JP3927800B2 - Voice recognition apparatus and method, program, and storage medium - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to the field of displaying speech recognition results using a graphic user interface (GUI).
[0002]
[Prior art]
Conventionally, a speech recognition technology for recognizing input speech by using a recognition dictionary in which a plurality of words is stored has been proposed. By using such speech recognition technology, an information processing apparatus or the like can be used. On the other hand, a technique for inputting a command desired by a user has also been proposed.
[0003]
In a speech recognition device or speech recognition system based on these technologies, in the case of a configuration in which other processing based on the recognition result is performed when the input speech recognition processing ends, the recognition processing If the result of misrecognition is included, the result of the other processing is also affected.
[0004]
For this reason, if it is assumed that commands that are important after the speech recognition processing or misrecognitions frequently occur, whether or not the recognition result is correct when the speech recognition processing ends. It is necessary to check whether or not
[0005]
As an example of such a recognition result confirmation process, for a certain recognition result, for example, a message prompting the user to confirm “Yes?” Is displayed on the display, and “Yes” is displayed. In addition, a method of prompting the user to operate a button by displaying a software button of “No” (hereinafter simply referred to as “button”) is generally used. Alternatively, a similar message is notified to the user with a synthesized sound, and recognition is performed using voice recognition of two types of words “Yes” and “No”, which are generally known to have relatively high recognition accuracy. There is a method of confirming the result by the user.
[0006]
According to the method for confirming the result of speech recognition in such a procedure, it is possible to prevent erroneous processing from being performed due to erroneous recognition. However, when erroneous recognition is repeated due to the influence of low recognition accuracy resulting from the use of only the recognition result of the first score in the speech recognition processing, the user-desired command in the processing performed after the speech recognition processing To get it done
(1) Notification to the user for confirmation of the recognition result from the system,
(2) “No” button selection operation that must be performed when the user is not satisfied with the recognition result, and re-input the same voice for re-recognition,
The user is forced to repeat the above procedure.
[0007]
Further, in the conventional speech recognition processing, even if the result of the first place score is a misrecognition for the user, the recognition result of the second place or lower may include a correct answer (a correct recognition result for the user). In this case, if a plurality of recognition results are simultaneously presented to the user as selection candidates, any candidate desired by the user can be selected, so that the troublesome procedure described above can be reduced.
[0008]
Furthermore, if these recognition result candidates are displayed as buttons and the corresponding command is executed when the button is pressed, a confirmation notification from the system or a response such as “Yes” or “No” from the user will be sent. The procedure can be omitted, and the number of user operations including voice input up to execution of a desired command can be reduced.
[0009]
As described above, a method for displaying a result of speech recognition as a graphic user interface (GUI) with a button on a display is proposed in, for example, Japanese Patent Laid-Open No. 10-21254. In the same publication, an information search device having a voice recognition function has been proposed, and according to the detailed description of the invention, when a keyword to be searched is input by voice and a search result by the recognized keyword is displayed, A search result for the first recognition result obtained as a result of the speech recognition is displayed, and a plurality of recognition result candidates of the first and lower are displayed so as to be selectable by a selection button in the order of candidates. When the first recognition candidate is incorrect, the user can select a correct speech recognition result as a keyword for information search by performing an operation of selecting a correct recognition result button using a mouse. it can.
[0010]
[Problems to be solved by the invention]
However, in the conventional technique in which a plurality of recognition result candidates described above are displayed, when the user operates buttons and switches in the graphic user interface, many buttons are arranged on the display screen, or the buttons themselves are small. In such a case, there is a high possibility that the operation will be mistaken, and care should be taken when operating a button for selecting a speech recognition result to perform particularly important processing.
[0011]
On the other hand, for buttons and switches attached to general physical devices (such as operation panels for home appliances and control panels for production facilities), for example, the buttons for performing important processes are enlarged. In addition, by arranging the size, shape, color, etc. according to the attributes of the buttons such as the frequency of use and functions, such as placing frequently used buttons at easy-to-press positions, it improves operability while preventing incorrect operations. I am trying. Therefore, even in the case of a configuration for displaying a voice recognition result using a graphic user interface, the same method is adopted, for example, with respect to a recognition result of a specific command, a display method is changed from another recognition result, It is thought that by changing the size of the GUI to be displayed according to the score of the recognition result, it is possible to prevent erroneous operations and improve user operability, but no such proposal has been made so far. .
[0012]
The present invention has been made in view of the above-described problems, and is a speech recognition that realizes good operability by improving a display mode when presenting a speech recognition result candidate to a user using a GUI. An object of the present invention is to provide an apparatus and method, a page description language display apparatus and control method thereof, and a computer program.
[0013]
[Means for Solving the Problems]
In order to achieve the above object, a speech recognition apparatus according to the present invention is characterized by the following configuration.
[0014]
For example, a speech recognition apparatus according to one aspect of the present invention is , Recognition result candidates by recognizing input speech And recognition score of each recognition result candidate For each acquisition result candidate acquired by the acquisition means and the acquisition means, Based on the recognition score, Display means Selectable When displaying Selectable area size Determining means, and determined by the determining means In size And display control means for controlling to display the recognition result candidates.
[0024]
DETAILED DESCRIPTION OF THE INVENTION
DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments of a speech recognition apparatus according to the present invention will be described in detail with reference to the drawings.
[0025]
[First Embodiment]
FIG. 1 is a block diagram illustrating a schematic configuration of the speech recognition apparatus according to the first embodiment.
[0026]
In the voice recognition apparatus shown in FIG. 1, reference numeral 100 denotes a voice input apparatus that inputs voice such as a microphone. Reference numeral 200 denotes a storage device such as a ROM, a RAM, and a hard disk that temporarily stores a program for operating the apparatus, data necessary for the operation of the apparatus, and data generated in the course of the operation.
[0027]
Reference numeral 300 denotes a display device such as a display used mainly for displaying recognition result candidates and the like. Reference numeral 400 denotes an operation input device such as a mouse or a keyboard used when a user inputs an operation.
[0028]
Reference numeral 501 denotes a voice recognition unit that recognizes input voice. A display control unit 502 controls the determination and display of the display mode (display format) of the recognition result. A recognition result selection unit 503 selects a recognition result in accordance with a user operation.
[0029]
Reference numeral 504 denotes a process control unit that executes a process corresponding to the recognition result based on the recognition result selected by the recognition result selection unit 503 or controls another program so as to execute the process. Reference numeral 505 denotes a display mode setting unit that sets a display mode.
[0030]
The storage device 200 describes an acoustic model 201 such as an HMM to be referred to when performing speech recognition, a recognition dictionary 202 describing pronunciation information of words to be recognized, and a method for determining a display mode by the display control unit 502. The display rule 203 and the processing rule 204 describing the processing method corresponding to the recognition result are stored.
[0031]
Here, an information processing apparatus such as a personal computer capable of inputting voice or a personal digital assistant (PDA) can be employed as the hardware of the voice recognition apparatus according to the present embodiment.
[0032]
Next, the operation of the speech recognition apparatus having the above-described configuration will be described with reference to FIG.
[0033]
FIG. 2 is a flowchart showing a control process of the speech recognition apparatus in the first embodiment, and a software program corresponding to each processing unit shown in FIG. 1 performed by a CPU (not shown) of the speech recognition apparatus. The processing procedure described in is shown below.
[0034]
In the figure, the voice input by the user using the voice input device 100 such as a microphone in step S101 is obtained by the function of the voice recognition unit 501 using the acoustic model 201 and the recognition dictionary 202 stored in the storage device 200. By being recognized, one or a plurality of recognition results are obtained (step S102).
[0035]
In step S103, the recognition result acquired in step S102 is determined as a recognition result selection candidate (recognition result candidate) according to the display format by determining the display mode on the display device 300 by the function of the display control unit 502. Display on the device 300. At this time, the method of determining the display mode may be described in a program in which the function of the display control unit 502 is described. For example, a display rule 203 as shown in FIG. You can leave it.
[0036]
FIG. 5 is a diagram for explaining an example of the display rule 203. In this rule, the type of GUI to be displayed and the display size are defined as rule 1 on the basis of the voice recognition score, and the arrangement at the time of display is a rule. 2 is specified. Such setting of the display rule 203 can be set by the user by the function of the display setting unit 505.
[0037]
For example, in step S102, the speech recognition process is performed using the recognition dictionary 202 in which the recognition vocabulary and pronunciation as illustrated in FIG. 3 are described, and the result of obtaining the top four candidates from the higher score is shown in FIG. It is assumed that
[0038]
In the above case, in step S103, the display rule 203 illustrated in FIG. 5 is referred to, and a diagram illustrating a plurality of recognition result selection candidates such as the GUI illustrated in FIG. 6 is displayed. That is, in the example of FIG. 6, according to the rules 1 and 2, “print” having the largest score is displayed as a button having a large size, and the following three candidates (recognition result candidates) are sequentially displayed according to the score values. It is displayed as a button.
[0039]
In step S104, a desired button is selected by the user from the plurality of buttons displayed as illustrated in FIG. 6 using the operation input device 400 such as a mouse, and the vocabulary corresponding to the selected button. (Recognition result candidate) is set as a correct recognition result.
[0040]
In step S105, the corresponding process is executed by referring to the processing rule 204 stored in the storage device 200 according to the set vocabulary as a correct recognition result according to the selection operation. For example, as illustrated in FIG. 7, the processing rule 204 is defined for each word described in the recognition dictionary 202, for example, printing processing is performed if the set recognition result is “print”.
[0041]
According to the present embodiment as described above, since the display mode when the result of speech recognition is presented to the user using the GUI is improved, good operability can be realized.
[0042]
In the present embodiment described above, the recognition result score of each vocabulary is used as the recognition result attribute shown in FIG. 4. However, the present invention is not limited to this, but the importance of each word and the type of part of speech. , Adopts various attributes such as whether it is a command for instructing a predetermined process, whether it is a keyword for information retrieval, frequency used in input voice, language type such as English / Japanese, etc. It is only necessary that at least one of these attributes is commonly used for each recognition result (the same applies to the following embodiments).
[0043]
[Second Embodiment]
Next, a second embodiment based on the speech recognition apparatus according to the first embodiment described above will be described. In the following description, the description similar to that of the first embodiment will be omitted, and the description will focus on the characteristic part of the present embodiment.
[0044]
In the first embodiment, the case where the display mode is mainly changed according to the recognition score has been described. However, in this embodiment, for example, as shown in FIG. 8, place names such as “Tokyo” and “Osaka” have a small size. Words that are displayed as text using normal fonts and that correspond to commands for selecting processing such as "Finish" and "Print" are displayed as buttons, and the font and button size at the time of display are The display mode is controlled using a display rule in which a display mode is set for each word, such as appropriately set according to the degree of importance.
[0045]
Also according to this embodiment, since the display mode when presenting the speech recognition result candidates to the user using the GUI is improved, good operability can be realized.
[0046]
[Third Embodiment]
Next, a third embodiment based on the speech recognition apparatus according to the first embodiment described above will be described. In the following description, the description similar to that of the first embodiment will be omitted, and the description will focus on the characteristic part of the present embodiment.
[0047]
In general, for example, in an application that edits text, voice recognition may be used for different purposes, such as inputting text or inputting a command for operating the application. In such a case, as a method for displaying recognition result candidates, it is easier for the user to display commands as buttons and text as text.
[0048]
Also, there may be no problem in operation even if a certain command selected by the user is mistakenly executed, such as displaying a menu. On the other hand, what is the user's original hope, such as closing an application? Some commands are difficult to return when different commands are executed.
[0049]
Therefore, in this embodiment, operability is improved by changing the display mode according to the executed command. For example, a command for ending an application can be prevented from being erroneously operated by improving the visibility by, for example, making the size larger than other recognition result candidates when displayed.
[0050]
In order to realize such a display mode, in the present embodiment, control control based on a display rule as illustrated in FIG. 10 is performed.
[0051]
In the display rule illustrated in FIG. 10, rule 1 describes that when there is a process corresponding to a recognition result candidate, it is displayed with a button, and in other cases, it is displayed as text. In rule 2, the size and font for display are defined using the recognition result candidate attribute of importance. And in rule 3, arrangement | positioning at the time of displaying a some recognition result candidate is prescribed | regulated.
[0052]
Here, the importance may be given in advance for each word as shown in FIG. 11 and may be included in the display rule, or the importance is described in the recognition dictionary itself and stored in the storage device 200 in advance. You can keep it.
[0053]
That is, in the example of FIG. 11, importance is set to a word corresponding to a command to avoid an erroneous operation such as “end”, “print”, and “delete”, and the importance is high in the display rule illustrated in FIG. 10. The word is described to be large in size and thickened to improve visibility.
[0054]
12 is a diagram illustrating a recognition result selection GUI displayed when a recognition result candidate based on the score shown in FIG. 9 is obtained based on the display rule shown in FIG. 10 in the third embodiment. A button is displayed for recognition result candidates with corresponding processing, other recognition result candidates are displayed as text, and the display size of words with high importance such as “End” and “Delete” is large. Has been.
[0055]
Also according to this embodiment, since the display mode when presenting the speech recognition result candidates to the user using the GUI is improved, good operability can be realized.
[0056]
[Fourth Embodiment]
Next, a fourth embodiment based on the speech recognition apparatus according to the first to third embodiments described above will be described. In the following description, the description similar to that of each of the above-described embodiments will be omitted, and description will be made focusing on characteristic portions in the present embodiment.
[0057]
In this embodiment, an example in which the speech recognition apparatus described in the first to third embodiments described above is realized in a page description language display apparatus that displays a page description language such as a WWW (World Wide Web) browser will be described. This will be described with reference to FIG. 13 and FIG.
[0058]
FIG. 13 is a block diagram showing a schematic configuration of a page description language display device according to the fourth embodiment, and the basic device configuration is the same as that of the speech recognition apparatus shown in FIG. Also in the present embodiment, an information processing apparatus such as a personal computer and a personal digital assistant (PDA) capable of inputting voice can be employed as hardware.
[0059]
That is, in FIG. 13, reference numeral 100 denotes a voice input device for inputting voice such as a microphone. Reference numeral 200 denotes a storage device such as a ROM, a RAM, and a hard disk that temporarily stores a program for operating the apparatus, data necessary for the operation of the apparatus, and data generated in the course of the operation.
[0060]
Reference numeral 300 denotes a display device such as a display mainly used for displaying recognition result candidates and the like. Reference numeral 400 denotes an operation input device such as a mouse or a keyboard used when a user inputs an operation.
[0061]
A page description control unit 507 controls data input in a predetermined page description language format, analyzes the input page description language, and displays a page on the display device 300 based on the analysis result. Reference numeral 501 denotes a voice recognition unit that recognizes input voice. Reference numeral 506 denotes a page description data generation unit that generates page description language data for displaying a voice recognition result by the voice recognition unit 501.
[0062]
Reference numeral 503 denotes a recognition result selection unit that selects any desired one of the recognition result candidates displayed on the display device 300 in accordance with a user operation. Based on the recognition result candidate selected by the recognition result selection unit 503, a processing control unit 504 executes a process corresponding to the recognition result candidate or controls another program so as to execute the process.
[0063]
The storage device 200 includes an acoustic model 201 such as an HMM to be referred to when performing speech recognition, a recognition dictionary 202 describing pronunciation information of words to be recognized, Page description data generation unit 506 Determine the display mode with for A display rule 203 describing the method and a processing rule 204 describing the processing method corresponding to the recognition result are stored. Here, various data including the acoustic models 201 to 204 stored in the storage device 200 are transmitted from a storage device such as a WWW server, which has a different configuration from the present device, via a communication network such as the Internet. It is good also as a structure which can be read.
[0064]
Next, the operation of the page description language display device having the above-described configuration will be described with reference to FIG. 14 using the same display rules, processing rules, and recognition results as those in the first embodiment.
[0065]
FIG. 14 is a flowchart showing a control process of the page description language display device in the fourth embodiment, and corresponds to each processing unit shown in FIG. 13 performed by a CPU (not shown) of the page description language display device. The processing procedure described in the software program is shown below.
[0066]
In the figure, in step S201, page description language data stored on the Internet server or the storage device 200 in the apparatus is input to the apparatus. The input page description language data is analyzed by the function of the page description language control unit 506 (step S202). In step S203, a page corresponding to the description content of the page description language data is based on the analysis result. Is displayed on the display device 300.
[0067]
Next, in the process of analyzing the page description language in step S202, it is determined whether a tag for performing speech recognition is described in the data of the page description language (step S204). Ends the process, and if it is described, in step S205, the voice input by the user using the voice input device 100 such as a microphone is accepted.
[0068]
In the present embodiment, tags other than the tag for performing speech recognition included in the page description language data have the same functions as those of a general page description language display device such as a WWW browser.
[0069]
In step S206, the voice input in step S205 is recognized by the function of the voice recognition unit 501 using the acoustic model 201 and the recognition dictionary 202 stored in the storage device 200, so that one or a plurality of voices are recognized. A recognition result candidate is obtained.
[0070]
In step S207, the function of the page description data generation unit 506 determines the display mode on the display device 300 as in the first embodiment, based on the recognition result candidate (FIG. 4) acquired by the voice recognition unit 501. In addition, page description language data corresponding to the contents of the determined display mode is further generated. Data generated by the function of the page description data generation unit 506 is set in the page description control unit 507.
[0071]
In step S208, the page description language data generated in step S207 is analyzed by the function of the page description control unit 507. In step S209, a page is displayed on the display device 300 based on the analysis result. . At this time, the method for determining the display mode of the page to be displayed on the display device 300 may be described in the program as in the first embodiment, or the display rule 203 as shown in FIG. 5 is stored. You may memorize | store in the apparatus 200 and refer to the display rule. The display rule 203 may be set by the user using the function of the display setting unit 505.
[0072]
FIG. 15 is a diagram illustrating a tag for executing speech recognition in the fourth embodiment.
[0073]
In the figure, the part shown in italics is an example of a speech recognition tag according to the present embodiment, and “<SpeechRcog .....>” is a description for executing input by speech recognition. In the above, “<SpeechRcog .....>” is interpreted as “recognize speech and display the recognized result”.
[0074]
In the page description language display device according to the present embodiment, the recognition dictionary 201 and the acoustic model 202 used for speech recognition can be specified by descriptions “grammar” and “acousticmodel”. Furthermore, it is assumed that the display rule 203 and the processing rule 204 described in the first embodiment can be specified by descriptions “resulttemplate” and “actiontable”.
[0075]
That is, in the example shown in FIG. 15, the function of the voice recognition unit 501 uses the recognition dictionary “command.gra” and the acoustic model “phone.mdl” according to the tag “<SpeechRcog .....>”. In addition to performing recognition, the function of the page description data generation unit 506 generates page description language data for displaying recognition result candidates with reference to the display rule “type1.dat” and the processing rule “command.tbl”. It is shown to do.
[0076]
FIG. 16 is a diagram illustrating page description language data generated by the function of the page description data generation unit 506 based on the display rule 203 shown in FIG. 5 in the fourth embodiment.
[0077]
In the figure, the italicized part is an extended part of the specification of a general page description language. In the page description language display device according to this embodiment, the description “input type = mybotton” and “size” Depending on the description, it is possible to specify the display of the button and the size of the button at the time of display. In addition, the range surrounded by “<p>... </ P>” is interpreted as being displayed as a single line button. In this embodiment, when the recognition result candidate is displayed on the display device 300 according to FIG. 16 by interpreting the extended specification, the display example shown in FIG. 6 is displayed as in the case of the first embodiment. Is done.
[0078]
Here, the description returns to the flowchart of FIG. In step S210, for the recognition result candidate displayed on the display device 300, the user selects a desired recognition result candidate using the operation input device 400 such as a mouse by the function of the recognition result selection unit 503.
[0079]
In step S211, processing corresponding to the selected recognition result candidate is executed by the function of the processing control unit 504. Here, the correspondence relationship between the recognition result candidate and the process may be described in the program, or the processing rule 204 in which the correspondence relationship illustrated in FIG. 7 is described is stored in the storage device 200 in advance. It may be referred to when the process is executed by the process control unit 504.
[0080]
In the example of the page description language data shown in FIG. 16, when the user presses the button displayed on the display device 300, the environment variable “com” specified by “name” is specified by “value”. The entered value (corresponding to search, print setting, and editing in the figure) is substituted, and execution of the program “ExecuteCommand” is started. The description “ExecuteCommand” corresponds to the processing control unit 504, and the “ExecuteCommand” and the program extract the value assigned to “com” from the environment variable specified by “name” and correspond to the extracted value. The process to be executed is specified by referring to the process rule 204 and executed.
[0081]
Also according to this embodiment, since the display mode when presenting the speech recognition result candidates to the user using the GUI is improved, good operability can be realized.
[0082]
As described above, according to each of the above-described embodiments, when one or a plurality of speech recognition result candidates are displayed on the display device 300 using a GUI including software buttons and text, individual speech recognition is performed. Since the format to be displayed for each recognition result candidate is determined based on attributes such as the recognition score and importance of the result candidate, the user interface using voice recognition and the operability of command input can be improved.
[0083]
In each of the above-described embodiments, the GUI illustrated in FIGS. 6 and 12 selects a user's desired one from a plurality of candidates displayed on the GUI using software buttons having different sizes and fonts. However, the present invention is not limited to this device configuration. For example, it may be configured to be selectable by a so-called radio button or check box provided in or near the display area of each selection candidate having a different size or font. good.
[0084]
Further, in each of the above-described embodiments, the GUI illustrated in FIG. 6 and FIG. 12 includes software buttons having different sizes and fonts when selecting a user-desired one from a plurality of candidates displayed on the GUI. The operability for the user has been improved by using the configuration to be selectable, but is not limited to this device configuration. For example, the operation can be performed by appropriately changing the display color of the button, the shape of the displayed button, Can be improved.
[0085]
[Other Embodiments]
An object of the present invention is to use a storage medium (or recording medium) that records a program code of software that realizes the functions of the above-described embodiments as a personal computer or a portable information terminal that operates as the above-described voice recognition device. This is also achieved by supplying the information processing apparatus and reading and executing the program code stored in the storage medium by the computer (or CPU or MPU) of the system or apparatus. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and is acquired as a computer program product via the storage medium storing the program code and the telecommunication line. The program code thus configured constitutes the present invention.
[0086]
Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an operating system (OS) or the like running on the computer based on the instruction of the program code. A case where part or all of the actual processing is performed and the functions of the above-described embodiments are realized by the processing is also included.
[0087]
Further, after the program code read from the storage medium is written in a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, the function is determined based on the instruction of the program code. This includes a case where the CPU or the like provided in the expansion card or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.
[0088]
【The invention's effect】
According to the present invention described above, a speech recognition apparatus and method for realizing good operability by improving a display mode when a speech recognition result candidate is presented to a user using a GUI, and a page description language display Provision of an apparatus, a control method thereof, and a computer program is realized.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a speech recognition apparatus according to a first embodiment.
FIG. 2 is a flowchart showing a control process of the speech recognition apparatus in the first embodiment.
FIG. 3 is a diagram illustrating a configuration example of a recognition dictionary 202 in the first embodiment.
FIG. 4 is a diagram illustrating an example of a recognition result candidate based on a score according to the first embodiment.
FIG. 5 is a diagram for explaining an example of a display rule 203 in the first embodiment.
6 is a diagram exemplifying a display mode of a recognition result selection GUI displayed on the display device 300 in the first embodiment. FIG.
7 is a diagram illustrating a configuration example of a processing rule 204 in the first embodiment. FIG.
FIG. 8 is a diagram for explaining an example of a display rule 203 in the second embodiment.
FIG. 9 is a diagram illustrating an example of a recognition result candidate based on a score according to the second embodiment.
FIG. 10 is a diagram for explaining an example of a display rule 203 in the third embodiment.
FIG. 11 is a diagram illustrating an example of setting importance for each word in the third embodiment.
FIG. 12 is a diagram illustrating a display mode of a recognition result selection GUI displayed on the display device in the third embodiment.
FIG. 13 is a block diagram showing a schematic configuration of a page description language display device according to a fourth embodiment.
FIG. 14 is a flowchart illustrating a control process of the page description language display device according to the fourth embodiment.
FIG. 15 is a diagram illustrating a tag for executing speech recognition in the fourth embodiment.
FIG. 16 is a diagram illustrating page description language data generated by the function of the page description data generation unit 506 based on the display rule 203 shown in FIG. 5 in the fourth embodiment.

Claims (12)

An acquisition means for acquiring a recognition result candidate and a recognition score of each recognition result candidate by recognizing the input voice;
For each recognition result candidate acquired by the acquisition unit, a determination unit that determines the size of a selectable region when displaying selectably on a display unit based on the recognition score;
And a display control unit configured to control the display of the recognition result candidate with a size determined by the determination unit. An acquisition means for acquiring a recognition result candidate by recognizing the input voice;
For each recognition result candidate acquired by the acquisition means, a determination means for determining the size of a selectable area when displaying selectably on the display means based on the importance of the recognition result candidate;
And a display control unit configured to control the display of the recognition result candidate with a size determined by the determination unit. The apparatus further comprises control means for controlling the process corresponding to the recognition result candidate to be executed in the own apparatus or an external apparatus according to the recognition result candidate selected from the recognition result candidates displayed by the display control means. The speech recognition apparatus according to claim 1 or 2 . In accordance with a recognition result candidate selected from the recognition result candidates displayed by the display control means, the apparatus further comprises control means for controlling the processing corresponding to the recognition result candidate to be executed in the own device or an external device,
The speech recognition apparatus according to claim 2 , wherein the importance is set to be large for a recognition result candidate corresponding to a process for which an erroneous operation is desired to be avoided. A pointing device, speech recognition apparatus according to any one of claims 1 to 4, further comprising a selection means for selecting the display means on the displayed recognition result candidates. An acquisition step of acquiring a recognition result candidate and a recognition score of each recognition result candidate by recognizing the input voice;
For each recognition result candidate acquired in the acquisition step, a determination step for determining the size of a selectable region when displaying selectably on a display unit based on the recognition score;
And a display control step of controlling to display the recognition result candidates with the size determined in the determination step. An acquisition step of acquiring recognition result candidates by recognizing input speech;
For each recognition result candidate acquired in the acquisition step, a determination step for determining the size of a selectable region when displaying selectably on a display unit based on the importance of the recognition result candidate;
And a display control step of controlling to display the recognition result candidates with the size determined in the determination step. According to the recognition result candidate selected from the recognition result candidates displayed by the display control step, the method further comprises a control step of controlling the processing corresponding to the recognition result candidate to be executed in the own device or an external device. The speech recognition method according to claim 6 or 7 . In accordance with a recognition result candidate selected from the recognition result candidates displayed by the display control step, further comprising a control step of controlling the processing corresponding to the recognition result candidate to be executed in the own device or an external device,
The speech recognition method according to claim 7 , wherein the importance is set to be large for a recognition result candidate corresponding to a process for which an erroneous operation is desired to be avoided. The speech recognition method according to claim 6 , further comprising a selection step of selecting a recognition result candidate displayed on the display unit by a pointing device.   A program for causing a computer to execute the speech recognition method according to any one of claims 6 to 10. A computer-readable storage medium storing the program according to claim 11 .

Images