JP2001109740A - Device and method for preparing chinese document - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve input efficiency by inputting a pinyin character string from a keyboard as in the conventional way and also inputting voice information by using voice. SOLUTION: A pinyin character string inputting means inputs the pinyin character string and also, a voice information inputting means inputs voice information of Kanji (Chinese character). A voice information analyzing means analyzes the inputted voice information and acquires tone information, and a pinyin converting means performs pinyin conversion on the basis of the tone information acquired by the voice information analyzing means and the pinyin character string inputted from the pinyin character string inputting means and acquires a conversion candidate.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention improves Chinese input operability by performing a Pinyin conversion using a Pinyin character string input from a keyboard and tone information input by voice when a Chinese document is input. And a Chinese document creation method.

[0002]

2. Description of the Related Art Chinese kanji correspond to Pinyin character strings in which phonetic notations, such as Japanese kanji readings, are represented in the alphabet. As shown in FIG. 1, a Pinyin character string corresponding to one kanji character is represented by a pair of an initial and a final.
Pinyin conversion in Chinese is performed by inputting a pinyin character string “gai” from a keyboard as shown in FIG. 2, converting the input pinyin character string, and converting the corresponding conversion candidates “the, phantom,
Break, lid, outline. However, similar to the problem of homonyms in Japanese Kana-Kanji conversion, there is also the problem that multiple homonyms are output as conversion candidates in Chinese Pinyin conversion. Has been reduced. FIG. 3 shows an example of a plurality of homonyms corresponding to the Pinyin character strings “gai” and “qi”.

In order to solve this problem, there has been proposed a method of inputting tone information such as Japanese accent information in addition to a Pinyin character string and performing Pinyin conversion. There are five types of tone information from the first voice to the fifth voice, and all kanji have any type of tone information. By inputting the type of tone information in addition to the Pinyin character string and performing Pinyin conversion, the number of conversion candidates output as conversion results can be reduced as shown in FIG.

As a method of inputting the type of tone information and performing pinyin conversion, see Japanese Patent Application Laid-Open No. Hei 8-18539.
No. 6 had been proposed. However, in this conventional technique, a number indicating the type of tone information (hereinafter referred to as a tone number) is input from a keyboard (that is, 1 to 5 corresponding to the first to fifth voices, which are the types of tone information). (For inputting numbers), it is necessary to input the tone number in addition to the input of the Pinyin character string from the keyboard, and there has been a problem that the key operation of the user becomes complicated. Furthermore, most users do not correctly recognize what kind of tone information the kanji has, and the method of inputting the tone number along with the pinyin character string from the keyboard reduces input speed, The current situation is that the accuracy is reduced.

[0005]

As described above, in the prior art, when performing pinyin conversion, a method of inputting a tone number along with a pinyin character string from a keyboard is employed. An extra number must be entered, and the input operation is complicated. In addition, if the tone number is erroneously input, only the conversion candidates that are not desired are output, so it is necessary to input the pinyin character string and the tone number again, so that the type of the tone information cannot be correctly recognized. For the user, there is a problem that the accuracy is lowered.

The present invention has been made in consideration of the above-mentioned circumstances, and it is possible to improve the input efficiency by inputting a Pinyin character string from a keyboard as in the past and inputting tone information using voice. It is an object to provide a Chinese document creation device and a Chinese document creation method.

[0007]

A pinyin character string is input from the pinyin character string input means, and voice information is input from the voice information input means. Voice information is analyzed by voice information analysis means, and tone information corresponding to the Pinyin character string input from the Pinyin character string input means is obtained. The Pinyin conversion means refers to the Pinyin conversion dictionary based on the Pinyin character string input from the Pinyin character string input means and the tone information obtained by the voice information analysis means, and acquires a corresponding Chinese character. The display means displays the kanji obtained by the pinyin conversion means as conversion candidates.

According to such a Chinese document creating apparatus and a Chinese document creating method, since tone information can be input by inputting voice information, key input is performed only when a Pinyin character string is input. Therefore, in pinyin conversion using tone information, an input operation can be easily and quickly performed.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. (First Embodiment) In the first embodiment, a pinyin character string is input by a key, tone information is input by voice input, and pinyin conversion is performed from these two pieces of information.

FIG. 5 is a block diagram showing the configuration of the Chinese document creating apparatus according to the first embodiment. In FIG. 5, a keyboard input unit 1 is composed of a keyboard or the like, and performs various input and editing operations for creating a Chinese document, such as input of a Pinyin character string. The voice input unit 2 is configured by a microphone or the like, and is for inputting voice data. Pointing device section 3
Is composed of a pen input device, a mouse, and the like, and selects a conversion candidate obtained by pinyin conversion. The control unit 4 includes a CPU or the like, and uses key data, voice data, and pointing data passed from the keyboard input unit 1, the voice input unit 2, and the pointing device unit 3 to generate a pin-in conversion unit 5 described later. , An audio data processing unit 7, an output unit 10, and the like. The Pinyin conversion unit 5 converts the Pinyin character string input from the keyboard input unit 1 and the tone information obtained from the voice data input from the voice input unit 2 into
Referring to the Pinyin conversion dictionary 6, it is converted into Kanji, and the conversion candidate is passed to the control unit 4. The Pinyin conversion dictionary 6 contains information necessary for converting a Pinyin character string into a corresponding Chinese character. The user word data registered by the user is also stored here. The voice data processing unit 7 analyzes the voice data input from the voice input unit 2 with reference to the voice data processing dictionary 8, thereby analyzing the analysis results such as the initials, finals, and tone information of the kanji from the voice data. Acquired and the analysis result is passed to the control unit 4. The audio data processing dictionary 8 stores information necessary for analyzing input audio data. The data storage unit 9 lists a Pinyin character string input from the keyboard input unit 1, data necessary for Pinyin conversion for the input Pinyin character string, conversion candidates obtained as a result of the Pinyin conversion, and a list of the conversion candidates. It stores candidate list data to be displayed, audio data input from the audio input unit 2, data necessary for analysis processing of the input audio data, various setting information set by the user, and the like. Output unit 1
Numeral 0 is composed of a CRT, LCD, or the like, and displays a conversion result obtained by pinyin conversion, a candidate list, and the like.

Next, the operation of the pinyin conversion process according to the first embodiment will be described with reference to the flowchart of FIG. First, the control unit 4 waits for a pinyin character string to be input from the user via the keyboard input unit 1, and when the pinyin character string is input, stores the pinyin character string in the data storage unit 9 (step). S101). Subsequently, the control unit 4 waits for the user to input a desired kanji phonetic notation as voice data through the voice input unit 2 within a predetermined time after the pinyin character string is input,
When voice data is input, the voice data is stored in the data storage unit 9 (step S102). The control unit 4 determines whether or not the input of the voice data from the user has been within a predetermined time after the input of the Pinyin character string (step S103), and when the voice has been input (step S10).
3), the input voice data is output to the voice data processing unit 7, and a process of extracting tone information from the voice data is performed (step S104).

The voice data processing unit 7 analyzes the voice data passed from the control unit 4 with reference to the voice data processing dictionary 8, and outputs a kanji initial, a final base, and a kanji corresponding to the voice data.
Obtain analysis results such as tone information. The audio data processing unit 7 outputs the obtained analysis result to the control unit 4.

The control unit 4 checks the analysis result output from the audio data processing unit 7 and determines whether tone information has been correctly extracted (step S105). That is, the control unit 4 determines whether or not the character strings of the kanji initial and the final in the analysis result of the audio data match the Pinyin character string input by the user. Judge that the information was correctly extracted.

If it is determined that the tone information has been correctly extracted (YES in step S105), the control unit 4 determines the tone information obtained from the analysis result of the voice data and the Pinyin character string stored in the data storage unit 9. Is output to the Pinyin conversion unit 5 to perform the Pinyin conversion (step S106). The Pinyin conversion unit 5 includes a control unit 4
The Pinyin conversion is performed by referring to the Pinyin conversion dictionary 6 from the Pinyin character string and tone information output from, and a conversion candidate corresponding to the Pinyin character string is obtained.
At this time, the Pinyin conversion unit 5 uses the tone information to narrow down the number of conversion candidates, and prevents kanji characters that do not match the tone information from being output as conversion candidates. Pinyin converter 5
Outputs the conversion candidate obtained by performing the Pinyin conversion to the control unit 4.

The control unit 4 stores the conversion candidates output from the Pinyin conversion unit 5 in the data storage unit 9, outputs the conversion candidates to the output unit 10, and displays the conversion candidates (step S108). The user can select a conversion candidate by designating a desired conversion candidate from the conversion candidates displayed on the output unit 10 with the pointing device 3.

On the other hand, if there is no voice input within a predetermined time after the pinyin character string is input in the determination of step S103 (NO in step S103), and as a result of voice data analysis processing in the determination of step S105,
If it is determined that the tone information could not be correctly extracted (NO in step S105), the control unit 4 outputs only the Pinyin character string to the Pinyin conversion unit 5. Pinyin converter 5
Performs Pinyin conversion with reference to the Pinyin conversion dictionary 6 based on the information of only the Pinyin character string output from the control section 4 (step S107).

FIG. 7 shows a specific example when pin-in conversion is performed by the operation described above. As shown in FIG. 7, the user inputs the pinyin character string “gai” and “ga”
By inputting the voice having the tone information of the third voice of "i",
"Kai" can be obtained as the conversion candidate.

As described above, the tone information among the information necessary for the Pinyin conversion is input by voice, so that only the Pinyin character string needs to be input from the keyboard input unit 1.
Complexity of input operation can be suppressed, and pinyin conversion with a high conversion rate using tone information can be easily performed without lowering the input speed.

In the above description, an example in which a single kanji is converted into a pinyin is shown. However, a word or a sentence composed of a plurality of kanji can be collectively converted into a pinyin. That is, first, the user is caused to collectively input a pinyin character string of a word or sentence, and then to input the same word or sentence collectively by voice. From the voice of the input word or sentence, the tone information of each kanji that composes them is extracted at a time, and the pinyin character string that has been input earlier is extracted for each kanji using the tone information of each kanji. Convert.

Further, the timing at which the voice input from the user is received is set within a fixed time after the input of the Pinyin character string, but the present invention is not limited to this. For example, the voice input is valid until the next Pinyin character string is input. It is good.

It is not necessary to input all tone information input by voice for each kanji character input as a pinyin character string, and kanji with many conversion candidates or conversion error when pinyin conversion is performed is not required. You only need to enter it if you want to enter a specific expected kanji.

Further, in general, in the speech recognition processing in Chinese, there is a problem that the recognition of the initial at the start of inputting the audio data is more difficult than the subsequent final. Therefore, in the voice data analysis process of the present embodiment, there is a possibility that relatively many errors occur in the recognition result of the initial (the recognition result of the initial differs from the Pinyin character string). However, even if the recognition result of the initial is incorrect, if the recognition result of the final following the initial is correct (the recognition result of the final matches the Pinyin character string), the recognition result having these initials and the final is correct. It is also possible to use the tone information obtained from the recognition result for pinyin conversion.

(Second Embodiment) Next, a second embodiment will be described. In the above-described first embodiment, tone information is input by voice input together with a pinyin character string when performing pinyin conversion.
In the embodiment, the tone information is input by voice to the conversion candidate obtained as a result of the Pinyin conversion, so that the conversion candidate of the kanji having the tone information is preferentially displayed. That is, when the Pinyin conversion is performed using only the Pinyin character string, a kanji having a plurality of tone information is obtained as a conversion candidate as a conversion result. Therefore, in the conversion candidate display, the kanji having the tone information desired by the user is displayed. Is not always displayed preferentially. For this reason, the desired kanji tone information can be input to the conversion candidate, and the kanji conversion candidate having the input tone information is preferentially displayed.

The configuration of the Chinese document creating apparatus according to the second embodiment is the same as the configuration of the first embodiment shown in FIG. The operation of the process for changing the display order of conversion candidates according to the input of tone information according to the second embodiment will be described with reference to FIG.

Here, it is assumed that Pinyin conversion using only the Pinyin character string is performed in the Pinyin conversion section 5 and conversion candidates for Pinyin conversion have already been displayed on the display screen of the output section 10. The display of this conversion candidate is
As shown in FIG. 9, it is assumed that among the conversion candidates having a plurality of tone information, the conversion candidate having a higher display priority (here, “the”) is displayed with priority.

First, when an arbitrary position on the display screen of the output unit 10 is designated by the user via the pointing device 3, information on the designated position on the display screen of the designated output unit 10 is received (step S201). It is determined whether or not a conversion candidate has been designated by the user based on the designated position information (step S202). When it is determined that a conversion candidate has been designated by the user (Y in step S202)
ES) Next, the control unit 4 determines whether or not there is another conversion candidate for the specified conversion candidate (step S203). That is, the control unit 4 determines whether a conversion candidate to be displayed on the display screen is stored in the data storage unit 9 in place of the designated conversion candidate (conversion candidate displayed on the display screen). I do.

If it is determined that there is another conversion candidate for the conversion candidate specified by the pointing device 3 (YES in step S203), then the control unit 4
Determines whether or not the designation of the conversion candidate by the pointing device 3 has been canceled by the user (step S204).

If the designation of the conversion candidate by the user has not been canceled (NO in step S204), the control unit 4
Waits for audio data to be input from the user via the audio input unit 2, accepts the input audio data, and outputs it to the audio data processing unit 7 (step S20).
5).

The voice data processing unit 7 analyzes the voice data passed from the control unit 4 with reference to a voice data processing dictionary, and obtains analysis results (such as the initials of kanji, finals, and tone information). The analysis result is output to the control unit 4 (step S206). The control unit 4 checks the analysis result output from the audio data processing unit 7, and determines whether or not the tone information has been correctly extracted (step S207). The method of checking the analysis result is the same as the method described in the first embodiment.

As a result of checking the analysis result, when it is determined that tone information has been correctly extracted (step S207)
YES), the control unit 4 extracts a conversion candidate having the extracted tone information from the conversion candidates stored in the data storage unit 9 and extracts the conversion candidate instead of the conversion candidate currently displayed on the output unit 10. Control is performed so that the conversion candidates are switched and displayed (step S208).

On the other hand, in the judgment of step S207,
When it is determined that the tone information has not been correctly extracted (NO in step S207), the control unit 4 proceeds to step S20.
4 again, it is determined whether or not the designation of the conversion candidate by the pointing device 3 has been canceled by the user.

If it is determined in step S202 that no conversion candidate has been specified by the user (NO in step S202), there is no other conversion candidate for the conversion candidate specified by the pointing device 3 in step S203. When it is determined (NO in step S203), in the determination in step S204, it is determined that the designation of the conversion candidate by the user has been canceled (YES in step S204), the control unit 4 displays the conversion candidate by inputting tone information. The order change processing ends.

FIGS. 9 and 10 show specific examples when the pinyin conversion is performed by the operation described above. FIG.
Shows an example in which one conversion candidate to be displayed is displayed in place of the conversion candidate “the”.
The figure shows a state in which “kai” is replaced and displayed by inputting the tone information of the voice. FIG. 10 shows an example in which there are two conversion candidates to be displayed in place of the conversion candidate “the”, and by specifying the conversion candidate “the” and inputting the tone information of the fourth voice. , “Lid” and “Overview” are displayed in a list. When there are a plurality of conversion candidates, a list is displayed in this way, and the user is caused to select a desired conversion candidate using the pointing device 3.

For the conversion candidates obtained by performing the Pinyin conversion using the Pinyin character string as described above, tone information is input by voice input, and conversion candidates having desired tone information are displayed preferentially. be able to. Therefore, a list of conversion candidates obtained without considering tone information is simply displayed as a list,
As compared with a case where a desired conversion candidate is searched from among them, it is possible to allow the user to efficiently perform a conversion candidate selection operation.

(Third Embodiment) Next, a third embodiment will be described. The configuration of the Chinese document creating apparatus according to the third embodiment is the same as the configuration of the first embodiment shown in FIG.

In the third embodiment, the dictionary 6 for pinyin conversion is used.
In user dictionary for registering a word that is not registered in the dictionary as a new user word. When registering a user word, there is a case where tone information is not specified for a user word to be registered because tone information is not known. In this case, when tone information relating to the user word is input during the subsequent pinyin conversion, this tone information is automatically registered as tone information of the user word. Here, the description will be given assuming that the registration of the user words is performed on the Pinyin conversion dictionary 6, but a user dictionary for registering the user words may be provided separately from the Pinyin conversion dictionary 6. .

Next, the operation of adding tone information to a user word according to the third embodiment will be described with reference to the flowchart of FIG. Here, the words normally registered in the conversion dictionary 6 are the initial, final, and tone information (in this example, the tone number is stored) of the Pinyin character string as shown in FIG. It is assumed that the kanji corresponding to is stored in a state in which the tone information is not stored in the user word registered by the user as shown in FIG. A process performed when tone information is input during pinyin conversion for a user word in which tone information is not stored will be described below.

The processing in steps S301 to S307 is the same as the operation of the pinyin conversion processing in the first embodiment. That is, the control unit 4 receives a Pinyin character string input from the user via the keyboard input unit 1, stores the Pinyin character string in the data storage unit 9 (step S301), and then inputs the voice input unit 2 from the user. Waits for the desired kanji phonetic notation to be input as voice data via the PC, and if voice data is input within a predetermined time after the pinyin character string is input, the voice data is stored in the data storage unit 9. (Steps S302, S30
3). The voice data processing unit 7 analyzes the input voice data with reference to the voice data processing dictionary 8 and obtains the analysis results of the kanji's initial, final, tone information and the like corresponding to the audio data. The obtained analysis result is output to the control unit 4 (step S304). The control unit 4 checks the analysis result output from the audio data processing unit 7 and determines whether tone information has been correctly extracted (step S30).
5). When the control unit 4 determines that tone information has been correctly extracted, the pinyin conversion unit 5 performs pinyin conversion by referring to the pinyin conversion dictionary 6 from the pinyin character string and tone information output from the control unit 4. , The conversion candidate corresponding to the Pinyin character string is obtained (step S3).
06). On the other hand, when there is no voice input within a certain period of time after the pinyin character string is input in the determination of step S307, and when it is determined in the determination of step S305 that tone information cannot be correctly extracted as a result of the voice data analysis process In either case, the Pinyin conversion unit 5 performs the Pinyin conversion with reference to the Pinyin conversion dictionary 6 based on the information of only the Pinyin character string (step S307).

Next, the control unit 4 determines whether or not there is a user word that is the same word as the conversion candidate obtained by pinyin conversion in step S306 and for which tone information is not stored (step S308). . That is, the control unit 4
Searches the Pinyin conversion dictionary 6 to determine whether or not a user word having the same combination as the combination of the Pinyin character string (initial, final) of the conversion candidate and the Chinese character is registered, and If the user word is registered,
Further, it is determined whether tone information is registered for the user word.

As a result, when it is determined that there is a user word which is the same as the combination of the Pinyin character string and the kanji of the conversion candidate and for which tone information is not registered (step S3).
08), the control unit 4 sets the pinyin conversion dictionary 6
The tone information of the conversion candidate having the same combination as the user word is registered as the tone information of the user word (step S3).
09). FIG. 14 shows an example of the Pinyin conversion dictionary 6 after registering the tone information of the user word.

Next, the control unit 4 converts the conversion candidates obtained in step S306 or step S307 into
It is displayed on the output unit 10 (step S310). A specific example of the operation described above will be described with reference to FIG.

It is assumed that the user word “reciprocal network” has already been registered in the Pinyin conversion dictionary 6 without specifying tone information as shown in FIG. At this time, the user inputs a pinyin character string “hu lian wang”, and then “hu (fourth voice) lian (second voice) w
ang (third voice) ”, and the pinyin character string and the tone information obtained from the voice data,
Pinyin conversion is performed to obtain a conversion candidate “interlink”. After the pinyin conversion result, the pinyin character string “hu lian w
The pinyin conversion dictionary 6 is searched for the combination of “ang” and the kanji “reciprocal network”, and a user word having the same combination and without specifying tone information is searched. As shown in FIG. 13, the pinyin conversion dictionary 6 stores a user word corresponding to the above-described combination and for which tone information is not specified, so that tone information is registered in this user word. FIG. 14 shows the state of the pinyin conversion dictionary 6 in which tone information is registered for a user word. Tone information “fourth voice, second voice, third voice” for the user word “reciprocal network” is registered in the Pinyin conversion dictionary 6.

According to the above-described processing, even when tone information is not registered at the time of registering a user word, if tone information is input by voice input for the same word during the subsequent pinyin conversion, the input tone information is input. Is automatically registered as user word registration information, so that the user can easily perform a user word registration operation.

The method described in each of the above-described embodiments is a program that can be executed by a computer, for example, a storage medium such as a magnetic disk (hard disk or the like), an optical disk (CD-ROM, DVD or the like), or a semiconductor memory. And can be applied to various devices, or transmitted by a communication medium and applied to various devices. The computer that realizes the present apparatus reads the program stored in the storage medium, and executes the above-described processing by controlling the operation of the program.

[0045]

As described in detail above, according to the present invention, when performing pinyin conversion using tone information, it is possible to input tone information by inputting desired kanji voice information. Therefore, the key input only needs to be performed for the Pinyin character string, and the input operation by the user can be performed easily and quickly.

After the Pinyin conversion is performed using only the Pinyin character string, tone information is input by inputting voice information for a predetermined conversion candidate, and the desired Kanji conversion candidate is displayed preferentially. Therefore, the user can easily and quickly select a conversion candidate.

Further, the tone information input by the speech information at the time of the pinyin conversion is registered as the tone information of the user word registered in the user word dictionary. It is not necessary to register in the word dictionary, and the convenience in registering user words is improved.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of correspondence between kanji and pinyin character strings.

FIG. 2 is a diagram showing an operation when pinyin conversion is performed using only a pinyin character string;

FIG. 3 is a diagram illustrating a kanji having the same Pinyin character string.

FIG. 4 is a diagram showing tone information of Chinese characters having the same Pinyin character string.

FIG. 5 is a diagram showing a configuration of a Chinese document creating apparatus according to the present invention.

FIG. 6 is a flowchart illustrating an operation of a pinyin conversion process according to the first embodiment;

FIG. 7 is a diagram showing a specific example of a pinyin conversion process according to the first embodiment;

FIG. 8 is a flowchart showing an operation of a display order conversion process of conversion candidates by input of tone information according to the second embodiment;

FIG. 9 is a diagram showing a first specific example of a display order conversion process of conversion candidates by input of tone information according to the second embodiment;

FIG. 10 is a diagram showing a second specific example of the display order conversion process of conversion candidates by inputting tone information according to the second embodiment;

FIG. 11 is a flowchart showing the operation of processing for adding tone information to a user word according to the third embodiment of the present invention.

FIG. 12 is a diagram showing a storage example of normal words stored in the conversion dictionary 6.

FIG. 13 is a diagram showing a storage example of user words to which tone information is not added, which is stored in the conversion dictionary 6.

FIG. 14 is a diagram showing a storage example of a user word to which tone information stored in a conversion dictionary 6 is added.

FIG. 15 is a diagram showing a specific example of a process of adding tone information to a user word according to the third embodiment of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 keyboard input unit 2 voice input unit 3 pointing device 4 control unit 5 pinyin conversion unit 6 pinyin conversion dictionary 7 voice data processing unit 8 voice data processing dictionary 9 data storage unit 10 output Department

Claims (6)

[Claims] 1. A pinyin character string input means for inputting a pinyin character string, a pinyin conversion dictionary for storing a combination of a pinyin character string and tone information and a kanji corresponding to the combination in association with each other; Voice information input means for inputting the kanji voice information corresponding to the pinyin character string input from, and the voice information input from the voice information input means is analyzed, and the pinyin character input from the pinyin character string input means is analyzed. Voice information analysis means for obtaining tone information corresponding to a column; and the pinyin conversion dictionary based on the pinyin character string input from the pinyin character string input means and the tone information obtained by the voice information analysis means. And converting the kanji obtained by the pinyin conversion means to obtain the corresponding kanji Chinese document creating apparatus characterized by comprising a display means for displaying as a complement. 2. A Pinyin character string input means for inputting a Pinyin character string, a Pinyin conversion dictionary for storing a combination of a Pinyin character string and tone information and a Chinese character corresponding to the combination in association with each other; Pinyin conversion means for referring to the Pinyin conversion dictionary based on the Pinyin character string input from the column input means to obtain corresponding kanji and tone information, and displaying the kanji obtained by the Pinyin conversion means as conversion candidates A voice information inputting means for inputting voice information of a desired kanji, and analyzing voice information inputted from the voice information inputting means, and corresponding to a pinyin character string inputted from the pinyin character string inputting means. Voice information analyzing means for obtaining tone information; and a conversion candidate obtained by the pinyin conversion means, Control means for controlling display of conversion candidates by the display means so as to preferentially display, as conversion candidates, kanji having tone information obtained by the voice information analysis means, . 3. A Pinyin character string input means for inputting a Pinyin character string, a Pinyin conversion dictionary for storing a combination of a Pinyin character string and tone information and a Chinese character corresponding to the combination in association with each other; Voice information input means for inputting the kanji voice information corresponding to the pinyin character string input from, and the voice information input from the voice information input means is analyzed, and the pinyin character input from the pinyin character string input means is analyzed. Voice information analysis means for obtaining tone information corresponding to a sequence, and the pinyin conversion dictionary based on the pinyin character string input from the pinyin character string input means and the tone information obtained by the voice information analysis means. Pinyin conversion means for referring to and acquiring the corresponding kanji, and a Pinyin character string and voice for a newly registered user word Searching the user word dictionary, a user word with the same kanji acquired Chinese character by the pinyin conversion means from the user word dictionary for storing in association with Chinese character corresponding to the combination and the combination with the information, respectively,
And a user word registration unit for registering tone information acquired by the speech information analysis unit as tone information of the searched user word. 4. A Chinese character having a Pinyin character string input means for inputting a Pinyin character string, a voice information input means for inputting Chinese character voice information, and a display means for displaying a Chinese character corresponding to the Pinyin character string as a conversion candidate. In the Chinese document creation method in the word document creation device, a Pinyin character string input from the Pinyin character string input means is received, voice information input from the voice information input means is analyzed, and the input Pinyin character string is input. The tone information corresponding to the obtained Pinyin character string and the acquired tone information are referred to a Pinyin conversion dictionary that stores the combination of the Pinyin character string and the tone information and the kanji corresponding to the combination respectively. A kanji corresponding to the received pinyin character string is acquired based on tone information, and the acquired kanji is used as a conversion candidate. Chinese document creating apparatus characterized by controlling so as to be displayed on said display means. 5. A Chinese character having a Pinyin character string input means for inputting a Pinyin character string, voice information input means for inputting Chinese character voice information, and a display means for displaying a Kanji character corresponding to the Pinyin character string as a conversion candidate. In the Chinese document creation method in the word document creation device, a Pinyin character string input from the Pinyin character string input unit is received, and a combination of the Pinyin character string and tone information based on the received Pinyin character string and the combination A kanji and tone information corresponding to the received pinyin character string are obtained by referring to a pinyin conversion dictionary that stores kanjis corresponding to the corresponding kanjis, and the obtained kanjis are displayed as conversion candidates on the display means. Controlling, and analyzing the voice information input from the voice information input means after displaying the conversion candidate, and The tone information corresponding to the obtained Pinyin character string, and a conversion candidate by the display means for preferentially displaying a kanji having the same tone information as the acquired tone information among the conversion candidates obtained by the Pinyin conversion as a conversion candidate. Controlling the display of a Chinese document. 6. A Pinyin character string input means for inputting a Pinyin character string, voice information input means for inputting Chinese character voice information, a display means for displaying a Chinese character corresponding to the Pinyin character string as a conversion candidate, and a user word In a Chinese document creation method in a Chinese document creation device having a user word dictionary capable of registering, a Pinyin character string input from the Pinyin character string input means is received, and voice information input from the voice information input means is received. To obtain the tone information corresponding to the input Pinyin character string, and refer to a Pinyin conversion dictionary that stores the combination of the Pinyin character string and the tone information and the kanji corresponding to the combination in association with each other. A Chinese character corresponding to the received pinyin character string based on the received pinyin character string and the acquired tone information. A character is obtained, a user word having the same kanji as the obtained kanji is searched from the user word dictionary, and tone information obtained by voice information analysis is registered in the user word dictionary as tone information of the searched user word. An apparatus for creating a Chinese document.