JP2001282778A - Device and method for processing information, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize speed-up and simplification of input processing of character string. SOLUTION: A voice recognizing part 31 recognizes a voice inputted by a microphone 11 and outputs the input character string candidates of a recognized character string to a predictive example part 35. Concerning the character string of highest possibility among the input character string candidates, the predictive example part 35 retrieves an input sentence data base stored in an HDD 35, and a character string following the retrieved character string is detected as a predictive example and displayed via a display processing part 37 onto an LCD 15. By selecting the predictive example, a user can input the character string and further, by similarly selecting the displayed predictive example concerning the selected character string, the user continuously inputs character strings.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing apparatus, method, and recording medium, and more particularly, to an information processing apparatus, method, and recording medium capable of realizing high-speed and simplified character string input operation. About.

[0002]

2. Description of the Related Art Techniques for speeding up and simplifying the input of character strings are becoming widespread. Recent advances in computer technology have led to miniaturization of portable computers. However, with the progress of miniaturization of computers, keyboards, which have conventionally been the mainstream of computer input means, have also been miniaturized. For this reason, as the size of the keyboard is reduced, it is difficult to operate the keyboard with the size of a human hand. Therefore, in order to solve this problem, a pen-input type computer using a pen having a handwritten character recognition function as an input means in addition to a keyboard has been developed.

Further, Japanese Patent Application Laid-Open Nos. 10-154033, 10-154144, and 11-16756
In step 9, the history of the character strings input so far is recorded as a database, and based on this database, a predicted character string that is likely to be continuously input is displayed, and a desired character is displayed from the displayed predicted character string. It has been proposed by the present applicant that the input of a character string is speeded up and simplified by selecting a column.

[0004]

However, in the above configuration, there is a problem that the character string at the beginning of the sentence needs to be input by a keyboard or a pen as in the past. In addition, since a character string that may be continuously input may not be predicted, there is also a problem that, at that time, the character string must be input using a keyboard, a pen, or the like.

The present invention has been made in view of such a situation, and allows a character string to be input by voice, and a candidate for a subsequent character string to be input to the input character string or to a character string which has just been selected immediately before. Is displayed as a prediction example (predicted character string), and a desired character string is selected from the displayed characters, so that the speed-up and simplification of character string input can be realized.

[0006]

An information processing apparatus according to the present invention comprises an input means for inputting a voice, and a recognition character string candidate for recognizing the voice input by the input means and detecting a corresponding recognition character string candidate. Detecting means, a recognized character string candidate display means for displaying a recognized character string candidate corresponding to the voice detected by the recognized character string candidate detecting means, and a desired character string candidate displayed from the recognized character string candidate display means. Character string selecting means for selecting a character string to be recognized, and a recognition character string selected by the recognition character string selecting means or a character string candidate which is predicted to be continued from the immediately preceding character string detected The predicted character string candidate detecting means, the predicted character string candidate displaying means for displaying the predicted character string candidate detected by the predicted character string candidate detecting means, and the predicted character string candidate displaying means Characterized in that it comprises a predicted character string selecting means for selecting a desired expected string from the character string candidate measurement.

[0007] The input means may further include a key or a pen.

The predictive character string candidate detecting means is configured to search the input character string for a selected recognized character string or a character string including the immediately preceding selected character string. The character string candidate display means can be displayed as predicted character string candidates in the order of appearance frequency from the search results of the predicted character string candidate detection means.

The predictive character string candidate detecting means searches the input character string for a selected recognized character string or a character string including the immediately preceding selected character string. 2. The predicted character string candidate display unit according to claim 1, wherein the predicted character string candidate display unit sequentially displays the predicted character string candidates from the search result of the predicted character string candidate detection unit, starting from the most recently input. Information processing device.

[0010] The recognition character string selecting means may further include a key or a pen.

[0011] The predicted character string selecting means may further include a key or a pen.

An information processing method according to the present invention includes an input step of inputting a voice, a recognition character string candidate detecting step of recognizing the voice input in the processing of the input step, and detecting a corresponding recognition character string candidate. A recognition character string candidate display step for displaying a recognition character string candidate corresponding to the voice detected in the processing of the recognition character string candidate detection step; and a recognition character string candidate displayed in the recognition character string candidate display step. Character string selection step of selecting a character string to be recognized, and a recognition character string selected in the processing of the recognition character string selection step, or a character string candidate that is predicted to follow the previous already selected character string A predicted character string candidate detecting step of detecting a predicted character string candidate detected in the processing of the predicted character string candidate detecting step, Characterized in that it comprises a prediction string selection step of selecting a desired expected string from the prediction string candidates displayed in the process of measuring the string candidate display step.

The program of the recording medium according to the present invention comprises an input control step for controlling the input of a voice, and a recognition for recognizing the voice input in the processing of the input control step and controlling the detection of a corresponding recognition character string candidate. A character string candidate detection control step, a recognized character string candidate display control step of controlling display of a recognized character string candidate corresponding to the voice detected in the processing of the recognized character string candidate detection control step, and a recognized character string candidate display A recognition character string selection control step of controlling selection of a desired recognition character string from recognition character string candidates displayed in the control step processing, and a recognition character string selected in the recognition character string selection control step or A predicted character string candidate detection control step for controlling detection of a character string candidate predicted to follow the already selected character string of A predicted character string candidate display control step for controlling the display of the predicted character string candidates detected in the processing of the loop, and a desired predicted character string from the predicted character string candidates displayed in the processing of the predicted character string candidate display control step. Predicting character string selection controlling step of controlling selection.

In the information processing apparatus and method and the recording medium according to the present invention, a voice is input, the input voice is recognized, a candidate for a corresponding character string is detected, and a recognition corresponding to the detected voice is performed. A character string candidate is displayed, a desired recognition character string is selected from the displayed recognition character string candidates, and the selected recognition character string or a character string that is predicted to follow the previous already selected character string Are detected, the detected predicted character string candidates are displayed, and a desired predicted character string is selected from the displayed predicted character string candidates.

[0015]

FIG. 1 is a diagram showing a configuration of an embodiment of a text input device 1 according to the present invention.

A voice uttered by the user is input to the microphone 11 of the text input device 1 at the time of inputting a character string or selecting a prediction example of the character string. Microphone 11
Turns on while the talk switch 16 is pressed,
As long as it is not pressed, it is off. Therefore, the user presses the talk switch 16 when inputting a character string,
By releasing the talk switch 16 at the end of the input, it is possible to input only the voice to be input as a character string.

The input key 12 is operated by a user at the time of inputting a character string, selecting an input candidate, or selecting an example of prediction, so that a desired character string or the like can be input. The input pen 13 is operated by the user when inputting a character string, selecting an input candidate, or selecting a prediction example, and traces a desired character string on the pressure-sensitive tablet 14 such as an input field or a selected portion on the LCD 15. It is input by touching or touching.

The talk switch 16 is a switch in which one short portion is fixed as an axis and the other short portion is movable in the direction of arrow A in the figure, and is pushed rightward in the direction of arrow A in the diagram. When the microphone 11 is turned on, the microphone 11 is turned on, and the arrow A in the figure is to the left (the state shown in the figure).
At this time, the microphone 11 is turned off.

The input selection unit 17 is composed of a tablet button or the like, and uses the microphone 11 used for input,
The input key 12 or the input pen 13 is selected and switched. Of course, the input selector 17
May be something other than a tablet button, for example, a dial method.

FIG. 2 is a block diagram showing the internal configuration of the text input device 1 of FIG. The speech recognition unit 31 is controlled by the control unit 101, and receives an HMM (Hidden Markov Model:
Hidden Markov model)
The voice input from the microphone 11 is recognized, and a candidate character string (input candidate) is output to the prediction exemplifying unit 35 based on the recognition result. The voice recognition unit 31 stores the input voice in combination with the selected character string supplied from the prediction exemplifying unit 35, and makes the voice recognition model learn.
Note that the voice recognition method may be a method other than the HMM.

The pen input unit 32 is controlled by the control unit 101. When a character string is input, an electric signal of the tablet pressure distribution traced by the input pen 13 supplied from the pressure-sensitive tablet 14 is written by handwriting characters. The recognizing unit 33 checks the input character string, and outputs the input character string candidate to the prediction example unit 35. Further, when selecting an input candidate or a prediction example, the pen input unit 32 selects the selected (pressed or traced by the input pen 13).
The signal corresponding to the portion is output to the prediction exemplifying unit 35.

The key input unit 34 is controlled by the control unit 101, and when inputting a character string, a character string candidate corresponding to the character input by the input key 12 is predicted by the prediction example unit 3.
5 is output. When selecting an input candidate or a prediction example, the key input unit 34 outputs a signal of a portion selected by moving a cursor or the like to the prediction example unit 35.

The prediction example unit 35 is controlled by the control unit 101, and stores a character string input from the voice recognition unit 31, the pen input unit 32, or the key input unit 34 in the HDD 35a. The input sentence database is searched, and a character string candidate that is predicted to be input subsequent to the input character string is searched as a prediction example, and the LCD 1 is displayed via the display unit 37 together with the input character string.
5 is displayed in the character input window 42 (FIGS. 3 to 6). In addition, the prediction exemplifying unit 35 receives the input from the voice recognition unit 31, the pen input unit 32, or the key input unit 34,
The character string selected (confirmed) from the prediction example candidates is output to the application processing unit 36. The details of the processing of the prediction exemplifying unit 35 will be described later.

The application processing unit 36 includes the control unit 1
01, which runs application software used as a so-called word processor. The application processing unit 36 outputs the selected (confirmed) character string input from the prediction example unit 35 to the display processing unit 37 in a state where the character string is input on the application software, and causes the LCD 15 to display the character string.

The display processing unit 37 is controlled by the control unit 101, processes a signal input from the prediction example unit 35 or the application processing unit 36, and causes the LCD 15 to display the processed signal.

The control unit 101 controls the entire operation of the text input device 1 and controls the RO connected to the main bus 102.
A basic program recorded in a read only memory (M) 103 is read into a random access memory (RAM) 104 as appropriate and executed. In addition, the control unit 101
A program such as an OS (Operating System) recorded in 105 is read into the RAM 104 as appropriate and executed. Further, the control unit 101 controls the magnetic disk 111, the optical disk 112, and the magneto-optical disk 1 mounted on the drive 110.
13 or a program recorded in the semiconductor memory 114 is read into the RAM 104 as appropriate and executed.

Next, the processing of the prediction example section 35 will be described. The prediction exemplifying unit 35 determines that the character string is
The content to be output to the display processing unit 37 (the content to be displayed) differs depending on whether the input is made from the pen input unit 32 or the key input unit 34.

FIG. 3 shows a case where the output content from the prediction example section 35 when a character string is input from the speech recognition section 31 is the LCD 1.
5 is an example displayed. Application display section 41
Is a portion where the content input from the application processing unit 36 is displayed. The character input window 42 is a portion in which the content input from the prediction example unit 35 is displayed.

Now, for example, when the user inputs "I am" as a voice to the microphone 11, the voice recognition unit 3
1 matches an input voice pattern with a voice recognition model, and selects “my”, “atashi no”, “akashi no”, and “takashi no” as character string candidates in descending order of possibility. It is assumed that the information is output to the prediction example unit 35. In this case, the character string candidates are ranked in the top four in descending order of the possibility.
Although it is set to the order, it may be set arbitrarily.

At this time, in the voice input display section 51, the character string candidates detected by the voice recognition section 31 are displayed in the order of the highest possibility from the top, "My", "I", and "Akashi". , And "Takashino", the input candidate display section 51a
"My", which is most likely to be an input candidate, is displayed in inverted characters.

In the prediction example display section 52 on the right side, a prediction example predicted to be continuously input to "my" which is most likely as a character string candidate displayed on the voice input display section 51 is displayed. "Name", "Schedule", "Thought", "Flex Holiday", "Next", "Report", and "Responsible" are displayed. still,
The search processing of the prediction example by the prediction example unit 35 will be described later. Further, in this case, the number of candidates for the prediction example is seven, but the number of candidates may be arbitrarily set.

The user can select “my”, “my”, or “my” displayed on the input candidate display section 51a of the voice input display section 51.
A desired content is selected from “Akashi no” and “Takashi no” by operating the input pen 13 and pressing the corresponding portion, or operating the input key 12 or the like. In this case, the prediction example display unit 52 displays a prediction example that is subsequently input to “my”, which is most likely to be an input candidate and is displayed as an inverted character at the top of the input candidate display unit 51a. Is displayed, but when other items are selected, the display content of the prediction example display section 52 changes in accordance with the selected character string.

The selection of the character string displayed on the input candidate display section 51a is to determine whether or not the voice recognition has been correctly performed in the case of the voice-recognized character string. It is necessary to use the input keys 12 or the input pen 13 other than (input). However,
In this case, since "my" is a desired character string, when the following character string (predictive example) is selected, "my" is also selected. In this case, it is not necessary to use a method other than the voice input.

Also, the user can display the "name is", "planned", "idea", "flex holiday", "next", "report" displayed on the prediction example display section 52, and A desired one from “charge” may be selected by voice input from the microphone 11, by operating the input pen 13, by pressing the corresponding portion, or by operating the input key 12. Alternatively, other character strings may be input by operating the microphone 11, the input keys 12, or the input pen 13.

For example, when "Name is selected", the prediction example unit 35 causes the application display unit 41 to display the input "My name is", as shown in FIG. Further, the prediction exemplifying unit 35 causes the immediately preceding input display unit 61 to display that “name is” has been selected as the immediately preceding input, instead of the voice input display unit 51. Further, the prediction example section 35 searches for a prediction example subsequent to “name is” and causes the prediction example display section 52 to display the prediction example. In the example of FIG. 4, “I am Sato”, “I am Kato”, “I am Suzuki”, “What is it”, “No”, “Difficult”, or “I do not know” is displayed in the prediction example display section 52. ing.

In the state shown in FIG. 4, the user displays “I am Sato”, “I am Kato” displayed on the prediction example display section 52,
From "Suzuki,""What,""No,""Difficult," or "I don't know," you can either use the microphone 11 to voice-input the desired character string, or press the input pen 13 to press that part. Or input key 1
2 may be selected, for example, or other character strings may be input by operating the microphone 11, the input keys 12, or the input pen 13.

FIG. 5 shows an example in which the output content from the prediction example unit 35 when a character string is input from the pen input unit 32 is the LCD1.
5 is an example displayed. Instead of the voice input display unit 51, a pen input display unit 71 is displayed, and further, an input field 71b is displayed therein. The user enters the input field 71
A desired character is input on the “b” by tracing a desired character with the input pen 13. Now, the input column 71b is in the process of inputting the radical "He", and is in a state of stroking "Thousands". At this time, the pen input unit 32 collates with the character data of the handwritten character recognition unit 33 based on the signal input from the pressure-sensitive tablet 14, and in this case, “I” and “ , "Interest" and "family"
Is output to the prediction example unit 35. The prediction example unit 35 includes:
The characters are displayed on the input candidate display section 71a on the LCD 15 via the display processing section 37. In this example, up to the top four input candidates are displayed on the input candidate display section 71a, but the number of input candidates displayed may be arbitrarily set.

The user, if any desired character is displayed in the input candidate display section 71a, puts it into the input pen 13,
Alternatively, selection may be made with the input key 12, or if there is no desired character, the input may be continued in the input field 71b. Regarding the selection of the prediction example in the prediction example display unit 52,
Same as above.

FIG. 6 shows an example in which the output content from the prediction example section 35 when a character string is input from the key input section 34 is the LCD 1.
5 is an example displayed. A key input display section 81 is displayed in place of the pen input display section 71, and an input field 81b is displayed therein. The user enters the information in this input field 81
A desired character is input on the key b with the input keys 12. Now, "wa" has been entered in the entry field 81b. At this time, the key input unit 34 matches the character data based on “wa” input from the input key 12, and based on the result, as input candidates, “I”, “Sum”, “Ring”, and “Wa” is displayed on the input candidate display section 81a. still,
In this example, up to the top four input candidates are displayed on the input candidate display section 81a, but the number of input candidates displayed may be arbitrarily set.

If the user has a desired character displayed on the input candidate display section 81a, the user presses the input key 12,
Alternatively, selection may be made with the input pen 13 or, if not, input may be continued on the input field 81b. The selection of a prediction example in the prediction example display unit 52 is the same as described above.

The microphone 11, the input keys 12,
Alternatively, which of the input pens 13 is used to input a character string can be switched by operating the input selection unit 17 as appropriate. Also, the talk switch 16
Is turned on, the input by the microphone 11 is performed. When the input key 12 is operated, the input key 1 is
2. When the input pen 13 touches the pressure-sensitive tablet 14, the input pen 13 may be used to enable input.

Next, referring to the flowchart of FIG.
The text input process of the text input device 1 will be described. still,
In this example, it is assumed that the microphone 11 is basically used for inputting a character string.

In step S1, speech recognition processing
Be executed.

Here, the input recognition processing will be described with reference to the flowchart of FIG. In step S21, the voice recognition unit 31 determines whether or not the talk switch 16 is turned on. If it is determined that the talk switch 16 is not turned on, the voice recognition unit 31 waits for an input until the talk switch 16 is turned on. If it is determined that there is, the process proceeds to step S22.

In step S22, the voice recognition unit 31
Starts a process of recording voice input from the microphone 11 as voice data. In step S23, the voice recognition unit 31 determines whether or not the talk switch 16 is turned off. If it is determined that the talk switch 16 is not turned off, the voice recognition unit 31 continues to record the voice input from the microphone 11, and determines that the voice is turned off. If it is determined, the process proceeds to step S24.

In step S24, the voice recognition unit 31
Ends the recording of the audio data. In step S25, the voice recognition unit 31 performs a recognition process on the recorded data, and generates a corresponding candidate. Step S2
In 6, the speech recognition unit 31 outputs the character string selected as a candidate to the prediction example unit 35.

Here, the description returns to the flowchart of FIG.

In step S2, the prediction example unit 35
Determines whether the input character string is a character string such as “input OK” registered in advance as a character string notifying the end of the input, and if it is “input OK”,
Terminates the text input process. When it is not “input OK”, the prediction exemplifying unit 35 determines whether or not the input character string candidate is included in the prediction exemplification. That is, whether the input character string is a newly input character string,
It is determined whether the input is a character string for selecting a candidate displayed as a prediction example.

If the input character string candidate is not included in the prediction example, in step S4, the prediction example unit 35 executes a search process of the prediction example.

Here, a search process of a prediction example will be described with reference to a flowchart of FIG.

For example, it is assumed that a sentence as shown in FIG. 10 has been input to the input sentence database of the HDD 35a.
Each sentence in the input sentence database, for each unit entered,
The data is recorded in sections, and is managed for each period by the time when the recording was started (each square in the text indicates the input unit). In this example, "I'm going to tell the story of a naked king"
"2000/1/14 17:08:06" states, "The king of a village was called a naked king by the villagers."
30 "," Because the king of this village "," 2000/1/14 1
7:13:06 "indicates that recording has begun.

Now, "King" surrounded by a bold line on the third line in the figure.
When searching for a prediction example following the character string, the prediction example unit 35 adds the input “king” and the already input character string in step S31 to form a character string for three input units, for example. Use to generate search terms. In other words, in this example, "the king of this village"
(= “This” + “village” + “king”) is the search target word. Therefore, the prediction exemplifying unit 35 sets the sentence already input in the input sentence database as the search target word “King of this village”.

In step S32, the prediction example section 35
Clears its own buffer that stores the retrieved prediction examples.

In step S33, the prediction example section 35
Searches for the search term "king of this village" from the front of the sentence recorded in the input sentence database.

In step S34, the prediction example section 35
Records the search result of the search target word. In the example of FIG. 7, the search target word “King of this village” is not included, so that there is no search result.

In step S35, the prediction example section 35
Determines whether or not the prediction examples are detected by the preset number of prediction example candidates. For example, if the number of candidates for the prediction example is set to three, no process is detected (the number is less than the set number of candidates), and the process proceeds to step S36. . Note that the number of candidates for the prediction example may be arbitrarily set.

In step S36, the prediction example section 35
Generates a search target word from which one input unit (“this”) of the search target word has been deleted. In other words, "the king of this village"
Means three input units ("this" + "village" + "king")
, The "king of the village" (= "village" +) consisting of two input units including the input character string
“King”) is generated as a search target word.

In step S37, the prediction example section 35
Determines whether the search target word has been deleted by deleting one input unit. In this case, the search target word exists as "King of the village",
It returns to step S33.

In step S33, the prediction example section 35
Searches forward in the input sentence database again with the search target word “King of the village”. In the example of FIG. 10, “King of the village” in a few input units from the left in the second row in the figure is detected.

In step S34, the prediction example section 35
Detects "ha" following this as a prediction example and stores it in the buffer.

In step S35, the prediction example unit 35
Determines whether the number of candidates for the prediction example searched is less than three. In this case, the number of predicted examples searched is one and less than three, so the processing is performed in step S36.
Proceed to. In step S36, the prediction exemplifying unit 35
Further, the input unit of the search target word is reduced, and only the immediately preceding input character string, that is, only “king” is set as the search target word.

In step S36, the prediction example section 35
Determines whether the search target word has been deleted by deleting one input unit. In this case, since the search target word exists as “king”, the process returns to step S33.

In step S33, the prediction example section 35
Searches the input sentence database using "king" as a search target word, and detects a predicted example following the searched portion. In step S34, the prediction exemplifying unit 35 searches for “NO” following “King” of the third input unit in the first row in the example of FIG. 10 and searches for “NO” in the seventh input unit of the second row. The subsequent "to" is detected and stored in the buffer. still,
The "ha" following the "king" of the third input in the second line is not detected because it is already recorded in the buffer.

In step S35, the prediction example section 35
Determines whether the number is less than three, which is the preset number of candidates for the prediction example. In this case, since three prediction examples of “ha”, “to”, and “no” have been detected as the prediction examples of “king”, the search result is not less than the preset three. Therefore, the process proceeds to step S38.

In step S38, the prediction example unit 35
Is "ha", "to", "no" stored in the buffer
Are output to the display processing unit 37, and the
5 is displayed on the prediction example display section 52, and the process returns to the original processing.

In step S37, for example, when there is no more search target word, the process proceeds to step S38. That is, the prediction example is displayed in a state where the number of candidates detected by the search is smaller than the set number.

Here, the description returns to the flowchart of FIG.

In step S5, the prediction exemplifying unit 35
Determines whether the input character string has a plurality of character string candidates. That is, in the case where the character string selected from the candidates of the prediction example is input, or in the case where the character string selected from the input candidate display unit 51a is input in step S6 described later,
There is only one input character string. On the contrary, the input character string is the microphone 11, the input key 12, or
When a new input is made from the input pen 13, the user has a plurality of character string candidates. Therefore, at this time, if an input candidate has been selected, it is only necessary to select a prediction example. If there are a plurality of input candidates, it is necessary to select an input candidate and select a prediction example.

In step S5, when there are a plurality of character string candidates, that is, a character string newly input from the microphone 11, the input key 12, or the input pen 13 (selection of input candidates and prediction example). If it is necessary to select), in step S6, the prediction example unit 35
Determines whether any of the input candidates other than the first is selected. That is, a prediction example is already shown for the first character string of the input candidate, but if any other character string is selected, the prediction example for the selected character string must be performed again. . Therefore, step S
In step 6, if a character string other than the first character string is selected from the input candidate character strings, the process returns to the processing of step S4, and a search for a prediction example of a newly selected input character string other than the first character string is performed. Processing will be executed.

If it is determined in step S6 that a character string other than the first character string of the input candidate character string has not been selected (no new input candidate has been selected), the process proceeds to step S7.

In step S7, the prediction exemplifying unit 35
Transmits the selected input character string to the application processing unit 3
6 to be displayed on the application display unit 41,
It returns to the process of step S1. That is, this process is repeated until "input OK" is recognized in step S2.

In step S3, when the input candidate is included in the prediction example, that is, it is determined that the character string displayed on the prediction example display section 52 has been input (selected) as the prediction example. In this case, the process proceeds to step S8.

In step S8, the prediction exemplifying unit 35
Outputs the character string selected from the prediction example to the application processing unit 36 and causes the application display unit 41 to display the character string.

In step S5, if there is not a plurality of input character strings, that is, whether a character string selected from the prediction examples has been input, or by the processing in step S6, one of the input candidates has already been input. If the character string is selected as something other than the place (only selection of a prediction example is necessary), the process returns to step S1, and the subsequent processes are repeated.

When many prediction examples are detected, the prediction example section 35 selectively outputs the prediction examples in descending order of appearance frequency, or outputs the prediction examples in the order of temporally recent input. It may be.

In the above-described sentence input processing, when a character string is input only with the input key 12 or the input pen 13, and when the input key 12 is used, the processing of the key input unit 34 is performed in the flowchart shown in FIG. A similar process is realized by replacing the process of step S1 with a process of outputting a character string input by the input key 12 to the prediction exemplifying unit 35. Further, when using the input pen 13, the pen input unit 32 performs the processing of step S 1 in the flowchart of FIG. 7 by inputting a character to the handwritten character recognition unit 33 based on the pressure distribution input from the pressure-sensitive tablet 14. A similar process is realized by recognizing a column and replacing the process of outputting the recognition result to the prediction exemplifying unit 35.

The processing using the input key 12 or the input pen 13 is described in Japanese Patent Application Laid-Open Nos. 10-154033 and 10-1541 which were previously proposed by the present applicant.
44 or JP-A-11-167569.

As described above, speeding up and simplification of character string input can be realized by voice input and prediction examples.

The above-described series of processing can be executed by hardware, but can also be executed by software. When a series of processing is executed by software, a program constituting the software may be executed by a computer built into dedicated hardware or by installing various programs to execute various functions. It is installed from a recording medium to a possible general-purpose personal computer or the like.

This recording medium is provided to the user in a state of being incorporated in the text input device 1 in advance as shown in FIG. 2, and is provided not only to the HDD 105 in which the program is recorded but also to the user separately from the computer. A magnetic disk 111 (including a floppy (registered trademark) disk) storing a program, an optical disk 112 (CD-ROM (Compact Disk-Read)
Only Memory), a DVD (including a Digital Versatile Disk), a magneto-optical disk 113 (including an MD (Mini-Disk)), or a package medium including a semiconductor memory 114 (including a Memory Stick).

In this specification, the step of describing a program recorded on a recording medium is not limited to processing performed in chronological order in the described order, but is not necessarily performed in chronological order. , And processing executed in parallel or individually.

[0082]

According to the information processing apparatus, method, and recording medium of the present invention, an input voice is recognized, a candidate for a corresponding recognized character string is detected, and a recognized character string candidate corresponding to the detected voice is recognized. Is displayed, a desired recognized character string is selected from the displayed recognized character string candidates, and a selected recognized character string or a character string candidate that is predicted to follow the immediately preceding already selected character string is detected. Since the detected predicted character string candidates are displayed and a desired predicted character string is selected from the displayed predicted character string candidates, the speed and simplification of character string input are realized.

[Brief description of the drawings]

FIG. 1 is a schematic view of a text input device to which the present invention is applied.

FIG. 2 is a block diagram of a text input device to which the present invention is applied.

FIG. 3 is a diagram illustrating a display example of a text input device.

FIG. 4 is a diagram illustrating a display example of a text input device.

FIG. 5 is a diagram illustrating a display example of a text input device.

FIG. 6 is a diagram illustrating a display example of a text input device.

FIG. 7 is a flowchart illustrating the operation of the text input device.

FIG. 8 is a flowchart illustrating an input recognition process.

FIG. 9 is a flowchart illustrating a search process.

FIG. 10 is a diagram illustrating a search process.

[Explanation of symbols]

1 sentence input device, 11 microphone, 12 input keys, 13 input pen, 14 pressure-sensitive tablet, 15
LCD, 16 talk switch, 17 input selector, 3
1 voice recognition unit, 31a HDD, 32 pen input unit, 3
3 Handwritten character recognition unit, 34 key input unit, 35 prediction example unit, 35a HDD, 36 application processing unit, 37 display processing unit, 41 application display unit, 42 character input window, 51 voice input display unit, 51a input candidate display unit , 52 prediction example display section,
61 last input display section, 71 pen input display section, 71a
Input candidate display section, 71b input field, 81 key input display section, 81a input candidate display section, 81b input field

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G10L 15/22 G10L 3/00 561C 561E

Claims (8)

[Claims] 1. An input unit for inputting a voice, a recognized character string candidate detecting unit for recognizing a voice input by the input unit and detecting a corresponding recognized character string candidate, and a recognized character string candidate detecting unit A recognition character string candidate display means for displaying a recognition character string candidate corresponding to the voice detected by the above, and selecting a desired recognition character string from the recognition character string candidates displayed by the recognition character string candidate display means Recognition character string selecting means, and the recognized character string selected by the recognition character string selecting means, or a predicted character string candidate for detecting a candidate of a character string predicted to be continued from the immediately preceding selected character string Detecting means, predicted character string candidate display means for displaying a predicted character string candidate, detected by the predicted character string candidate detecting means, and displayed by the predicted character string candidate display means An information processing apparatus comprising: a predicted character string selecting unit that selects a desired predicted character string from the predicted character string candidates. 2. The information processing apparatus according to claim 1, wherein the input unit further includes a key or a pen. 3. The predictive character string candidate detecting means, wherein the selected recognized character string is selected from character strings already input,
Alternatively, a search is performed using a character string that includes the character string that has already been selected immediately before, and the predicted character string candidate display means predicts the character strings in order of appearance frequency from the search results of the predicted character string candidate detection means. 2. The information processing apparatus according to claim 1, wherein the information is displayed as a character string candidate. 4. The predictive character string candidate detecting means, wherein the selected recognized character string is selected from character strings already input,
Alternatively, a search is performed using a character string that includes the character string that has just been selected immediately before, and the predicted character string candidate display means includes a search result of the predicted character string candidate detection means, from a search result that is temporally recently input. 2. The information processing apparatus according to claim 1, wherein the information is sequentially displayed as a predicted character string candidate. 5. The information processing apparatus according to claim 1, wherein the recognition character string selection unit further includes a key or a pen. 6. The information processing apparatus according to claim 1, wherein the predicted character string selecting unit further includes a key or a pen. 7. An input step of inputting a voice, a recognition character string candidate detecting step of recognizing the voice input in the processing of the input step, and detecting a corresponding recognition character string candidate, and the recognition character string candidate. Detected in the processing of the detection step,
A recognition character string candidate displaying step of displaying a recognition character string candidate corresponding to the voice, and a recognition character string for selecting a desired recognition character string from the recognition character string candidates displayed in the processing of the recognition character string candidate display step A selection step, a predicted character string candidate detection for detecting the recognized character string selected in the processing of the recognized character string selection step or a character string candidate predicted to be continued from the immediately previously selected character string Step, detected in the processing of the predicted character string candidate detection step,
A predicted character string candidate displaying step of displaying a predicted character string candidate; and a predicted character string selecting step of selecting a desired predicted character string from the predicted character string candidates displayed in the processing of the predicted character string candidate displayed step. An information processing method, comprising: 8. An input control step of controlling input of a voice, and a recognition character string candidate detection control step of recognizing the voice input in the processing of the input control step and controlling detection of a corresponding recognition character string candidate. A recognition character string candidate display control step of controlling display of a recognition character string candidate corresponding to the voice detected in the processing of the recognition character string candidate detection control step; and a processing of the recognition character string candidate display control step A recognition character string selection control step of controlling selection of a desired recognition character string from the recognition character string candidates displayed in the above, and the recognition character string selected in the processing of the recognition character string selection control step, or A predictive character string candidate detection control step for controlling detection of a character string candidate predicted to follow the already selected character string; A predicted character string candidate display control step of controlling the display of predicted character string candidates detected in the processing of the loop, and a desired predicted character from the predicted character string candidates displayed in the predicted character string candidate display control processing. A predictive character string selection control step of controlling selection of a column. A recording medium storing a computer-readable program.