WO2002035518A1

WO2002035518A1 - On-vehicle speech recognizing apparatus

Info

Publication number: WO2002035518A1
Application number: PCT/JP2001/008945
Authority: WO
Inventors: Noriyuki Tsunashima
Original assignee: Kabushiki Kaisha Kenwood
Priority date: 2000-10-24
Filing date: 2001-10-11
Publication date: 2002-05-02
Also published as: JP2002132290A

Abstract

An on-vehicle speech recognizing apparatus basically has a dictation function by speech recognition and a function of controlling a device. The apparatus controls a device by using a command control dictionary before recognition of a dictation start word and carries out speech recognition by using a dictation dictionary after recognition of a dictation start word until recognition of a dictation end word so as to perform dictation. When a signal for judging whether a vehicle is traveling or not is inputted and if the vehicle is judged to be traveling, the dictation start word is deleted from the command control dictionary, and the device is controlled according to the command control dictionary. If the vehicle is judged to be not, the dictation start word is added to the command control dictionary.

Description

Technical field

The present invention relates to voice recognition having a dictation function and a device control function, and more particularly to a voice recognition device suitable for use in a vehicle.

Light

Fine

Landscape technology

FIG. 8 shows an example of a conventional speech recognition device having a dictation function and a device control function. FIG. 8 (a) is a block diagram showing an outline of a hardware configuration of a conventional speech recognition device. As shown in this figure, the prior art device basically converts a microphone into an electric signal, an amplifier to amplify the microphone output, an amplified analog signal, and a digital signal. An AZD converter and a microcomputer (CPU) that processes digital signals

It consists of dictionary memories 3a and 3b (for command control and for dictation) and a display device. In operation, the digital signal input to the microcomputer is compared with the voice pattern stored in the command control dictionary memory 3a or the dictionary memory 3b for dictation, and the input voice is recognized. A device (not shown) is controlled by the voice command recognized as described above. In the case of a dictation, the device is converted into characters and displayed on the display device 7.

FIG. 8 (b) is a diagram showing a conventional speech recognition device in functional blocks. As shown in the figure, the voice input to the voice input unit 1 is compared with the voice pattern stored in the command control dictionary 3a or the dictation dictionary 3b by the voice recognition unit 2 and input. Voice is recognized. The command by the voice recognized in this way controls the corresponding operation to the control target device 5 by the command control unit 4, and in the case of a dictation, the voice recognized by the text conversion unit 6 is used. It is converted to a character string and displayed on the text display section 7. The operation of the speech recognition device will be described more specifically with reference to FIG. When the voice recognition software is started, first, in step S20, it is determined whether or not a voice has been input, and when the voice has been input, the process proceeds to step S21. In step S21, voice recognition processing is performed by comparing the voice pattern stored in the command control dictionary 3a with the pattern of the input voice.

The command control dictionary 3a contains a dictation start word. Next, proceed to step S22, and determine whether the voice pattern recognized in step S21 is a dictation start word, and if it is a dictation start word, proceed to step S24. I do. If the speech pattern recognized in step S21 is not a dictation start word, the process proceeds to step S23.

In step S23, control corresponding to the command obtained from the recognition result is performed, and steps S20 and subsequent steps are repeated.

In step S24, it is determined whether or not a voice has been input, and when a voice has been input, the process proceeds to step S25. In step S25, voice recognition processing is performed by comparing the voice pattern stored in the dictation dictionary 3b with the input voice pattern, and the process proceeds to step S26.

The dictation dictionary 3b contains a dictation end word, and in step S26, it is determined whether or not the voice pattern recognized in step S25 is the dictation end word. If the word is a dictation end word, the flow shifts to step S20 to end the dictation. If the voice pattern recognized in step S25 is not the dictation end word, the process proceeds to step S27. In step S27, the voice pattern obtained from the recognition result is converted into characters, and in step S28, the text data obtained in S27 is displayed on the text display section 7, and then in step S24. The following is repeated.

When the above-mentioned conventional speech recognition is used while driving a vehicle for in-vehicle use, there is a possibility that the voice uttered to control the device may be erroneously recognized as a dictation start keyword (dictation start word) due to noise. is there. In the dictation state, noise resistance is lower than that of voice recognition for commands, it is not suitable for use under the noise while driving a car, and it is difficult to recognize the dictation end word. There is a danger that it will not be possible to get out of the dictation state, in which case the dictation state will be created, resulting in confusion for the user.

Also, when used by the driver, the dictation displays the contents of the utterance and utters long sentences, which hinders the driver's concentration. Dictation causes a hindrance to safe driving, and dictation during driving should be prohibited.

The present invention has been made in view of the above points, and an object of the present invention is to provide an in-vehicle voice recognition device in which dictation is prohibited while a vehicle is running. Disclosure of the invention

In order to achieve the above object, an in-vehicle voice recognition device of the present invention has a dictation function by voice recognition, a device control function, and means for determining whether or not a vehicle is running. If it is determined that the vehicle is running, the dictation function is disabled, and only the control function of the device is enabled. If it is determined that the vehicle is not running, the dictation function is disabled. It is configured to operate to enable both the application function and the control function of the device. The operation of disabling the dictation function is performed by an operation for preventing the start of dictation. More specifically, the device of the present invention comprises a command control dictionary used before dictation starts, and a dictation dictionary used after dictation starts or before ending. The operation for preventing the dictation from being started is performed by invalidating or deleting the dictation start word in the command control dictionary. The operation of preventing the start of the dictation may be performed by disabling the dictation start switch. Further, in the device of the present invention, when it is determined that the vehicle is not running after the dictation function is disabled, the dictation start word of the command control dictionary is enabled or added. Has become. On the other hand, whether or not the vehicle is running is determined based on one of the following signals indicating that the vehicle is running, which is output from an external accessory device. (i) Signal output from parking switch

(ii) Signal output from vehicle speed sensor

(i i i) Signal output from GPS receiver

Since the vehicle-mounted speech recognition device of the present invention is configured as described above, according to the present invention, (i) a dictation state does not occur due to erroneous recognition during traveling, and a command is always issued during traveling. Will not cause confusion for the user, and

(i i) Since the dictation operation is prohibited while the vehicle is running, operational effects such as ensuring safe driving can be obtained. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a block diagram showing a configuration of a vehicle-mounted voice recognition device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the speech recognition apparatus according to the first embodiment of the present invention.

FIG. 3 is a block diagram showing a configuration of a vehicle-mounted speech recognition device according to a second embodiment of the present invention.

FIG. 4 is a front view showing the operation of a vehicle-mounted speech recognition apparatus according to a second embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration of a vehicle-mounted speech recognition device according to a third embodiment of the present invention.

FIG. 6 is a flowchart showing the operation of the vehicle-mounted voice recognition device according to the third embodiment of the present invention.

FIG. 7 is a block diagram showing a configuration of a vehicle-mounted speech recognition device according to a fourth embodiment of the present invention.

No. 80 is a block diagram showing a configuration example of a conventional speech recognition device.

FIG. 9 is a flowchart showing an operation example of the conventional speech recognition apparatus. Embodiment of the Invention

Hereinafter, an embodiment of a vehicle-mounted speech recognition device according to the present invention will be described in detail with reference to the drawings. explain.

FIG. 1 is a block diagram showing a speech recognition apparatus according to a first embodiment of the present invention. Here, FIG. 1 (a) is a block diagram showing an outline of the hardware configuration. Compared with the configuration of the conventional technology shown in FIG. A distinctive feature is that a switch is added.

As in the prior art device, the sound input to the microphone is converted to an electric signal, amplified by an amplifier, and converted to a digital signal by an A / D converter.

The digital signal is input to the microcomputer, and is compared with the voice pattern stored in the command control dictionary memory 3a or the dictionary memory 3b for dictation to recognize the input voice. A device (not shown) is controlled by the voice command recognized in this way, and in the case of a dictation, it is converted into characters and displayed on the display device 7.

During the microcomputer, signals from the power switch 8 and the like are input, and when the parking switch 8 is on, that is, when the vehicle is stopped, the dictation start word is stored in the command control dictionary memory 3a. When the perkinda switch 7 is off, that is, during operation, the dictionary start word is deleted from the command control dictionary memory 3a.

FIG. 1 (a) is a diagram showing the configuration of a device according to a first embodiment of the present invention by functional blocks.

The voice input to the voice input unit 1 shown in the figure is compared with the voice pattern stored in the command control dictionary 3a or the dictionary for dictation 3b by the voice recognition unit 2 and input. Voice is recognized. The command based on the voice recognized in this way controls the corresponding operation of the control target device 5 by the command control unit 4 or, in the case of dictation, the voice recognized by the text conversion unit 6. The text is converted to a character string and displayed on the text display section 7. A signal from the parking switch 8 is input to the voice recognition unit 2 to detect whether or not the vehicle is running. Next, the operation of the speech recognition apparatus will be described with reference to FIG. When the voice recognition software is started, first, in step S1, it is determined whether or not a voice is input, and when the voice is input, the process proceeds to step S2. Step S 2 is for command control The voice recognition process is performed by comparing the voice pattern stored in the dictionary 3a with the input voice pattern.

The command control dictionary 3a includes a dictation start word. Next, the process proceeds to step S3, and it is determined whether or not the voice pattern recognized in step S2 is a dictation start word, and if it is a dictation start word, the process proceeds to step S5. If the voice pattern recognized in step S2 is not a dictation start word, the process proceeds to step S4.

In step S4, control corresponding to the command obtained from the recognition result is performed, and steps S1 and subsequent steps are repeated.

In step S5, the state of the parking switch 8 is determined, and if it is determined that the vehicle is on, that is, the vehicle is stopped, the dictation processing of step S6 and subsequent steps is executed. Step S1 and subsequent steps are repeated to invalidate the dictation function.

In step S6, it is determined whether or not a voice has been input, and when the voice has been input, the process proceeds to step S7. In step S7, voice recognition processing is performed by comparing the voice pattern stored in the dictionary for dictation 3b with the input voice pattern, and the process proceeds to step S8.

The dictation dictionary 3b contains a dictation end word. In step S8, it is determined whether or not the voice pattern recognized in step S7 is a dictation end word. If it is not an end word, the process proceeds to step S9. In step S9, the voice pattern obtained from the recognition result is converted into characters, and in step S10, the text data obtained in S9 is displayed on the text display unit 7, and then the dictation in step S5 and below is performed. The function is repeated. If the voice pattern recognized in step S7 is the dictation end word, the process proceeds to step S1 to end the dictation.

As described above, when the parking switch 8 is off, that is, during driving, there is no transition to the dictation operation, so that confusion such as a dictation state against the user's intention does not occur. Also, since dictation is prohibited while driving, the collection of drivers by using the dictation function while driving is prohibited. Safe driving is ensured because it is possible to prevent a decrease in medium power.

FIG. 3 is a diagram showing a function block of a vehicle-mounted speech recognition device according to a second embodiment of the present invention. In this example, instead of starting the dictation by recognizing the dictation start word in the first embodiment, it is determined whether the dictation switch 9 controls the command or performs the dictation. The operation of the vehicle-mounted speech recognition device according to the second embodiment will be described with reference to FIG. When the voice recognition software is started, first, the state of the dictation switch 9 is detected in step S11, and if ON, that is, the dictation is selected, the process proceeds to step S15. If it is off, that is, if the dictation has not been selected, the process proceeds to step S12.

In step S12, it is determined whether or not a voice has been input, and when the voice has been input, the process proceeds to step S13. In step S13, a voice recognition process is performed by comparing the voice pattern stored in the command control dictionary 3a with the input voice pattern. In step S14, control corresponding to the command obtained from the recognition result is performed, and step S1 and subsequent steps are repeated.

In step S15, the state of the parking switch 8 is determined, and if it is determined that the parking switch 8 is off, that is, the vehicle is running, steps S1 and subsequent steps are repeated to disable the dictation function. If it is determined that the voice is in the middle, the process proceeds to step S16.If there is a voice input in step S16, in step S17, the voice pattern is compared with the voice pattern stored in the dictation dictionary 3b. Performs speech recognition processing for dictation. In step S18, the voice pattern obtained from the recognition result is converted into characters, and in step S19, the text data obtained in S18 is displayed on the text display unit 7, and then in step S1 and below. Is repeated.

As described above, even if the dictation switch 9 is turned on, that is, dictation is selected, the parking switch 8 is off, that is, the vehicle does not shift to the dictation operation while the vehicle is running. There is no confusion because no dictation occurs against the intention of the user. Also, dictation operation is prohibited while driving, so during driving By using the dictation function, it is possible to prevent the driver's concentration from being reduced, thus ensuring safe driving.

FIG. 5 (a) is a block diagram showing an outline of the construction of an eighty-one window of a vehicle-mounted speech recognition apparatus according to a third embodiment of the present invention. In this example, instead of the output signal of the parking switch 8 in the first embodiment, operation is being performed from the speedometer 10.

A signal indicating whether or not the speed is not 0 is input to the microcomputer, the dictation start word is deleted from the command control dictionary memory 3a during driving, and the command control dictionary memory 3 during stoppage. A dictation start word is added to a. Other configurations and operations are the same as those of the first embodiment.

FIG. 5 (b) is a functional block diagram of the vehicle-mounted speech recognition device according to the third embodiment of the present invention. In this example, a vehicle speed sensor 110 is used instead of the output signal of the parking switch 8 in the first embodiment, a signal indicating the vehicle speed from the vehicle speed calculation unit 11 is input to the voice recognition unit, and the vehicle travels. A determination is made whether it is in the middle. Other configurations are the same as in the first embodiment.

FIG. 6 is a flowchart for explaining the operation of the vehicle-mounted speech recognition apparatus according to the third embodiment of the present invention. In this example, instead of using the parking switch 8 in step S5, the vehicle speed sensor 110 is used to determine whether the vehicle is traveling at the vehicle speed from the vehicle speed calculation unit 11 or is stopped. Other operations are the same as in the first embodiment. FIG. 7 (a) is a block diagram showing an outline of a hardware configuration of a vehicle-mounted speech recognition device according to a fourth embodiment of the present invention. In this example, instead of the output signal of the parking switch 8 in the first embodiment, a signal from the GPS (global positioning system) 12 indicating that the vehicle is in operation (speed is not 0) Is input to the micro computer 4 and the dictation start word is deleted from the command control dictionary memory 3a during driving, and the dictation start word is stored in the command control dictionary memory 3a while the vehicle is stopped. Will be added. Other configurations and operations are the same as those of the first embodiment.

FIG. 7 (b) is a functional block diagram of a vehicle-mounted voice recognition device according to a fourth embodiment of the present invention. In this example, a GPS (global positioning system) 12 is used in place of the output signal of the parking switch 8 in the first embodiment, and The position calculation unit 13 calculates the vehicle speed from the moving distance, outputs a signal indicating the vehicle speed to the voice recognition unit, and determines whether the vehicle is running by the voice recognition unit. Other configurations and operations are the same as those of the third embodiment. Industrial applicability

ADVANTAGE OF THE INVENTION According to the vehicle-mounted speech recognition device of this invention, a dictation state does not go against a user's intention while driving, and an operation corresponding to a command is always performed during driving. No confusion.

In addition, since the dictation operation is prohibited while the vehicle is running, the use of the dictation function during driving does not hinder safe driving.

Claims

The scope of the claims

1. A vehicle-mounted voice recognition device having a dictation function by voice recognition and a device control function,

A means for determining whether or not the vehicle is running; if it is determined that the vehicle is running, the dictation function is disabled and only the control function of the device is enabled; If it is determined that the vehicle is not traveling, the operation is performed so as to enable both the dictation function and the control function of the device.

2. The apparatus according to claim 1,

The operation of disabling the dictation function is performed by an operation that prevents the dictation from starting.

3. The apparatus according to claim 2,

A command control dictionary for use before dictation starts, and a dictation dictionary for use after dictation starts or before ending, and

The operation of preventing the dictation from being started is performed by invalidating or deleting the dictation start word in the command control dictionary.

4. The apparatus according to claim 2,

The operation for preventing the dictation from being started is performed by disabling the dictation start switch.

5. The apparatus according to claim 3,

An in-vehicle voice recognition device for enabling or adding the dictation start word of the command control dictionary when it is determined that the vehicle is not running after the dictation function is disabled.

6. The apparatus according to claim 1,

The in-vehicle voice recognition device determines whether or not the vehicle is traveling based on a signal indicating that the vehicle is traveling, which is output from an external attached device.

7. The apparatus according to claim 6, wherein

The signal indicating that the vehicle is running is output from the parking switch.

8. The apparatus according to claim 6, wherein

A vehicle-mounted voice recognition device, wherein the signal indicating that the vehicle is running is output from a vehicle speed sensor.

9. The apparatus according to claim 6, wherein

A vehicle voice recognition device, wherein the signal indicating that the vehicle is running is output from a GPS receiver.