JP2000163427A - Processor and method for information processing and provision medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain the dictionary main sentence of a word, which is inputted by being voiced, accurately at a high speed. SOLUTION: A speech recognition part 42 recognizes a voice inputted from a microphone by referring to a recognition object word storage part 43 and determines N speech-recognized word candidates. In a recognition object word storage part 43, about 5000 words (less than the words collected in a general Japanese dictionary) are registered. An index word detection part 44 detects the index words corresponding to the speech-recognized word candidates by referring to a detection object storage part 45. In the detection object word storage part 45, words more than the words registered in the recognition object word storage part 43, e.g. tens of thousands of words (nearly as many as the words collected in a general Japanese dictionary) are registered. A dictionary main sentence detection part 46 detects the dictionary main sentences corresponding to the index words by referring to a dictionary main sentence storage part 47. The dictionary main sentence storage part 47 is stored with dictionary main sentences (explanatory sentence) of the words registered in the detection object word storage part 45.

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 and method, and a providing medium, and more particularly, to an information processing apparatus and method capable of accurately and quickly obtaining information corresponding to designated information input by voice. Information processing apparatus and method, and a providing medium.

[0002]

2. Description of the Related Art There has been developed an electronic dictionary device capable of pronouncing a dictionary text (explanation) instead of inputting a word to be searched by operating a keyboard and inputting it by voice. In this electronic dictionary device, a built-in ROM
The dictionary body (description) of the word input by voice is read from a Japanese language dictionary, a Japanese-English dictionary, or the like stored in, for example, and displayed on the display unit. In this way, by inputting by voice, the user's input operation on the electronic dictionary device is
It will be easier.

[0003]

By the way, in the speech recognition processing, more information for speech recognition of more words is required in order to accurately recognize the inputted speech. However, there is a problem that the more the information for voice recognition to be referred to, the longer the time for voice recognition.

On the other hand, if the information for speech recognition to be referred to is limited in order to execute the speech recognition processing at a higher speed, the input will not be accurately recognized, and eventually the required dictionary will be required. There was a problem that text could not be obtained (search was not performed).

[0005] The present invention has been made in view of such circumstances, and it is an object of the present invention to obtain a required dictionary text quickly and accurately.

[0006]

According to a first aspect of the present invention, there is provided an information processing apparatus comprising: an input unit for inputting designation information by voice; a first storage unit for storing word data and information for voice recognition; A second storage unit for storing word data having a larger vocabulary than the word data stored in the first storage unit;
A third storage unit for storing information corresponding to the word data stored in the storage unit, and converting the designation information input by the input unit into information for voice recognition stored in the first storage unit. A first detecting means for performing voice recognition based on the first word data and detecting the N word data; and a predetermined word data selected from the N word data detected by the first detecting means. A second detecting means for detecting M word data from the second storage means based on the similarity criterion; and a word data selected from the M word data detected by the second detecting means. Information corresponding to the third
And N word data detected by the first detecting means, M word data detected by the second detecting means, or the third detecting means. Display control means for controlling display of information corresponding to the detected word data.

According to a third aspect of the present invention, in the information processing method, an input step of inputting designated information by voice, a first storage step of storing word data and information for voice recognition, and a first storage step are performed. A second storage step of storing word data having a greater vocabulary than the input word data, a third storage step of storing information corresponding to the word data stored in the second storage step, and an input step. A first detecting step of performing voice recognition on the specified information based on the voice recognition information stored in the first storing step, and detecting N word data; In correspondence with the predetermined word data selected from the selected N pieces of word data, based on a predetermined similarity criterion, M word units stored in the second storage step are used. A second detection step of detecting data; and information corresponding to the selected word data among the M word data detected in the second detection step, the information stored in the third storage step. From the third detection step, the N word data detected in the first detection step, and the M word data detected in the second detection step.
And a display control step of controlling display of information corresponding to the individual word data or the word data detected in the third detection step.

According to a fourth aspect of the present invention, a providing medium is stored in an input step of inputting designated information by voice, a first storage step of storing word data and information for voice recognition, and a first storage step. A second storage step of storing word data having a larger vocabulary than the word data obtained, a third storage step of storing information corresponding to the word data stored in the second storage step, and an input step. The specified information is subjected to voice recognition based on the voice recognition information stored in the first storage step, and a first detection step of detecting N word data; Corresponding to the predetermined word data selected from the N pieces of the word data, based on a predetermined similarity criterion, M word data from the word data stored in the second storage step. A second detecting step of detecting the data, of the M word data detected by the second detection step, the information corresponding to the word data selected, third
Detecting from the information stored in the storage step of the third
And the N detected in the first detection step
A display control step of controlling display of information corresponding to the word data, the M word data detected in the second detection step, or the word data detected in the third detection step. A computer-readable program to be executed by a processing device is provided.

[0009] The information processing apparatus according to claim 1, claim 3,
In the information processing method described in the above, and the providing medium described in the claim 4, the designation information is input by voice, the word data and the information for voice recognition are stored, Word data is stored, information corresponding to the stored word data is stored, and input designation information is subjected to voice recognition based on the stored voice recognition information, and N word data are detected. M word data is detected and detected from the stored word data based on a predetermined similarity criterion in accordance with predetermined word data selected from the detected N word data. Information corresponding to the selected word data among the M word data is detected, and information corresponding to the detected N word data, the detected M word data, or the detected word data is detected. Display of information is controlled.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below. In order to clarify the correspondence between each means of the invention described in the claims and the following embodiments, each means is described. When the features of the present invention are described by adding the corresponding embodiment (however, an example) in parentheses after the parentheses, the result is as follows. However, of course, this description does not mean that each means is limited to those described.

An information processing apparatus according to a first aspect of the present invention includes an input unit (for example, a microphone 5 in FIG. 2) for inputting designated information by voice, and a first storage unit for storing word data and information for voice recognition. (For example, the recognition target word storage unit 43 in FIG. 3) and a second storage unit (detection target word storage unit 45) that stores word data having a larger vocabulary than the word data stored in the first storage unit. And third storage means for storing information corresponding to the word data stored in the second storage means (for example, the dictionary body storage unit 47 in FIG. 3), and designation information input by the input means. A first detection unit (for example, the voice recognition unit 42 in FIG. 3) that performs voice recognition based on the voice recognition information stored in the first storage unit and detects N word data; N words detected by the detection means Corresponding to a predetermined word data selected from over data, based on a predetermined similarity criteria, second
The second detection means (for example, the headword detection unit 44 in FIG. 3) for detecting the M word data from the storage means, and the selected word data from the M word data detected by the second detection means A third detection unit (for example, a dictionary body detection unit 46) for detecting information corresponding to the extracted word data from the third storage unit, and N word data detected by the first detection unit. Display control means for controlling the display of information corresponding to the M word data detected by the detection means or the word data detected by the third detection means (for example, the input / output display control section 48 in FIG. 3) And characterized in that:

[0012] The information processing apparatus according to the second aspect of the present invention provides a changing means for changing the similarity criterion (for example, step S in FIG. 4).
6, S9, S12, S15), wherein the display control means is based on the word data detected by the first detection means and the similarity criterion changed by the change means by the second detection means. An information processing apparatus for displaying a combination of each of the detected word data and information corresponding to the word data detected by the third detection means (for example, FIGS. 17 and 18).

FIG. 1 shows an example of the external configuration of a portable electronic dictionary 10 to which the present invention is applied. Portable electronic dictionary 10
A display unit 1 configured by an LCD (Liquid Crystal Display) and displaying information to be presented to the user, such as a processing result corresponding to an input from the user, is provided at an upper front portion of the LCD.

A cursor button 2A operated to move the cursor displayed on the display unit 1 upward, a cursor button 2B operated to move the cursor downward, and a cursor below the display unit 1. A decision button 3 operated when selecting an item and a cancel button 4 operated when canceling a predetermined process are provided.

On the front right side of the portable electronic dictionary 10, a microphone 5 for a user to input voice and a reference change button 6 operated when changing a similarity reference (described later) are provided.

FIG. 2 shows an example of the internal configuration of the portable electronic dictionary 10. The interface 21 includes a display unit 1,
The cursor button 2, the enter button 3, the cancel button 4, the microphone 5, and the reference change button 6 are connected, and the interface 21 executes an input / output interface process.

The CPU 22 executes various processes according to a program stored in the ROM 23. The ROM 23 stores a program to be executed by the CPU 22 and also stores predetermined information necessary for executing the program. The RAM 24 appropriately stores data and programs necessary for the CPU 22 to execute various processes.

FIG. 3 shows an example of a functional configuration of an application program for executing the dictionary function with voice recognition of the portable electronic dictionary 10.

The speech recognition unit 42 is used for the recognition target word storage unit 4.
3, speech recognition is performed on the speech input from the microphone 5, and N speech recognition word candidates (described later) are determined. About 50 words are stored in the recognition target word storage unit 43.
Information of 00 words (the number of words smaller than the number of words collected in a general Japanese dictionary) and information for voice recognition corresponding to the information are registered.

The headword detection unit 44 refers to the detection target word storage unit 45, and searches for a candidate word for speech recognition word (described later).
Is detected. The detection target word storage unit 45 stores words of a larger vocabulary than the information of the words registered in the recognition target word storage unit 43, for example, tens of thousands of words (words that are included in a general Japanese language dictionary). Number) is registered.

The dictionary text detecting section 46 includes a dictionary text storage section 4.
7, the dictionary text (description of the headword) corresponding to the headword is detected. The dictionary text storage unit 47 stores the dictionary text of the words registered in the detection target word storage unit 45.

The input / output display control section 48 controls information to the display section 1. For example, the N voice recognition word candidates determined by the voice recognition unit 42, the headwords detected by the headword detection unit 44, and the dictionary text detected by the dictionary text detection unit 46 are input / output display control units 48. Is displayed on the display unit 1. The input / output display controller 48 also controls inputs from the cursor button 2 to the reference change button 6.

The control section 41 controls each section to execute each process described above.

The recognition target word storage unit 43, the detection target word storage unit 45, and the dictionary text storage unit 47 among the units that execute these functions are stored in the ROM 23, as shown in FIG. The remaining control unit 41, voice recognition unit 42,
The headword detection unit 44, the dictionary text detection unit 46, and the input / output display control unit 48 are held in the CPU 22.

Next, a processing procedure when the dictionary function with voice recognition is executed will be described with reference to the flowchart of FIG. In step S1, the control unit 41
Audio is input from the microphone 5 by the user, and an audio signal corresponding to the input audio is input / output display control unit 4.
Then, the process waits until the audio signal is supplied from the input / output 8 and when the audio signal is input from the input / output display control unit 48, the process proceeds to step S2. In this example, it is assumed that the user pronounces “Kamehameha” and inputs the sound from the microphone 5.

In step S2, the control section 41 transfers the voice signal supplied in step S1 to the voice recognition section 42, and executes voice recognition processing. Details of the speech recognition processing in step S2 are shown in the flowchart of FIG.

In step S31, the voice recognition unit 42
Performs an AD conversion process on the audio signal supplied from the input / output display control unit 48 to convert the audio signal into a digital audio signal.

In step S32, the voice recognition unit 42
Refers to the recognition target word storage unit 43, detects N words closer to the digital voice signal obtained in step S31 among words registered therein, and determines the words as voice recognition word candidates. That is, in the case of this example, N words that are acoustically closer to "Kamehameha" are set as speech recognition word candidates.

The speech recognition word candidates are determined in this way, and the recognition target word storage unit 43 referred to at this time is
Does not require a lot of time to determine speech recognition word candidates, since only a smaller number of words are registered than those in a general Japanese dictionary.

In step S33, the voice recognition unit 42
Reads the text data of each of the speech recognition word candidates determined in step S32 from the recognition target word storage unit 43, and supplies it to the input / output display control unit 48. As a result, the voice recognition processing is terminated, and step S3 in FIG.
Proceed to.

In step S3, the input / output display control section 48 outputs the text data supplied from the voice recognition section 42 to the display section 1 and causes the display section 1 to display the text data. Display unit 1 shown in FIG.
, Among the voice recognition word candidates corresponding to “Kamehameha” input by the user as voice, four words “from”, “Kame”, “Kame”, and “ha” are displayed. .

When there are four or more speech recognition word candidates, the user operates the cursor button 2 to move the cursor up and down and scroll the display on the display unit 1 to browse them. can do.

Next, in step S4, the control unit 41
Waits until the user operates the decision button 3 or the cancel button 4, and if the cancel button 4 is operated, the process returns to step S <b> 1, and the user again inputs voice from the microphone 5. Wait until it is done.

For example, if a word that was previously input by voice from the microphone 5, in this example, “Kamehameha” was not found in the voice recognition word candidates displayed on the display unit 1, the user , Cancel button 4
Can be input again by voice from the microphone 5.

Further, even if the user cannot find a word input by voice from the microphone 5 among the voice recognition word candidates displayed on the display unit 1, the user is acoustically close to the word. You can choose one. For example, in the case of this example, the user
Is recognized as a word close to "Kamehameha", and
By operating the cursor button 2 and moving the cursor to "Kame" as shown in FIG. 6 and operating the enter button 3, it can be selected. In such a case, in step S4, the control unit 41 determines that the determination button 3 has been operated, and proceeds to step S5, in which one of the speech recognition word candidates selected at this time, "Kame".
Is determined as a speech recognition word.

Next, in step S6, the control unit 41
Notifies the headword detection unit 44 of the speech recognition word determined in step S5, and causes the headword detection process to be executed based on the first similarity criterion. Here, among the words registered in the detection target word storage unit 45, the speech recognition words include
N words in the order of the Japanese syllabary are detected as headwords. Hereinafter, this processing will be described more specifically with reference to the flowchart of FIG.

In step S41, the headword detection unit 44 detects the order (number) of the words determined as the voice recognition words in step S5 in the order of the Japanese syllabary in the detection target word storage unit 45.

In step S42, the headword detecting unit 44 detects the 50th of the speech recognition words detected in step S41.
Each of the numerical values indicating the order of the syllabary order of the words registered in the detection target word storage unit 45 is subtracted from the numerical value indicating the order of the sound order, and the absolute value of the subtraction result (hereinafter, referred to as inter-word distance) ) Is calculated.

For example, if the order of the "Kame" in the Japanese syllabary is the 100th and the order of the "Kamo" after the "Kame" in the Japanese syllabary is the 103rd, "Kame" and " The distance between words of "Kamo" is 3 (= | 100-103 |)
It is said. Also, if the order of “Kamukoda” ahead of “Kameko” in the Japanese syllabary is the 99th order, the inter-word distance between “Kameko” and “Kamukoda” is 1 (= 100−99). )
It is said. In this way, the inter-word distance between the speech recognition word and each of the words registered in the detection target word storage unit 45 is calculated.

If the word recognized as the speech recognition word is not registered in the word-to-be-detected storage unit 45, in this example, if "Kame" is not registered in the word-to-be-detected storage unit 45, the process proceeds to step S41. , “Kame” are registered in the detection target word storage unit 45, and 50
The order of the sound order is determined. Then, the inter-word distance is calculated based on the determined order.

In step S43, the headword detection unit 44 determines the distance between words calculated in step S42.
From the smaller value, N inter-word distances are selected, the word when the distance is calculated is detected, and determined as a headword.

In this way, N words that are closer to the voice recognition word in the order of the Japanese syllabary are detected as headwords.

Next, in step S44, the headword detection unit 44 supplies the text data of the N headwords determined in step S43 to the input / output display control unit 48. Thus, the headword detection processing based on the first similarity criterion is terminated, and the process proceeds to step S7 in FIG.

In step S7, the input / output display control unit 48 outputs the text data supplied from the headword detection unit 44 to the display unit 1 and displays it. The display unit 1 shown in FIG. 8 displays a part of the headword detected based on the first similarity criterion of “Kame”, which is determined as the voice recognition word in step S5. For example, 50 from "Kame"
The first word in the phonetic order, "Kamukooda", is
"Kamehameha" is displayed above "Kame", and "Kamehameha" after "Kame" in the Japanese syllabary order is displayed below it.

It should be noted that "Kame", which is regarded as a speech recognition word,
If not registered in the detection target word storage unit 45, “Kame” itself is not detected as a headword (step S4).
3) It is not displayed on the display unit 1.

Next, in step S8, the control unit 41
Waits until the user operates the enter button 3 or the reference change button 6, and if the reference change button 6 is operated, the process proceeds to step S9 to control the headword detection unit 44. Then, the headword of the speech recognition word determined in step S5 is detected based on the second similarity criterion. Details of the headword detection process based on the second similarity criterion in step S9 are shown in the flowchart of FIG.

In step S51, the headword detection unit 44 replaces one character with another character, inserts one character, and performs processing for each word registered in the word storage unit 45. The process of deleting characters,
Alternatively, four processes of replacing a predetermined one character and a character adjacent thereto are performed, the word is converted into a word determined as a speech recognition word in step S5, and the number of times of the process performed at this time is counted. In the following, the counted number of processes is referred to as a MED (Minimun Edit Distance) value. When there are a plurality of combinations of the processing for converting to the speech recognition word (the processing with 4 described above), the smallest number of the processings is set as the MED value.

For example, "Ame" is the first character "
By performing the process of replacing “a” with “ka” once, it is converted into “kame”, that is, the MED value of “ame” for “kame” is set to “1”. , “Gamera” means that “ra” is deleted, and “
Is converted to “Kame” by executing the processing twice in total with the processing in which is replaced with “K”. That is, the MED value of “Game” for “Kame” is “2”.

Next, in step S52, the headword detecting unit 44 selects N MED values from the smaller one of the MED values counted in step S51, and extracts the words in which the MED values are counted. Detect and decide on a headword. Next, in step S53, the headword detection unit 44 supplies the text data of the N headwords determined in step S52 to the input / output display control unit 48.
Thus, the headword detection process based on the second similarity criterion is terminated, and the process proceeds to step S10 in FIG.

In step S10, the input / output display control unit 48 outputs the text data supplied from the headword detection unit 44 to the display unit 1 and displays it. The display unit 1 shown in FIG. 10 displays the word recognized as the voice recognition word in step S5.
A part of the headword of “Kame” detected based on the second similarity criterion is displayed.

Next, at step S11, the control unit 4
1 waits until the user operates the enter button 3 or the reference change button 6, and if the reference change button 6 is operated, the process proceeds to step S 12 and controls the headword detection unit 44. The headword corresponding to the speech recognition word determined in step S5 is detected based on the third similarity criterion. The details of the headword detection processing based on the third similarity criterion in step S12 are described in FIG.
1 is shown in the flowchart.

In step S61, the headword detection unit 44 selects a word including a speech recognition word from the words registered in the detection target word storage unit 45. For example, "nakameguro" includes "kame" in the character string, and in this case, is a headword corresponding to "kame".
Similarly, “Kame” and “Kamehameha” are also headwords because “Kame” is included in the character string.

In step S62, the headword detection unit 48 supplies the text data of the headword detected in step S61 to the input / output display control unit 48. Thereby, the headword detection processing based on the third similarity criterion is terminated, and the process proceeds to step S13 in FIG.

In step S13, the input / output display control unit 48 outputs the text data supplied from the headword detection unit 44 to the display unit 1 and displays it. The display unit 1 shown in FIG.
A part of the headword of "Kame" detected based on the third similarity criterion is displayed.

Next, in step S14, the control unit 4
1 waits until the user operates either the enter button 3 or the reference change button 6, and if the reference change button 6 is operated, the process proceeds to step S15, where the headword detection unit 44 is controlled. A headword corresponding to the speech recognition word determined in S5 is detected based on a fourth similarity criterion. Here, the thesaurus distance between the word recognized as the speech recognition word in step S5 and the word registered in the detection target word storage unit 45 is calculated. Hereinafter, this processing will be described with reference to the flowchart in FIG.

In step S71, the headword detection unit 44 forms a five-level thesaurus structure as shown in FIG. 14 using the words registered in the word storage unit 45 to be detected.

In step S72, the headword detection section 44 calculates a thesaurus distance between the word registered in the word storage section 45 to be detected and the speech recognition word. For example, “Kame” and “Suppin” in the sixth hierarchy belong to the category 1.5630-1 in the fifth hierarchy one level higher. In such a case, the thesaurus distance between “Kame” and “Suppin” is set to 1. Also, the sixth level “Kame” and “Snake”
Is 1.5630 in the fourth layer, which is two layers higher.
(Reptiles). In this case, "Kame"
And the thesaurus distance between “the snake” is 2. In this way, the thesaurus distance from the speech recognition word is calculated.

Next, in step S73, the headword detection unit 44 selects N thesaurus distances from the smaller one of the thesaurus distances calculated in step S72, and extracts the word at the time of calculation. Detect and decide on a headword.

Next, in step S74, the headword detection unit 44 supplies the text data of the N headwords determined in step S73 to the input / output display control unit 48. Thus, the headword detection processing based on the fourth similarity criterion is terminated, and the process proceeds to step S16 in FIG.

In step S16, the input / output display control section 48 outputs the text data supplied from the headword detection section 44 to the display section 1 for display. The display unit 1 shown in FIG. 15 displays a part of the headword detected based on the fourth similarity criterion of “Kame”, which is a speech recognition word.

Next, in step S17, the control unit 4
1 waits until either the enter button 3 or the reference change button 6 is operated, and when the reference change button 6 is operated, returns to step S6 and executes the subsequent processing.

In step S8, step S11, step S14, or step S17, the enter button 3
Is determined to have been operated, the control unit 41 proceeds to step S18, notifies the dictionary text detection unit 46 of the currently selected headword, and outputs the dictionary text (explanatory text) of the headword to the dictionary text. The data is read from the storage unit 47 and supplied to the input / output display control unit 48.
8 to output the text data supplied from the dictionary body storage unit 47 to the display unit 1 for display.

In the case of this example, the display unit 1 displays "Kamehameha" input in step S1 by the processing in step S7 and the processing in step S13, so that the user can proceed to step S8 or step S14.
By selecting “Kamehameha” and operating the enter button 3, the dictionary text of “Kamehameha” can be displayed on the display unit 1 as shown in FIG.

As described above, when the dictionary text is displayed on the display unit 1, the processing is terminated.

As described above, even if the word input from the microphone 5 is not displayed on the display unit 1 in step S5, the reference change button 6 is operated (toggle operation), and the headword detection processing is performed. Can be changed, the previously entered word can be displayed, and its dictionary text can be displayed.

In the above, the first to fourth similarity criteria
In the above description, the headwords detected based on the similarity criterion are separately displayed on the display unit 1. However, as shown in FIG. 17, they can be displayed on the display unit 1 at the same time. . In FIG. 17, headings detected based on a first similarity criterion, a second similarity criterion, a third similarity criterion, and a fourth similarity criterion from the left end to the right end of the display unit 1. The word is displayed.

Further, as shown in FIG. 18, the voice recognition word candidate (left side), the headword (center), and the dictionary sentence (right side) can be simultaneously displayed on the display unit 1.

[0068] Examples of a providing medium for providing a user with a computer program for performing the above-described processing include:
In addition to recording media such as magnetic disks, CD-ROMs, and solid-state memories, communication media such as networks and satellites can be used.

[0069]

According to the information processing apparatus according to the first aspect, the information processing method according to the third aspect, and the providing medium according to the fourth aspect, the word data corresponding to the information for voice recognition includes: Since the word data including the word information is separately stored, information corresponding to the specified information can be obtained accurately and at high speed.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration example of an appearance of a portable electronic dictionary 10 to which the present invention is applied.

FIG. 2 is a block diagram illustrating an example of an internal configuration of the portable electronic dictionary 10 of FIG.

FIG. 3 is a diagram illustrating a functional configuration example of the portable electronic dictionary 10 of FIG.

FIG. 4 is a flowchart illustrating a processing procedure when a dictionary function is executed.

FIG. 5 is a flowchart illustrating a speech recognition process.

FIG. 6 is a diagram illustrating a display example of the display unit 1.

FIG. 7 is a flowchart illustrating a headword detection process based on a first similarity criterion.

FIG. 8 is a diagram illustrating another display example of the display unit 1.

FIG. 9 is a flowchart illustrating a headword detection process based on a second similarity criterion.

FIG. 10 is a diagram illustrating another display example of the display unit 1.

FIG. 11 is a flowchart illustrating a headword detection process based on a third similarity criterion.

FIG. 12 is a diagram illustrating another display example of the display unit 1.

FIG. 13 is a flowchart illustrating a headword detection process based on a fourth similarity criterion.

FIG. 14 is a diagram illustrating a thesaurus structure.

FIG. 15 is a diagram illustrating another display example of the display unit 1.

FIG. 16 is a diagram illustrating another display example of the display unit 1.

FIG. 17 is a diagram illustrating another display example of the display unit 1.

FIG. 18 is a diagram illustrating another display example of the display unit 1.

[Explanation of symbols]

1 display unit, 2 cursor buttons, 3 enter button, 4 cancel button, 5 microphone,
6 Standard change button, 10 Portable electronic dictionary, 2
1 interface, 22 CPU, 23 ROM, 2
4 RAM, 41 control unit, 42 voice recognition unit, 43
Recognition word storage unit, 44 Headword detection unit, 4
5 Detection target word storage unit, 46 Dictionary text detection unit,
47 dictionary text storage unit, 48 input / output display control unit

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 15/40 310G 15/403 310Z 350C F-term (Reference) 5B009 KB00 VA00 5B075 ND03 PP07 PP22 PR06 QM05 UU01 5B091 AA03 BA02 CA22 CB12 CC02

Claims (4)

[Claims] An input unit for inputting designated information by voice; a first storage unit for storing word data and information for voice recognition; a first storage unit for storing word information and voice recognition information; A second storage unit for storing word data of a large vocabulary; a third storage unit for storing information corresponding to the word data stored in the second storage unit; First detection means for performing voice recognition on the designation information based on the voice recognition information stored in the first storage means and detecting N pieces of the word data; Corresponding to the predetermined word data selected from the N word data detected by the means,
A second detection unit that detects M word data from the second storage unit based on a predetermined similarity criterion; and, among the M word data detected by the second detection unit, Third detection means for detecting information corresponding to the selected word data from the third storage means, N word data detected by the first detection means, the second detection And a display control means for controlling display of the information corresponding to the M word data detected by the means or the word data detected by the third detection means. apparatus. 2. The image processing apparatus according to claim 1, further comprising a change unit configured to change the similarity criterion, wherein the display control unit sets the word data detected by the first detection unit, the word data detected by the second detection unit, The respective word data detected based on each of the changed similarity criteria;
The information processing apparatus according to claim 1, wherein the information corresponding to the word data detected by the detecting means is displayed in combination. 3. An input step of inputting designated information by voice; a first storage step of storing word data and information for voice recognition; and a larger than the word data stored in the first storage step. A second storage step of storing vocabulary word data; a third storage step of storing information corresponding to the word data stored in the second storage step; and the designation input in the input step A first detecting step of recognizing information based on the information for voice recognition stored in the first storing step and detecting N word data; and detecting in the first detecting step Corresponding to the predetermined word data selected from the N pieces of word data thus obtained,
A second detection step of detecting M word data from the word data stored in the second storage step based on a predetermined similarity criterion; and a M number of word data detected in the second detection step. A third detection step of detecting information corresponding to the selected word data from the word data selected from the information stored in the third storage step; and detecting the information in the first detection step. Controlling the display of the N pieces of word data, the M pieces of word data detected in the second detection step, or the information corresponding to the word data detected in the third detection step. An information processing method, comprising: a display control step. 4. An input step of inputting designated information by voice, a first storage step of storing word data and information for voice recognition, and a step of storing the word data and the word data stored in the first storage step. A second storage step of storing vocabulary word data; a third storage step of storing information corresponding to the word data stored in the second storage step; and the designation input in the input step A first detecting step of recognizing information based on the information for voice recognition stored in the first storing step and detecting N word data; and detecting in the first detecting step Corresponding to the predetermined word data selected from the N pieces of word data thus obtained,
A second detection step of detecting M word data from the word data stored in the second storage step based on a predetermined similarity criterion; and a M number of word data detected in the second detection step. A third detection step of detecting information corresponding to the selected word data from the word data selected from the information stored in the third storage step; and detecting the information in the first detection step. Controlling the display of the N pieces of word data, the M pieces of word data detected in the second detection step, or the information corresponding to the word data detected in the third detection step. A providing medium for providing a computer-readable program that causes an information processing apparatus to execute a process including a display control step.