JP2000267694A - Voice recognition device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice recognition device having a satisfactory operability in which even the canceling of a voice command can be performed by voice, which can by dispense with the operating of a canceling switch and which can eliminate the canceling key. SOLUTION: This device has a voice command group which has a hierarchical structure and includes a final voice command for commanding final equipment operations to respective hierarchies and an intermediate voice command for an intermediate command in which the selecting of final commands existing in lower hierarchies is needed in order to perform equipment operations. In each hierarchy of the voice command group, a voice command canceling the inputted voice command or a voice command making voice recognition complete is included.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition apparatus for controlling the operation of a device based on a voice command, and more particularly to a voice recognition apparatus having a hierarchical structure of voice commands.

[0002]

2. Description of the Related Art A speech recognition apparatus is a system which can identify a voice command from an instruction input by a user's voice and send an operation signal based on the voice command to the apparatus, thereby controlling the operation of the apparatus only by the user's voice.

[0003] A car navigation system is an example of a speech recognition device currently on the market. The car navigation system is a system for setting and inputting a destination of an automobile, thereby displaying a route from a current traveling position obtained by using the GPS to the destination on a map on a dedicated screen to provide road guidance. As a destination setting method, voice input is used instead of the button input operation or panel touch operation that was originally used, and voice guidance of the destination is used, so that road guidance to the destination is automatically displayed on the screen. Display can be performed, and operability is further improved.

[0004] Generally, in order to simplify the commands so that the driver can easily understand the operation of the equipment and to improve the recognition rate of the voice command by the voice recognition device, the voice recognition device is used in the voice recognition device. The voice command has a hierarchical structure. For this reason, it is necessary to sequentially input a plurality of voice commands before final device operation control is performed.

FIG. 4 shows an example of a hierarchical structure of a voice command used for a voice recognition device for controlling an air conditioner and an audio device for a vehicle. As air-conditioning control for automobiles, switching of blow-off outlets (FACE, FOO)
T, DEF, etc.), switching between inside and outside air, temperature setting, ON / OFF of an air conditioner, and the like. Audio control includes switching between compact discs, tapes, AM / FM radios, etc., reproduction of compact discs and tapes, radio selection, volume control, and the like. On the first level, "air conditioning" and "audio""frontdefroster""insideair""outsideair"
There is a selection command called. "Air conditioning" and "Audio"
Is not an operation instruction to the final device, but an intermediate command that requires selection of a branched voice command in the second and subsequent layers. In the second and subsequent hierarchies, a final branched operation instruction command and an intermediate command having a further lower-layer branched command are mixed. On the first level, operation instruction commands for final equipment that the driver wants to use immediately, such as “front defroster”, “inside air”, and “outside air” are also prepared.

FIG. 5 is a block diagram showing a schematic configuration of a conventional voice recognition device used for air conditioning and audio of a car. The speech recognition device 200 includes a speech recognition unit, a speech synthesis unit 110, and a control unit 120. The voice recognition unit of the voice recognition unit and the voice synthesis unit 110 is connected to the microphone 130 for the driver to input a command by voice, and the voice synthesis unit is connected to the speaker 140.
The control unit 120 is connected with a recognition start switch 150 for starting voice recognition and a cancel switch 160 for canceling a command by a voice command. The control unit 120 is connected to the air conditioning 170 and the audio 180 so as to output an operation control signal.

[0007] When the driver presses the recognition start switch 150, a recognition start signal is sent to the control unit 120, and voice recognition is started. First, a guide voice prompting selection of a voice command of the first hierarchy is played from the speaker 140. The guide voice is synthesized by the voice synthesizer. The driver uses the microphone 130
When a voice command is selected and answered, the voice recognition unit 110 compares the voice command with a voice command stored and registered in a storage unit provided in the control unit 120.
If the input voice command is not recognized as any of the first-level voice commands, the input is prompted again by the guide voice as an error in the voice command input. When the voice command is recognized as a first-layer voice command, a guide voice prompting selection of a second-layer voice command is further played. When the final operation instruction command is recognized,
An operation control signal is sent from the control unit 120 to the air conditioner 170 or the audio 180, and an operation desired by the driver is executed.

At the time of voice command input, when the driver erroneously answers a voice command which is different from the desired voice command and which is registered in advance, or the voice command is a desired voice command, If the device erroneously recognizes another voice command, a device operation not desired by the driver will be performed. Therefore, a cancel button for canceling the voice command is provided, and when the driver presses the button, the immediately preceding voice command is canceled.

Hereinafter, a cancel operation procedure will be described.

[0010] 1. Driver: Press the recognition start switch.

2. Apparatus: Plays a guide to start recognition. "Start recognition process. Do you operate air conditioning or audio?" Driver: Voice command of the first layer is input by voice.
"Air conditioning" Apparatus: prompts the user to select the second level voice command.
"Perform the air conditioning operation. Please select the operation item." Driver: Here, since "front differential" on the first level is originally desired, the cancel button is pressed to cancel the voice command on the first level.

6. Apparatus: Return to the first level,
Prompt for selection of hierarchical voice command. "Start recognition process. Do you operate air conditioning or audio?" Driver: Voice command of the first layer is input again. "Front differential" Apparatus: Flow the operation guide for starting the front defroster and end the operation. "Remove the fogging of the windshield."

However, in the conventional voice recognition apparatus having a cancel switch, when canceling a voice command, the user must press the cancel switch each time the command is canceled, which is troublesome. In this case, there is a problem that the advantage of the voice recognition device that the device can be operated only by voice input is halved.

Accordingly, it is an object of the present invention to provide a voice recognition apparatus capable of canceling voice commands by voice, eliminating the need for cancel switch operation, and eliminating the cancel switch. I do.

[0014]

SUMMARY OF THE INVENTION The object of the present invention is to control the operation of a device based on a voice command.
A voice command group having a hierarchical structure, in which it is necessary to select a final voice command for a final device operation command in each layer and a final voice command existing in a lower layer for device operation. A voice command group including an intermediate voice command for a command, and each layer of the voice command group includes a voice command for canceling the input voice command or a voice command for ending the voice recognition process, Achieved effectively.

A voice recognition apparatus for controlling the operation of a device based on a voice command is a voice command group having a hierarchical structure, wherein a final voice command for giving a final device operation command to each layer. Storage means for storing a voice command group including an intermediate voice command for an intermediate command, and voice recognition means for recognizing the input voice command by comparing the voice input of the user with the voice command in the storage means When the input voice command is an intermediate voice command, in order to select a lower-layer voice command, guidance means for prompting voice input using a synthesized voice synthesized by the voice synthesis unit, and the input voice command Control means for sending an operation command signal to the corresponding device in the case of the final voice command, and the input voice command is stored in each layer of the voice command group. By including a voice command to terminate a voice command or voice recognition processing erasing are achieved more effectively.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 Hereinafter, a first embodiment of a speech recognition device according to the present invention will be described with reference to the drawings. FIG. 1 is an overall configuration diagram showing an embodiment of a voice recognition device according to the present invention, which controls the air conditioning of a car and the operation of audio by voice commands. The speech recognition device 100 includes a speech recognition unit 10, a speech synthesis unit 15, and a control unit 20. The voice recognition unit 10 is connected to a microphone 30 for the driver to input voice commands, and the voice synthesis unit 15 is connected to a speaker 40. A recognition start switch 50 for starting speech recognition is connected to the control unit 20. The driver has the microphone 30
When a voice command is input toward the voice recognition unit 10
Then, the voice command stored in the storage unit provided in the control unit 20 is compared with the registered voice command, the input voice command is recognized, and an operation control signal is transmitted from the control unit 20 to the air conditioner 70 or the audio 80. Then, the operation desired by the driver is performed.

In the present embodiment, unlike the prior art, there is no cancel switch for canceling a command by a voice command.

FIG. 2 is a system diagram showing an example of a command group having a hierarchical structure for controlling the operation of air conditioning and audio in the embodiment of the present invention shown in FIG. On the first level, “air conditioning”, “front differential”, “front differential off”, “audio”, and the like are arranged. “Air conditioning” and “audio” are not final instruction commands, but are intermediate commands having options in the second hierarchy. Other commands are final instruction commands that directly instruct the operation of the device. Also, a command of “processing end” is provided. When the command of “processing end” is selected, all the voice recognition operations end.

In the second hierarchy, "set temperature", "air conditioner", "rear differential", etc., which are intermediate commands of the first hierarchy, "air conditioning", etc., and a selection command of an intermediate command, "audio", are provided. A CD changer, a tape, an FM, an AM, and the like are arranged. Here, "set temperature", "tape", "FM" and "AM" are intermediate commands. The second layer has commands of “processing end” and “cancel”, respectively. The “cancel” command is a command for canceling the command command of the previous layer.
Hereinafter, a third layer, a fourth layer, and a command group follow.

Next, a procedure for canceling a command command according to this embodiment will be described.

1. Driver: Press the recognition start switch.

2. Apparatus: Plays a guide to start recognition. "Start recognition process. Do you operate air conditioning or audio?" Driver: Voice command of the first layer is input by voice.
"Air conditioning" Apparatus: prompts the user to select the second level voice command.
"Perform the air conditioning operation. Please select the operation item." Driver: Here, since "front differential" on the first level is originally desired, "cancel" is uttered to cancel the voice command on the first level.

6. Apparatus: Return to the first level,
Prompt for selection of hierarchical voice command. "Start recognition process. Do you operate air conditioning or audio?" Driver: Voice command of the first layer is input again. "Front differential" Apparatus: Flow the operation guide for starting the front defroster and end the operation. "Remove the fogging of the windshield." Next, the operation of the voice recognition device of the present invention by a series of voice command operations following the above operation procedure is shown in FIG.
The operation will be described with reference to the operation flowchart shown in FIG.

When the driver presses the recognition start switch 50, the system starts operating, and the variable N is set at step S10.
, And then guide the N-layer guide in step S11. Since N = 1 at the moment, a one-layer guide, that is, "Start recognition processing. Which air conditioning / audio should be operated?" In accordance with this guide, when the driver voice-inputs "air conditioning" and the first hierarchical voice command in step S12, it is determined in step S13 whether "air conditioning" is an N-layer command. In this case, the process returns to step S11. If the command is an N-layer command, the process proceeds to step S14, where it is determined whether or not the voice input is an end command. If the command is not the end command, the process proceeds to step S15, and it is determined whether the command is a cancel command. Similarly, whether the command is an intermediate command is determined in step S16. Execute the command of For example, when N = 1, the “air conditioning” command of the first level is executed.

If it is determined in step S16 that the voice input is an intermediate command, the flow advances to step S18, where the variable N is incremented by 1, and the flow returns to step S11. If the voice input is a cancel command in step S15, step S1
The process proceeds to step 9, where the variable N is decremented by one, and then returns to step S11.
In this way, the voice commands of the second and third layers are sequentially selected and executed. If there is a speech input of the cancel command in the middle, the process returns to the first level and the steps from S11 are executed again.

After executing the command of the N-th layer in step S17, it is determined in step S20 whether or not a predetermined time has elapsed, and in step S21, whether or not a cancel voice has been input. If YES in step S20, If NO in step S21, the series of operations is terminated. Otherwise, the process proceeds to step S22, where the execution of the command of the N layer is stopped, and the process returns to step S11 to guide the N layer.

As described above, according to the configuration of the present embodiment, the operation of pressing the cancel switch becomes unnecessary, and the cancel switch can be omitted, so that the operability is improved and the number of parts can be reduced. In addition, since the “processing end” command is provided at each level, the voice recognition processing can be ended at any stage of the voice recognition operation, so that it is troublesome to move the hand to cancel the processing operation. Can be eliminated.

The words of the voice command for canceling are preferably longer words which are not confused with the words of other voice commands. This is because erroneous recognition is less likely to occur. For example, it is preferable that “speech recognition process is completed” and “hierarchy up” is more than “cancel” than “process end”.

Embodiment 2 Each layer of the voice command group includes a voice command for canceling the input voice command or a voice command for ending the voice recognition process, and a cancel switch for canceling the input voice command is provided. The cancel operation may be performed in any of the above.

Embodiment 3 Further, the configuration of the present invention can be applied to an apparatus in which voice recognition is started by voice input instead of the recognition start switch. In that case, it becomes possible to perform air conditioning or audio operation completely hands-free. Further, even when a command to start "recognition" is given by mistake, the present invention is particularly effective because it is not necessary to search for the cancel switch in a hurry, and it is possible to immediately cancel by voice.

Embodiment 4 FIG. The configuration of the present invention can be applied to a car navigation system, a television, a telephone with a hands-free function, and the like, in addition to an air conditioner and an audio device of a car.

[0032]

The speech recognition apparatus according to the present invention has a command for canceling the speech command of the previous layer or a command for terminating speech recognition for terminating speech recognition for each layer of the speech command group having a hierarchical structure. However, by canceling the voice command by voice, the cancel switch is eliminated, and the operation of canceling the voice command by the cancel switch becomes unnecessary, and the operability in the voice recognition processing can be significantly improved.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an overall configuration of a voice recognition device and peripheral devices according to a first embodiment of the present invention;

FIG. 2 is a system diagram showing an example of a command group having a hierarchical structure for performing air conditioning and audio operation control according to the embodiment of the present invention shown in FIG. 1;

FIG. 3 is a diagram showing an operation flow of the speech recognition device of the present invention.

FIG. 4 is a system diagram showing an example of a command group for performing air conditioning and audio operation control in a conventional voice recognition device.

FIG. 5 is a block diagram showing the overall configuration of a conventional voice recognition device and its peripheral devices.

[Explanation of symbols]

10 voice recognition unit, 15 voice synthesis unit, 20 control unit,
30 microphones, 40 speakers, 50 recognition start switch, 70 air conditioning, 80 audios.

Claims (2)

[Claims] 1. A voice recognition apparatus for controlling operation of a device based on a voice command, comprising: a voice command group having a hierarchical structure, wherein a final voice command for a final device operation command is provided in each layer; And a voice command group including an intermediate voice command for an intermediate command necessary to select a final voice command existing in the lower layer for operation. A voice command for canceling the voice command or a voice command for terminating the voice recognition process. 2. A voice recognition apparatus for controlling operation of a device based on a voice command, comprising: a voice command group having a hierarchical structure, wherein a final voice command for a final device operation command is provided in each layer; Storage means for storing a voice command group including an intermediate voice command for an intermediate command required to select a voice command for a final device operation command existing in a lower layer for operation; Voice recognition means for recognizing the input voice command by comparing the voice input with the voice command in the storage means; and, when the input voice command is an intermediate voice command, a voice for selecting a lower-layer voice command. Guidance means for prompting voice input using the synthesized voice synthesized by the synthesizer, and when the input voice command is the final voice command, the corresponding device operates And a control means for sending a decree signals, each layer of the voice commands, the voice recognition device characterized by comprising a voice command to terminate a voice command or speech recognition process canceling a voice command input.