JPH05341951A - Voice input operation unit - Google Patents

Abstract

PURPOSE:To quickly and smoothly perform recognition confirming processing and erroneous recognition processing, etc., accompanied by a voice input operation. CONSTITUTION:When a voice recognition button on the display screen of a display part 4 is selected with the left button of a device (mouse) 5, the shape of a cursor in the position is changed by a control part 3 and a user is urged to vocalize. In this case, when the user input one of selectable operation buttons in voice, the voice is recognized at a voice recognition part 2 and the result is delivered to the control part 3. The control part 3 urges the recognition processing by the user by moving the cursor to the operation button corresponding to the first option of the recognized result and changing the shape of the cursor. When the user presses the left, center or right button of the device (mouse) 5 at this time, the control part 3 selects the operation button at the position of the cursor, moves the cursor to the operation button corresponding to the next option of the recognized result or moves the cursor to a voice recognition button.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice input operation device suitable for performing various controls by using position indicating means such as a mouse and voice input means in combination.

[0002]

2. Description of the Related Art Speech recognition technology is an excellent man-machine
It plays an important role in realizing the interface. In recent years, with the high performance and miniaturization of personal computers, these devices are often operated by voice input.

How to perform recognition confirmation and erroneous recognition processing when operating such a device by voice input,
The problem of how to encourage vocalization is one of the most important practical issues. In the conventional voice input operation device, the above-described recognition confirmation, erroneous recognition processing, and notification prompting utterance have been realized by the following methods.

First, in the conventional voice input operation device,
For example, when the user wants to select one of a plurality of selectors (selectable display cells such as operation buttons displayed for command selection) displayed on the display unit, the user selects the selector name ( Input the display cell name, command name, etc. by voice.

The voice input operation device recognizes the voice uttered by the user and displays the recognition result as a character string in a predetermined area of the display screen. At the same time, the voice input operation device displays a confirmation guide as to whether or not the recognition result is correct.

[0006] The user determines whether or not the input voice is correctly recognized from the display character string indicating the recognition result, and inputs the determination result with an input device such as a keyboard, thereby inputting to the voice input operation device. Notice.

[0007] The voice input operation device, when notified of correct recognition (recognition confirmation) by the user's keyboard operation, performs processing corresponding to the recognized selector,
When the erroneous recognition is notified, the next candidate (of recognition) is displayed as a recognition result in a character string, and the confirmation guidance is displayed again.

Further, when the recognition result by the voice input operation device is erroneous recognition and it is desired to correct the recognition result,
The user performs a two-step process of correcting the recognition result by moving to the erroneous recognition process by operating the keyboard in the confirmation process. On the other hand, the notification prompting the user to speak is realized by a method of displaying "Please speak" or the like after entering the speech reception process by operating the keyboard.

In the recognition confirmation and the erroneous recognition processing in the conventional voice input operation device as described above, the recognition confirmation cannot be intuitively and easily performed, and the correction in the erroneous recognition processing cannot be performed quickly. It had a major drawback.

[0010]

As described above, in the conventional voice input operation device, recognition confirmation cannot be performed intuitively and easily, and when the recognition result is erroneous recognition,
In the confirmation process, the keyboard is operated to move to the misrecognition process,
There is a problem that a two-step process of correcting the recognition result is necessary. In addition, the notification that prompts the user to speak requires an input operation such as a keyboard.

That is, in the conventional voice input operation device,
There is a large operability drawback that quick recognition confirmation and erroneous recognition processing cannot be performed, and troublesome preparation for starting vocalization is required, and there is a problem that a quick and smooth operation by voice input is hindered.

The present invention has been made in consideration of such circumstances, and its purpose is to display a selector (operation button) on a command selection screen, a building, a place, etc. on a map screen. When a display cell that is not the target of selection on the screen is specified by voice, a position indication display that can be moved by a position indication means such as a mouse is displayed on the display cell on the display screen that corresponds to the voice recognition result. An object of the present invention is to provide a voice input operation device that allows quick recognition and confirmation by moving a cursor.

Another object of the present invention is to provide a voice input operation device in which an operation for erroneous recognition processing of the above voice recognition result can be performed quickly and smoothly by using a position indicating means such as a mouse. ..

Still another object of the present invention is to provide a voice input operation device capable of giving a notification prompting the user to utter a voice input, in particular, a notification prompting a re-voice due to misrecognition without burdening the user. To do.

[0015]

A voice input operation device according to the present invention is a selector (operation button) on a command selection screen.
In addition, when selecting a display cell such as a building or place on the map screen that is to be selected and specified on the display screen based on the recognition result of voice input, move it by operating the position indicating means such as a mouse. The controllable position indication display cursor is moved to the position of the corresponding display cell on the display screen, and the recognition result for the input voice is indicated by the position indication display cursor.

Further, according to the present invention, after the recognition result for the input voice is pointed by the position pointing display cursor, the input from the position pointing means is accepted, and the "recognition" or "erroneous recognition process" of the recognition result is performed according to the input information. It is also characterized in that it is configured.

Furthermore, the present invention provides "erroneous recognition processing" for the recognition result.
In accordance with the operation contents of the position indicating means, the position indicating display cursor is moved to the display cell corresponding to the next candidate of the recognition result, and the voice input start can be specified by the position indicating operation from the position indicating means. It is also characterized in that the display cell selection designation according to the movement of the position indication display cursor or the position indication input from the position indication means is accepted.

Further, according to the present invention, when the position indicating display cursor is moved to the corresponding display cell according to the recognition result,
It is also characterized in that display cells corresponding to the top N (N is an integer equal to or greater than 1) of the subsequent candidates of the recognition result are displayed in a display mode specific to the candidate rank.

[0019]

In the above configuration, when the specific display cell displayed on the display screen is positioned by operating the position pointing means and the specific display cell is selected, the color or shape of the position pointing display cursor is changed. It changes and prompts the user to speak.

Here, when the user utters an attribute name unique to the target display cell in order to select any one from the display cell group on the display screen, the generated content is recognized, and the generated content is recognized. The position indicating display cursor is moved to the display cell corresponding to the first candidate of the recognition result. The user can intuitively judge the recognition result for the voice uttered by the user from (the display cell of) the position pointed by the position pointing cursor.

If the display cell at the position pointed by the position indicating cursor is the intended one, that is, if the recognition result is correct, the user performs a specific operation of the position indicating means to select the display cell ( If the position indicating means is an example of a mouse, a specific button of the mouse is operated. As a result, while the user operates the position indicating means to move the position indicating cursor to the target display cell, the display is performed as if the specific operation of the position indicating means for selecting the display cell is performed. The cell is selected.

If the display cell at the position pointed by the position pointing cursor is not the intended one, that is, if the recognition result is incorrect, the user specifies the position pointing means for "erroneous recognition processing". Do the operation. If the first specific operation is performed here, the position indicating display cursor is moved to the display cell corresponding to the next candidate of the recognition result, and “confirmation” or “erroneous recognition processing” for the next candidate becomes possible. .. In addition, when the second specific operation is performed, the position indicating display cursor is moved to the specific display cell, and it becomes possible to give a notification prompting a re-voice. Further, when the third specific operation is performed, the display cell selection designation mode according to the position indication input from the position indication means is entered, and the position indication means can be operated to directly specify the position of the target display cell. It will be possible.

Further, in the above configuration, the position indicating display cursor is moved to the corresponding display cell in accordance with the recognition result for the input voice, and at the same time, the upper N candidates of the second and subsequent candidates of the recognition result (N is 1 Display cells corresponding to the above integers) are displayed in a display form specific to the candidate rank.
Therefore, the user can confirm the content of the recognition result candidate on the screen, and thus can smoothly determine whether to re-voice or select the next candidate.

As described above, according to the above configuration, the user only operates the position indicating means with respect to the recognition result of the voice input to confirm the recognition result and erroneously recognize the recognition result from the notification prompting the utterance of the voice input. The processing up to the processing can be performed, and a quick and smooth voice input operation can be performed.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment to which the present invention is applied will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a voice input operation device according to an embodiment of the present invention. In the figure, 1 is a microphone for inputting voice, and 2 is a voice recognition unit for recognizing voice input by the microphone 1. Reference numeral 3 denotes a control unit that receives the recognition result of the voice recognition unit 2 and the operation content of a pointing device 5 described later, and controls the entire apparatus including display control.

A display unit 4 is controlled by the control unit 3,
Reference numeral 5 denotes a pointing device capable of pointing an arbitrary position on the display screen of the display unit 4 by a user's operation and moving the position pointing display cursor (pointing cursor) to the designated position. This pointing device 5 is used for moving the pointing cursor,
It is possible to input selection information. A typical pointing device 5 is a mouse, a touch panel, a pen input device, or the like, but here, for simplicity, a three-button mouse will be assumed and described.

FIG. 2 shows an example of the screen configuration of the display screen on the display unit 4 of FIG. The display screen shown in FIG. 2 is composed of a work screen 21 in which sentences and figures can be input, and a command selection screen 22 in which commands can be selected. On this command selection screen 22, a plurality of command selectors are displayed as operation buttons 23. Command selection screen 22
In a predetermined area of, for example, an operation button having a lip shape pattern used to notify the start of voice recognition (hereinafter,
A voice recognition button) 24 is displayed. By selecting the voice recognition button 24 with the pointing device (mouse) 5, the user is prompted to speak by changing the shape of the pointing cursor (mouse cursor).

When the user wants to select a desired operation button 23 on the command selection screen 22 shown in FIG. 2 by voice input, first, the user selects the voice recognition button 24 with the pointing device (mouse) 5. After that, an attribute name (here, command name) unique to the attribute (here, command) of the desired operation button 23 is uttered toward the microphone 1.

The attribute name uttered by the user, that is, the command name is input by the microphone 1, and the voice recognition unit 2
Sent to. The voice recognition unit 2 performs acoustic analysis, feature extraction on the input voice (command name) from the microphone 1,
Well-known voice recognition processing including processing such as similarity calculation is performed, and the recognition result (first to nth candidates) is passed to the control unit 3.

Upon receiving the voice recognition result for the input voice (command name) from the voice recognition unit 2, the control unit 3 receives a mouse cursor (pointing button) on the display screen of the display unit 4.
The cursor) is moved so as to overlap the operation button 23 of the command corresponding to the recognition result (first candidate).

The user determines from the display position of the mouse cursor, that is, the position of the operation button 23 pointed to by the mouse cursor, whether or not the recognition result is correct. Then, if the recognition result is correct, the user inputs confirmation by using, for example, the left button of the pointing device (mouse) 5 at that position. Then, the operation button 23 pointed by the mouse cursor is selected, and the operation corresponding to the selected button 23 is executed by the control unit 3.

On the other hand, in the case of erroneous recognition, that is, the mouse is pressed on the operation button 23 that does not correspond to the command name uttered by the user
When the cursor moves, the user performs erroneous recognition processing by any one of (1) selection of next candidate, (2) re-voiced voice, and (3) direct input.

First, in order to perform the erroneous recognition process by selecting the next candidate in (1) above, the user presses, for example, the middle button of the pointing device (mouse) 5. Then, under the control of the control unit 3, the mouse cursor moves to the operation button 23 corresponding to the next candidate of the recognition result. In this state, confirmation processing or erroneous recognition processing again can be performed.

When the user wants to perform the erroneous recognition process due to the re-voiced voice of (2), the user presses the right button of the pointing device (mouse) 5, for example. Then, under the control of the control unit 3, the mouse cursor moves onto the voice recognition button 24 on the display screen shown in FIG. 2 and the shape thereof changes to prompt the user to speak.

Next, when it is desired to perform the erroneous recognition processing by the direct input of the above (3), the user directly moves the pointing device (mouse) 5 to move the mouse cursor to the target operation button 23. Is moved, and the left button of the pointing device (mouse) 5 is pressed at that position. Then, the operation button 23 pointed by the mouse cursor is selected, and the operation corresponding to the selected button 23 is executed by the control unit 3.

Next, a series of processes relating to the above-mentioned voice input operation will be described more specifically with reference to the flow chart shown in FIG. 3 and the mouse / cursor shape change explanatory diagram shown in FIG.

The user selects the command selection screen 2 shown in FIG.
When it is desired to select a desired operation button 23 on the display 2 by voice input, first, the pointing device (mouse) 5 is moved and operated, and the voice recognition button 24 on the selection screen 22 is selected.
4A, the mouse cursor 25 is moved, and the left button of the pointing device (mouse) 5 is pressed at that position to select the voice recognition button 24.

When the control section 3 detects that the voice recognition button 24 has been selected by operating the left button of the pointing device (mouse) 5 (step S2), FIG.
As shown in (b), the shape of the mouse cursor 25 on the voice recognition button 24 is changed (here, the pattern of the arrow is changed to a pattern indicating a microphone), and a display prompting the user to speak is displayed (step S3). .. The color may be changed without changing the cursor shape. Further, both the cursor shape and the color may be changed.

The user can display the display shown in FIG.
Recognizing that the start of voice input is prompted, a command name unique to the operation button 23 to be selected and specified is uttered.

The command name uttered by the user is input by the microphone 1 and sent to the voice recognition unit 2, where the recognition process is performed, and the recognition result (first to nth candidates) is controlled. It is passed to the section 3 (step S4).

When the control unit 3 receives the voice recognition result for the input voice (command name) from the voice recognition unit 2, the control unit 3 generates the position information of the operation button 23 corresponding to the recognition result (first candidate). , The mouse cursor 25 is moved to the position indicated by the information (that is, the position of the recognized operation button 23) (step S5), and as shown in FIG.
The shape of the cursor 25 is changed (the color may be changed) to prompt the user for confirmation processing (step S6).

The user determines from the position of the operation button 23 pointed by the mouse cursor 25 whether or not the recognition result is correct. Then, if the recognition result is correct, the user inputs a confirmation with the left button of the pointing device (mouse) 5 at the position of the operation button 23.

When the control section 3 detects that the left button of the pointing device (mouse) 5 has been pressed on the operation button 23 (in which the mouse cursor 25 is pointing) (step S7), the mouse Command processing corresponding to the operation button 23 pointed by the cursor 25 is executed (step S8).

On the other hand, if the recognition result is wrong, if the user wants to perform the erroneous recognition process by selecting the next candidate in (1), the user points the pointing device (mouse) on the erroneously recognized operation button 23. Press the 5 middle button.

When the control unit 3 detects that the middle button of the pointing device (mouse) 5 has been pressed on the operation button 23 (the state where the mouse cursor 25 is pointing) (step S9), the recognition result The mouse cursor 25 is moved to the operation button 23 corresponding to the next candidate (step S10). As a result, the user can perform the confirmation processing of the next candidate in the same manner as for the recognition result of the first place, and there is no need to re-voice.

Further, if the recognition result is wrong and if the erroneous recognition process by the re-voiced voice of (2) is desired,
The user presses the right button of the pointing device (mouse) 5 on the operation button 23 that has been erroneously recognized.

When the control unit 3 detects that the right button of the pointing device (mouse) 5 is pressed on the operation button 23 (the state where the mouse cursor 25 is pointing) (step S11), the mouse Cursor 25
To the voice recognition button 24 and change the cursor shape (or color) to the initial state (state before selecting the voice recognition button 24) shown in FIG. 4A (step S1).
2). As a result, the next voice input (re-voice) can be smoothly performed. The process of step S12 is also performed when the process corresponding to the recognized operation button 23 (step S8) is executed as a result of the recognition confirmation (step S7).

If the recognition result is wrong and the user wants to perform the erroneous recognition processing by the direct input in (3) above, the user can directly use the pointing device (mouse).
By moving 5 the mouse cursor 25 is moved onto the desired operation button 23, and the left button of the pointing device (mouse) 5 is pressed at that position.

When the control section 3 detects that the left button of the pointing device (mouse) 5 is pressed on the operation button 23 (the state where the mouse cursor 25 is pointing) (step S7), the above-mentioned operation is performed. As in the case where the left button is pressed while the mouse cursor 25 is moved to the corresponding operation button 23 according to the voice recognition result, command processing corresponding to the operation button 23 pointed to by the mouse cursor 25 is executed. Yes (Step S
8).

The present invention is not limited to the above embodiment, but can be variously modified and implemented without changing the gist thereof. For example, by displaying the voice recognition result for other than the first-ranked candidate, the selection of the next candidate can be smoothly processed. That is, the operation buttons 23 on the display screen respectively corresponding to the top N (N is an integer of 1 or more) of the second and subsequent candidates of the recognition result are displayed in a display mode specific to the candidate order, for example, by color coding. It may be displayed by reversing or blinking. In such a case, since the content of the next candidate can be confirmed, it becomes possible to smoothly judge which of the re-voice and the selection of the next candidate is more advantageous.

Further, what is pointed to by the mouse cursor according to the recognition result is not limited to the operation button 23 on the command selection screen 22, but the target of selection designation on the display screen, that is, the unique attribute name. The same applies to any display cell that has an attribute and is a target of selection specification. An example thereof is shown in FIG.

FIG. 5 is an example of a map screen on which buildings and places on the map are displayed, and what the mouse cursor points to according to the voice recognition result is the buildings and places on the map. Here, confirmation and erroneous recognition processing are performed on the screen of the building or place pointed by the mouse cursor. In addition, the same applies to the case where the destination (station) is designated by voice on the screen of the railway route map in which the fare is stated. In this case, what is indicated by the mouse cursor according to the voice recognition result is the station on the railway route map.

In the above embodiment, the pointing device 5 is assumed to be a mouse for the sake of simplicity, but other position indicating means such as a touch panel or a pen input device can also be applied. When a touch panel or a pen input device is used as the pointing device 5, in the function of the mouse button described in the above embodiment,
For example, by making the left button correspond to direct input to the screen with a finger or a pen, for the middle button and the right button, by displaying the corresponding operation button on the screen and selecting it directly with the finger or pen, It is possible to obtain the same operability as that of a mouse.

[0055]

As described above, according to the present invention,
Corresponds to the speech recognition result when the display cell that is the target of selection specification on the display screen such as the selector (operation button) on the command selection screen, the building, the place, etc. on the map screen is specified by voice. Since the position indicating display cursor that can be moved and operated by the position indicating means such as a mouse is moved to the display cell on the display screen to be displayed, the user can change the position indicating cursor of the position indicating cursor (the display cell thereof) from The recognition result for the voice uttered by the user can be intuitively determined, and quick recognition confirmation can be performed.

Further, according to the present invention, after the recognition result for the input voice is pointed by the position pointing display cursor, the input from the position pointing means is accepted, and the "confirmation" or "erroneous recognition process" of the recognition result is performed according to the input information. Especially, in the "erroneous recognition process", the position indication display cursor is moved to the display cell corresponding to the next candidate of the recognition result according to the operation content of the position indication means, and the voice is inputted by the position indication operation from the position indication means. Since the position indication display cursor is moved to a specific display cell whose start can be designated, or the display cell selection designation according to the position indication input from the position indication means is accepted, the erroneous recognition process for the voice recognition result is performed. Various operations and notifications prompting the user to speak voice input,
The position indicating means can be used quickly and smoothly without burdening the user.

As described above, according to the present invention, the user is urged to speak by moving the position indicating display cursor to the specific display cell in which the voice input start can be designated, and the recognition result is indicated by the position indicating display cursor to indicate the position. Confirm according to the operation of the means,
By performing the misrecognition process, it becomes possible to perform a quick and smooth voice input operation.
The interface can be dramatically improved, and a great effect can be obtained in practical use.

Further, according to the present invention, the input function by voice recognition can be easily incorporated into the application using the existing position pointing means (pointing device).

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a voice input operation device according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a screen configuration of a display screen in the embodiment.

FIG. 3 is a flowchart for explaining the operation of the embodiment.

FIG. 4 is a view for explaining a shape change of a mouse cursor in the same embodiment.

5 is a diagram showing another example of a screen configuration of a display screen applicable to the device of FIG.

[Explanation of symbols]

1 ... Microphone (voice input means), 2 ... Voice recognition unit,
3 ... Control unit, 4 ... Display unit, 5 ... Pointing device (position indicating means), 22 ... Command selection screen, 23 ...
Operation button (display cell), 24 ... Voice recognition button (specific display cell), 25 ... Mouse cursor (position indication display cursor).

Claims (7)

[Claims] 1. A display device having a unique attribute name and capable of displaying a plurality of display cells to be selected and designated, and an arbitrary position on a display screen of the display device is designated to position at the designated position. Position indicating means capable of moving the indication display cursor, voice input means for inputting voice, voice recognition means for recognizing voice input by this voice input means, and recognition by this voice recognition means And a control means for performing corresponding control including display control of the display device by receiving the input voice recognition result and the input information from the position indicating means, the control means on the display screen of the display device. When the attribute name of an arbitrary display cell is voice-input by the voice input unit while the plurality of display cells are being displayed, and the attribute name is recognized by the voice recognition unit. As a result of recognition in accordance with the display screen of the corresponding voice input operation apparatus characterized by moving the position indicating cursor on the display cell. 2. The control means receives an input from the position designating means after moving the position designating display cursor to the display cell, and “confirms” or “misrecognizes the recognition result” according to the input information. The voice input operation device according to claim 1, wherein 3. The control means, when the input information from the position designating means is predetermined first information, the position designating display cursor is displayed on the display cell corresponding to the next candidate of the recognition result. The voice input operation device according to claim 2, wherein the first erroneous recognition process of moving the voice input device is performed. 4. The control means, when the input information from the position pointing means is predetermined second information, performs a second erroneous recognition process for prompting a re-voice. The voice input operation device according to claim 3. 5. The second control unit displays in advance on the display screen a specific display cell in which a voice input start can be designated by being position-instructed by the position instructing unit, and urges the re-voice. 5. The voice input operation device according to claim 4, wherein in the erroneous recognition process, the position indicating display cursor is moved to the specific display cell. 6. The control means, when the input information from the position instructing means is predetermined third information, selects and designates the display cell according to the position instructing input from the position instructing means. The voice input operation device according to claim 5, wherein the third erroneous recognition process is performed. 7. The control means moves the position-indicating display cursor to a corresponding display cell on the display screen according to the recognition result, and at the same time, the upper N candidates (N is 1 or more) of subsequent candidates of the recognition result. 7. The voice input operation device according to claim 6, wherein the display cells corresponding to (integer) are displayed in a display form unique to the candidate rank.