 41 is expanded toward the higher address side each time one line is input. 43 is a homophone buffer whose first address is H3AD and whose final address is HEAD. The homophone buffer 43 is expanded to the lower address side each time a homophone is generated by kana-kanji conversion during document input. 42 represents an empty area that is not yet used.

FIG. 5 is a diagram showing the structure of the document management data area 40.

Each word is composed of 16 bits, and 50 holds the start address DSAD of the document data area 41, and 51 holds the address DEAD of the last line of the document data area 41. In addition, 52 is the start address H3AD of the homophone buffer 43,
53 stores the final address (IEAD) of the homophone buffer 43.

FIG. 6 is a diagram showing the structure of the document data area 41. Document data is managed in units of lines, with each line having the number of characters specified by "number of characters in one line."

FIG. 7 is a diagram showing the structure of each line of data in the document data area 41.

The line data consists of a 3-byte control data section 7o storing line feed pitch, etc., and 3-byte character data 71 to 74 each representing each character.

Figure 8 (A) and (B) are character data sections 71 to 74 in Figure 7.
In the configuration diagram, each character data part consists of 3 bytes. FIG. 8(A) is a diagram showing the structure of normal character data 80, and FIG. 8(B) is a diagram showing the structure of homophone pointer data 81.

Normal character data 8o is control byte 82 (OCOH, H
represents a hexadecimal number) and the 2-byte character code 83
．． It consists of 84. The character code 83.84 is here the JISC6226 code. On the other hand, the homophone pointer data 81 is composed of a control byte 85 (OEOH), a homophone pointer 86 and a homophone number 87. The homophone pointer 86 is data indicating at what position in the homophone buffer 43 the homophone data corresponding to the homophone is stored, and takes a value of 1 to 255. The homophone number 8 is data indicating which character of the homophone pointer data string the character data corresponds to.

Figure 9 shows homophone pointer data 8 corresponding to document data.
FIG. 1 is a diagram showing an example of No. 1;

In the figure, four characters 90 to 93 from the mth character to the m+3rd character indicate a homophone pointer data string in which a homophone exists and is not selected, and those homophone pointer data 8
94 to 97 showed 1. Each pointer data 9
4 to 97, “0EOH” of 85 is a control byte,
"See" of the homophone pointer 86 indicates the position of the homophone data corresponding to the character in question, and is common to all pointer data strings 94-97. In addition, numbers 1 to 4 are stored in the homonym number column 87 in order from the beginning, and each character data 90 to 93 is the first to 4th number in the homophone pointer data string.
This indicates that it is the fourth character.

FIG. 10 is a diagram showing the data structure of the homophone buffer 43.

The homophone buffer 43 consists of homophone data with a fixed length of 128 bytes per homophone word, and the homophone data 1
00 to 103 are numbered sequentially starting from 1 from the high address side. Each time a homophone is generated, it is extended by 128 bytes toward the lower address, and the value of the start address H3AD is also communicated accordingly, and each code is deleted each time a homophone is selected.

FIG. 11 is a diagram showing the data structure of homophone data 100 as an example, and the data structures of other homophone data are exactly the same.

In the same figure, the in-use flag 110 is a flag indicating whether homophone data is stored in the area of the homophone data 100; if it is 1°, it means that there is data;
If so, it indicates that there is no data. IKLIII is an area for storing the number of characters in kana reading, and 112 on the IC is an area for storing the number of characters for homophone candidate kanji. However, if the number of candidate kanji characters differs for each candidate, the maximum number of characters shall be stored. ILN 113 is an area that stores the total number of homophone candidates. Further, DISPN 114 is data indicating the number of the homophone candidate currently displayed on the display.

The kana reading character string 115 stores a character string in kana reading input from the keyboard 5, and the number of characters corresponds to the value of IKL. Also, Kanji code (1-n) 116-1
18 stores the kanji code strings of the candidate kanji characters read from the word dictionary of the DISK 4 in numerical order corresponding to the priority order of the candidate kanji characters. Here, n is the same as the value of ILN113.

However, the number of characters in the kanji code string of the kanji code is 11 on the IC.
If it is smaller than 2, O is stored in the remaining part. Finally, empty area 119 represents an area that is not in use.

[Explanation of homophone creation, display, and selection process (Figure 12~
FIG. 18)] FIG. 12 is a schematic flowchart of the kana-to-kanji conversion process executed when kana characters are input from the keyboard 5 and the kanji conversion key is pressed.

When the kana-kanji conversion process is started by inputting from the keyboard 5, candidate kanji extraction process is first executed in step S1, and the candidate kanji corresponding to the specified kana reading is extracted from the word dictionary, and multiple homophones are extracted. When doing so, they are stored in the homophone buffer 43 in priority order. Step S2
Then, check the number of candidate kanji as a result of the candidate kanji extraction process, and if there is only one, proceed to step S3, write the candidate kanji in the document data after the cursor position, and move the cursor position to the next kanji after the cursor position in step S4. Update to point to the character.

If there are two or more candidate kanji in step S2, the process advances to step S5 to create a homophone buffer 43, and if there is an error in step S6, nothing is done. If there is no error, the process proceeds to step S7, where homophone pointer data 81 is created, and then candidate characters are displayed in step S8.

FIG. 13 shows the creation process of the homophone buffer 43 in step S5.

First, in step SIO, the pointer PTR of RAM3 that holds the address of the homophone buffer 43 is set to (HEAD-12).
8) Store the start address of homophone 1. Next, in step Sll, the RAM 3 that holds the number of the homophone buffer 43
Assign 1 to REC. In step S12, the final address and the first address of the homophone buffer 43 are compared, and if they match, it means that there is no homophone, and the process jumps to step 316. If not, the process proceeds to step S13, and the phase-in-use flag 11 of the homophone data area pointed to by PTR is
0 is checked, and if it is 0 (unused), the process advances to step S17. If not, step S14 checks the next homophone data area from the PTR.
Subtract 8 to update the address and add 1 to REC. As a result, if PTR does not become smaller than I(SAD) in step S15, the processes of steps 313 to S15 are repeated to search for an empty area in the homophone buffer 43 in which no homophone has been set.

When the value of pointer PTR becomes smaller than HSAD in step S15, the process proceeds to step S16, and HSAD and DEAD are
If the value is smaller than the final address of the document data area 41, it means that the homophone buffer 43 is full, so error processing is performed in step S18, and the process ends.

Otherwise, the process advances to step S19, sets the in-use flag 110 of the homophone data area pointed to by PTR to 1, stores the number of characters in the kana reading in IKLIII in step S20, and stores the maximum number of candidate kanji characters in ICL 112 in step s21. do. Next, in step S22, after storing the total number of candidate kanji in ILN113, in step S23, DISP
Store 1 in N. Finally, in step 324, candidate kanji are stored in the kanji sections 116 to 118 of the homophone data in order from the first candidate, as shown in FIG.

FIG. 14 is a flowchart of the homophone pointer creation process in step s7 of FIG. 12.

First, in step S30, an address within the document data area 41 is obtained from the cursor position and stored in ADR of the RAM 3. Next, in step S31, the value of the maximum length ICL of the candidate kanji is stored in I of the RAM 3, and 1 is stored in J of the RAM 3. Next, in step S32, "EOH" is stored as a control byte at the address pointed to by ADH. Step S33
Then, the number (RFC) of the homophone buffer 43 created above is stored in the next address, and furthermore, in step s34, J is stored in the next address [ADR+2]. This sounds like homophone number 87 in FIG. Next, in step S35, the value of I is subtracted by 1, J is added by 1, and ADR is added by 3. As a result, if 1≦0 in step 336, the process ends. If not, then in step S37 it is checked whether ADR does not exceed the end of one line, and if it does not, the process returns to step S32. If it exceeds, add 3 more to ADR in step 33B (
After that, the process returns to step S32.

FIG. 15 is a flowchart of the candidate character display process in step S8 of FIG. 12.

When the previous candidate/next candidate key is pressed, the process proceeds to step S40, and the value of the homophone pointer 86 is extracted from the homophone pointer data 81 pointed to by the cursor (set to n). Step S
In step 41, the address of the nth homophone data is determined. If the key input in step S42 is the next candidate key, in step 343 the DISPN in the homophone data is
Add 1 to. In step S44, DISPN is compared with the total number of homophone candidates ILN, and if the value of DISPN exceeds the value of ILN, the value of ILN is changed to DIS in step S45.
Substitute into PH. However, if DTSPN exceeds ILN as a result, DISPN=ILN.

On the other hand, if the key input in step S42 is the previous candidate key, the process proceeds to step 346, where 1 is subtracted from DISPN. In step S47, it is determined whether DISPN has become smaller than 1. If it is less than 1, the process advances to step S48 and DI
Let SPN=1.

Next, in step 34.9, a homophone display process is executed to redisplay the homophone pointed to by the cursor, and in step S50, a homophone list display process is executed to display a list of homophones at the bottom of the CRT display 7. .

FIG. 16 shows a flowchart of the homophone display process in step S49 of FIG. 15.

In step S60, the first address of the homophone pointer data string is determined from the cursor position. Next step S61
The homophone pointer 8 is obtained from the homophone pointer data 81.
Take out the value of 6 (let this be n). Next step S
In step 62, the address of the n-th homophone data in the homophone buffer 43 is determined, and in step 563, the DISPN-th candidate kanji code string from the homophone data is extracted. Next, in step S64, the characters in the ICL 112 after the cursor position are replaced with the candidate kanji code string, and displayed in reverse black and white.

FIG. 17 is a flowchart showing the process when the selection key is pressed to select a homophone.

When the selection key is pressed, first, in step s70, the start address of the homophone pointer data string 81 pointed to by the cursor is obtained, and then in step S71, the value of the homophone pointer 86 is extracted from the homophone pointer data (this is referred to as n). do). Next, the nth homophone data address in the homophone buffer 43 is determined (step 572). Step S73
Then, the DISPN-th candidate kanji code string in the n-th homophone data is extracted, and in step S74, the document data corresponding to the characters of the ICL 112 after the cursor position are replaced with the candidate code string. In step S75, the in-use flag 110 indicating that the homophone data is vacant is cleared. In step S76, the address of the homophone data and the value of the start address H3AD of the homophone buffer 43 are compared,
If they do not match, the process ends.

If they match, the process advances to step S77 to clear the homophone data. In step S77, the address of the homophone data is stored in the PTR. Next, in step 378, the in-use flag 11 of the homophone data pointed to by PTR is
Checks for 0, and if it is 1 (in use), ends the process. If 0 (unused), proceed to step S79 and H3
The value of PTR is assigned to AD, and 128 is added to PTH in step S80. As a result, if PTR does not exceed HEAD in step s81, the processes of steps 878 to S80 are repeated. In this way, continuous selected and confirmed homophone data areas are erased from the start address H3AD of the homophone buffer 43.

FIG. 18 is a flowchart showing the operation of printing processing to the printer 9.

When printing is activated from the keyboard 5 or the like, the process advances to step S90, and the start address of the document data 41 is stored in the document data pointer ADR. Step S91
Now clear the line counter (LINEC). Next, in step S92, one line of the document data 41 indicated by the ADH is transferred to the line buffer of the RAM 3, which is a work for printing.

Here, the line buffer is a work area for holding data for one line of a document, and has a byte size of 3×(number of characters in one line). In step S93, a printing homophone selection process (details will be described later) is performed based on one line of document data stored in the line buffer. In step S95, an output image (dot pattern) for one line is created based on the contents of the line buffer after the printing homophone processing and is output to the printer 9.

Thereafter, 1 is added to LINEC in step 396, and the number of bytes for one line ((number of characters in one line+1)×3) is added to ADR in step S97. After that, in step S98, ADR
and HEAD, and if ADR exceeds HEAD, processing of all document data has been completed, so step S
At 99, a page break is performed and the process ends. If not, the process advances to the STELAB 5100, and then LINEC is compared with the number of lines on one page, and if LINEC does not exceed the number of lines on one page, the process returns to step S92. If it is exceeded, step 5
After a page break is performed in step 101, the process returns to step S91.

[Explanation of the printing homophone selection process (FIGS. 19 and 20)] FIG. 19 is a flowchart of the printing homophone selection process according to the embodiment of the present invention, and this program is stored in the ROM2. .

In the printing homophone selection process, first, in step 5110, the )' address of the first character of the line buffer is stored in the pointer PTR that holds the address of the line buffer, and in step 5111
Clear the variable I that holds the count of characters. Next, in step 5112, the control byte 85 of the character data pointed to by PTR is ticked, and if it is not "EOH" (ie, homophone pointer data 81), the process jumps to step 5119. If 'EOI(''), step 51
Proceed to step 13 and retrieve the byte pointed to by the content of PTH + 1 (the content of the homophone pointer 86) (temporarily assume this is a)
. In step 5114, the address of the homophone data in the homophone buffer 43 is determined based on the value of a.

In step 5115, the content of the byte pointed to by the content of PTR +2, that is, the homophone number 87 is extracted (set to b). In step 5116, the first candidate kanji code string (kanji code 1) in the homophone data obtained in step 5115
The character code (for example, JIS code) of the b-th character is extracted (this is assumed to be C). Next, in step 5117, the PT
Store 'Cot('' in the byte pointed to by H, and step 51
At step 18, store the above character code C in the next 2 bytes.

In step 5119, add 1 to ■, resulting in step 5
If at step 120, ■ exceeds the number of characters in one line, the process ends. If not, the process returns to step 5112.

As explained above, if homophone pointer data exists in the row buffer, it will be replaced with the first candidate Kanji code in the corresponding homophone data.

For example, in Figure 20, the mth character 90 to m+3
If the four characters of the th character 93 are homophone pointer data, the homophone pointer data is replaced with the first candidate kanji code string 202 of the corresponding homophone data 201, and the character string indicated by 200 is replaced with the first candidate kanji code string 202. printed.

In this way, when homophone pointer data is detected in one line during printing, it is replaced with the first candidate Chinese character of the homophone data corresponding to the homophone pointer data and printed.

However, this process is only performed during printing, and the document data itself does not change.

[Other Examples (Figures 21 to 22)] In the above embodiment, when a homophone is detected in a document, the first candidate kanji string of the corresponding homophone data is unconditionally selected. However, the DISPNth candidate kanji of the homophone data (the kanji string displayed on the CRT 7) may be printed instead.

The operation in this case is shown in FIG. In the figure, step 51
Steps 30 to 5135 correspond to step 8110 in FIG.
~Step 5115 is exactly the same, so the explanation will be omitted. In step 5136, DISPN in the homophone data is
Extract the character code of the b-th character of the candidate kanji ``C'' string (let this be C). Step 813
7. In step 813B, the homophone pointer data pointed to by PTH is written in override mode. The subsequent processing is also the same as that shown in FIG. 19.

FIG. 22 is a schematic diagram showing the operation of the flowchart in FIG. 21. mth character 90 to mth 1,000th character 9
Assuming that the four characters of 3 are homophone pointer data,
The homophone pointer data is replaced with the first candidate Kanji code string 222 of the homophone data 221 corresponding to the number of the DISPN 114, and the character string indicated by 200 is printed.

As a result, at the time of printing, the candidate kanji characters that were displayed on the screen of the CRT display at the time of document creation are selected and printed.

As explained above, according to this embodiment, printing can be performed even if homophones remain in the document data, so even if a homophone is accidentally selected and forgotten, normal printing can be performed. It has the effect of being able to.

Furthermore, although the homophones are selected and printed when printing, the homophones remain as they are in the document data, so there is an effect that the homophones can be selected after viewing the print result.

Furthermore, even when a homophone candidate Chinese character is selected and displayed and the confirmation process is forgotten, the displayed homophone can be checked by printing, so that normal printing can be performed.

[Effects of the Invention] As described above, according to the present invention, there is an effect that normal printing can be performed even when homophones are left unselected in a document.

Further, according to the present invention, even if a homophone candidate word is selected in a document and left unconfirmed, the candidate word can be printed.

[Brief explanation of drawings]

Fig. 1 is a hardware configuration diagram of a document processing device according to an embodiment of the present invention, Fig. 2 (A) and (B) are diagrams showing an example of display operation of homophones, and Fig. 3 is a conventional example of unselected homophones. 4 is a diagram showing the structure of the document memory DBUF. FIG. 5 is a diagram showing the structure of document management data. FIG. 6 is a diagram showing the structure of document data. Figure 8 (A) shows the structure of one line of document data; Figure 8 (A) shows the structure of character data; Figure 8 (B) shows the structure of character data;
) is a diagram showing the configuration of homophone pointer data, Figure 9 is a diagram showing an example of homophone pointer data corresponding to a homophone in one line of data, Figure 10 is a diagram showing the structure of a homophone buffer, Figure 11 is a diagram showing the structure of each homophone data in the homophone buffer, Figure 12 is a flowchart showing an overview of the kana-kanji conversion process stored in the ROM, and Figure 13 is the creation of the homophone buffer in step S5. 14 is a flowchart of the homophone pointer creation process in step S7; FIG. 15 is a flowchart of the candidate character display process in step S8; FIG. 16 is a flowchart of the homophone display process in step S49; Figure 18 is a flowchart showing the outline of the homophone selection process, Figure 18 is a flowchart showing the print processing operation, Figure 19 is a flowchart showing the operation of the homophone selection process during printing in this embodiment, Figure 20 is the homophone selection process. A schematic diagram showing the operation of the selection process. FIG. 21 is a flowchart showing the operation of the homophone selection process during printing in another embodiment. FIG. 22 is a schematic diagram showing the operation of the homophone selection process in another embodiment. It is. In the figure, 1...CPU, 2...ROM, 3...RA
M. 4...DISK, 5...Keyboard, 6...DB
U.F. 7...CRT, 8...CG, 9...Printer, 1
0...Bus, 40...Document management data area, 41.
...Document data area, 42...Empty area, 43...Homophone buffer, 70...Control data section, 80...Character data, 81...Homophone pointer data, 85...
・Control byte, 86...Homophone pointer, 87...
Homophone number, 110... In-use flag, 111...
IKL, 112...ICL, 113...ILNl
114...DISPN. 2゜(A) I 3rXJ Fig. 4 4゜Fig.

Claims (2)

[Claims] (1) an input means for inputting document information; a conversion means for performing kana-kanji conversion by referring to a dictionary section based on the reading information input from the input means; and a storage means for storing the document information; A document processing device comprising: printing means for printing out the document information; determining means for determining a priority order of homophones corresponding to the reading information; and extracting the homophones from the dictionary section according to the priority order. a determining means for selecting and determining one of the homophones; and means for printing the highest priority homophone when there is an undetermined homophone in the document information during printing; A document processing device comprising: (2) an input means for inputting document information; a conversion means for performing kana-kanji conversion by referring to a dictionary section based on the reading information input from the input means; and a printing means for printing out the document information. A document processing device comprising: a display means for extracting and displaying a homophone corresponding to the reading information; a selection means for selecting one of the homophones; and a document processing device for determining the selected homophone. 1. A document processing apparatus, comprising: a storage means for storing, when an undetermined homophone exists in the document information at the time of printing, a homophone selected by the selection means is printed.