JP2008242261A - Speech recognition device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a speech recognition device that can reduce the possibility that driver's concentration on driving operation declines as a result of output of a recognition result, and facilitates canceling operation even if the recognition result is wrong. <P>SOLUTION: It is judged whether a speech that a user utters is output as character data (recognition result) (S10) and when the recognition result is output and not a priority command, it is decided whether the user should concentrate on the driving operation (S40, S50) from a travel-related signal showing the travel state of the vehicle and state information output from an external state acquisition device 52. When it is decided that the user should concentrate on the driving operation, the recognition result is not output and a judgment of whether the user should concentrate on the driving operation is repeatedly made. When it is judged that the user need not concentrate on the driving operation any more, the recognition result is output. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a speech recognition device mounted on a vehicle, and particularly relates to an output timing of a speech recognition result.

  Voice recognition devices are widely used to operate various devices in situations where switch operation is difficult. Since it is difficult to operate the switch while driving the vehicle, the voice recognition device is mounted on the vehicle and used to operate a navigation device or the like (for example, Patent Document 1).

  The speech recognition apparatus outputs the recognized character data to the operation target device, and the device to which the character data is input executes a function determined by the input character data. A speech recognition device is also widely known in which recognition results, that is, recognized character data are output to a speaker or a display.

The speaker confirms whether or not the word he / she spoke is correctly recognized by confirming the operation status of the device or the output content from the speaker or display. If the recognized character data is different from the spoken word, the utterance is executed after performing a predetermined cancel operation, and the voice recognition process is executed again.
JP 2006-184421 A

  While the vehicle is driving, the situation changes every moment, and there is a situation that should be concentrated on the driving operation, such as during the steering operation. Even in this situation, when the operation of the device based on the output of the voice recognition result is performed, the driver must check whether the recognition result is correct. In other words, when the voice recognition result is output to the operation target device and the device operates based on the result, it is necessary to check whether the operated device is the intended device and the operation content. When the character data thus output is output from a speaker or a display, the voice output from the speaker or the character displayed on the display must be confirmed. For this reason, in a situation where the driver should concentrate particularly on the driving operation, the attention required for the driving operation may be insufficient.

  In particular, if the recognition result is a misrecognition, as described above, the utterance must be executed again after canceling, so the attention required for driving operation is particularly insufficient. There is a possibility.

  In addition, even if it turns out that it is a misrecognition, it is necessary to concentrate on the driving operation, so the canceling operation and the subsequent utterance cannot be executed, and the speech recognition operation must be performed again from the beginning. It may be necessary.

  The present invention has been made based on this circumstance, and the purpose of the present invention is to reduce a reduction in attention to driving operation by outputting a recognition result, and the recognition result is An object of the present invention is to provide a voice recognition device that can easily perform a cancellation operation even in the case of erroneous recognition.

  In order to achieve the object, the invention according to claim 1 is directed to a recognition processing means for extracting the inputted voice as character data by executing a recognition process when a voice spoken by a user is input, and the recognition processing means. A speech recognition apparatus comprising: an output control means for outputting a recognition result to a predetermined device, further comprising a determination means for sequentially determining whether or not the driving operation requires attention; and the output control means Is based on the fact that the determination means determines that the situation requires attention to the driving operation, and delays the timing of outputting the recognition result as compared with the case where it is not determined that the situation requires attention to the driving operation. It is characterized by that.

  According to the first aspect of the present invention, it is determined whether or not the driving operation requires attention. When it is determined that the driving operation requires attention, the timing for outputting the recognition result is delayed. As a result, the probability that the time when the recognition result is output is a situation that requires attention to the driving operation decreases. Therefore, it is possible to reduce the lack of attention to the driving operation by outputting the recognition result, and the time when the recognition result is output is a time when the driving operation has a relatively large margin. Since the possibility increases, even when the recognition result is a misrecognition, the cancel operation can be easily performed.

  Here, the time when the recognition result is output may be delayed by a predetermined time from the time when it is not determined that the driving operation requires attention. It is preferable to determine the output time based on the determination result.

  The invention according to claim 2 is a situation in which, in claim 1, the output control means requires attention to the driving operation by the determination means when it is determined that the determination means requires the driving operation. The recognition result is output after it is not determined that it is.

  In this way, the voice recognition result is not output while it is determined that the driving operation requires attention. Therefore, it is possible to prevent the attention to driving operation from becoming insufficient by outputting the recognition result. In addition, since the time when the recognition result is output is a time when the driving operation has a relatively large margin, even when the recognition result is erroneous recognition, the canceling operation can be performed particularly easily.

  According to a third aspect of the present invention, in the second aspect, the output control unit is configured such that when the elapsed time after the voice is recognized by the recognition processing unit exceeds a preset allowable standby time, Regardless of the determination result in the determination means, the recognition result is output.

  In this way, it is possible to avoid the inconvenience that the situation in which the judgment means judges that the situation requires the driving operation is not resolved for a long period of time, and the recognition result cannot be output for a long period of time.

  According to a fourth aspect of the present invention, in the speech recognition apparatus according to any one of the first to third aspects, the output control means is a content to which the character data extracted by the recognition processing means is to be output quickly. Is set in advance, the recognition result is output at the same time as when it is not determined that the driving operation requires attention regardless of the determination result by the determination means. In this way, it is possible to prevent the output of the recognition result that requires quickness from being delayed.

  The determination means may determine whether or not the driving operation requires attention based on a traveling-related signal indicating a traveling state of the vehicle. As described in 7, whether or not the situation information of the road on which the vehicle is traveling is obtained from the external situation acquisition device mounted on the vehicle, and the driving operation needs attention based on the obtained information. It is good also as determining. Moreover, you may determine whether it is a condition which requires attention for driving operation using both the driving | operation related signal which shows the driving | running state of a vehicle, and the situation information from an external condition acquisition apparatus.

  As the travel-related signal, there is a signal indicating a steering angle, and in claim 6, it is determined that the driving operation needs attention based on the fact that the steering operation can be determined from the signal indicating the steering angle. is doing. In this way, since the output of the voice recognition result is delayed during the steering operation, the voice recognition result is output during the steering operation, thereby reducing the decrease in concentration on the steering operation. Can do.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a navigation device 10 having a function as a voice recognition device of the present invention.

  The navigation device 10 includes a sound input device 11, a voice detection device 12, a position information detection device 13, a voice recognition processing device 20, a navigation ECU 30, and an output device 31. The sound input device 11 is a so-called microphone, and inputs a voice spoken by a vehicle occupant. The sound detection device 12 is a device that detects sound input to the sound input device 11 as an electrical signal (hereinafter referred to as a sound signal).

  The position information detection device 13 is supplied with a position signal from the position signal sensor 50, and sequentially detects the current position of the vehicle based on the supplied position signal. Then, position information indicating the detected current position is output to the speech recognition processing device 20. For the position signal sensor 50, for example, a gyroscope, a geomagnetic sensor, a GPS receiver, a distance sensor, or the like can be used one or more kinds of various devices used for detecting the current position in a car navigation device.

  The voice recognition processing device 20 includes a computer and a storage device including a CPU, a ROM, a RAM, and the like, and the CPU executes a program stored in advance in the ROM while using a temporary storage function of the RAM. It functions as a recognition processing unit 21, a determination unit 22, and an output control unit 23. The recognition processing unit 21 converts the speech signal into character data by analyzing the speech signal detected by the speech detection device 12 using the speech recognition dictionary stored in the storage device.

  The determination unit 22 includes position information output from the position information detection device 13, a steering angle signal output from a steering angle sensor 51 that detects a steering angle, a travel-related signal other than the steering angle signal, and an external situation acquisition device 52. From the status information output from, it is determined whether or not the current situation requires attention to the driving operation.

  The above-mentioned driving-related signals include a signal indicating ON / OFF of the accessory switch, a signal indicating ON / OFF of the ignition switch, a signal indicating the operation of the antilock brake system, a signal indicating the tilt angle of the host vehicle, and the operation of the front wiper There are signals indicating the situation and vehicle speed signals. The external situation acquisition device 52 is a device that acquires the situation of the road on which the vehicle is traveling. For example, an in-vehicle camera that images the vehicle periphery such as the front of the vehicle, and weather at the current position of the vehicle from an external facility by wireless communication. There are communication devices that acquire information.

  Here, the determination in the determination unit 22 using the various signals will be specifically described. First, an example will be described in which it is determined that the driving operation requires attention based on the travel-related signal. For example, when it can be determined that the steering wheel is being operated based on the steering angle signal output from the steering angle sensor 51, it is determined that the driving operation requires attention. However, even if it can be determined that the steering wheel is being operated, when the vehicle speed can be determined to be extremely slow (for example, 10 km / h or less) based on the vehicle speed signal, or when the accessory switch or ignition switch is off (in this case, If the vehicle speed is considered to be zero), it may be determined that the situation does not require attention to the driving operation.

  The steering wheel may be operated only when the steering wheel is turned (when the change speed of the steering rudder angle is equal to or higher than a predetermined value), or when the steering wheel is held (the steering steering angle is at an angle). The above may also be included during the steering operation.

  In addition, when it is determined that the driving operation needs attention based on the driving-related signal, when it can be determined that the anti-lock brake system is operating, when the inclination angle of the host vehicle is greater than or equal to a predetermined angle, There are times when the operating speed of the front wiper is at its maximum speed.

  Next, an example will be described in which it is determined that the driving operation requires attention based on the situation information of the road on which the vehicle is traveling. The position information output from the position information detection device 13 is also one of the road status information, and based on this position information, when it can be determined that the current position is at or near a branch point such as an intersection, It is determined that the situation requires attention for driving operation.

  As another case where it is determined that the driving operation requires attention based on the situation information, the vehicle front image captured by the in-vehicle camera is analyzed to determine that the obstacle is within a predetermined distance from the own vehicle. When it is possible, there is a case where the weather at the current position acquired by the communication device is a storm or a storm.

  The output control unit 23 outputs the recognition result in the recognition processing unit 21, that is, the character data to the navigation ECU 30 and the output device 31. The processing of the output control unit 23 will be described in detail later with reference to FIG.

  The navigation ECU 30 is a computer including a CPU, a ROM, a RAM, etc., and a map scale change process, a menu display selection process, a destination setting process, a route search execution process, a route guidance process, a display screen change process, a volume adjustment process, etc. The various processes are executed. In these processes, information necessary for the passenger is output from the output device 31 as appropriate. In order for the vehicle occupant to cause the navigation ECU 30 to execute a desired process, a predetermined voice command is input to the sound input device 11 or a switch (not shown) is operated.

  The output device 31 is an interface that transmits various kinds of information from the navigation device 10 to the occupant, and specifically, a display and a speaker.

  FIG. 2 is a flowchart illustrating the processing of the output control unit 23 and the determination unit 22 of the speech recognition processing device 20. Next, processing of the output control unit 23 and the determination unit 22 will be described according to this flowchart. In FIG. 2, steps S10 to S30 and S60 to S70 are processes of the output control unit 23, and steps S40 and S50 are processes of the determination unit 22. The processing shown in FIG. 2 is repeatedly executed at a predetermined cycle.

  First, in step S <b> 10, it is determined whether a recognition result is output from the recognition processing unit 21. If this determination is negative, the routine shown in FIG. 2 is terminated. On the other hand, when the determination is affirmative, the process proceeds to step S20, and it is determined whether or not the recognition result output from the recognition processing unit 21, that is, the character data is a preset priority command. This priority command is a command that the driver is expected to desire to execute the function quickly regardless of the driving situation. For example, a command that instructs to display a map around the current position on the display corresponds to the priority command. The above command is a priority command because the driver is instructed to display a map around the current position on the display, and is difficult to understand if he / she is lost or if only voice route guidance is used. This is because there can be a situation.

  If step S20 is affirmative, the process proceeds to step S30, and the recognition result, that is, character data is output to the navigation ECU 30 and the output device 31, respectively. As a result, the recognized character is output as a sound from the speaker which is the output device 31, or the recognized character is displayed on the display. When the character data is supplied, the navigation ECU 30 executes control determined based on the character data, and appropriately outputs necessary information from the output device 31 to the passenger.

  On the other hand, when step S20 is negative determination, it progresses to step S40 and it is determined from the driving | running | working relevant signal which shows the driving | running | working state of a vehicle whether it is the situation which requires attention for driving operation. For example, if it can be determined that the vehicle is traveling at a predetermined vehicle speed or more and the steering wheel is being operated, an affirmative determination is made in step S40. In addition, since the specific content of determination of step S40 has been mentioned above, description of another specific example is abbreviate | omitted. If the determination in step S40 is negative, the process proceeds to step S50. If the determination is affirmative, the process proceeds to step S60.

  In step S50, it is determined from the status information output from the external status acquisition device 52 whether or not the driving operation requires attention. Since the specific content of the determination in step S50 has also been described above, description thereof is omitted here. If the determination in step S50 is negative, it is determined that neither the traveling state of the vehicle nor the state of the road is a situation that requires attention to the driving operation. Therefore, the process proceeds to step S30 described above, and the recognition result is immediately displayed. Output. On the other hand, if step S50 is affirmative, the process proceeds to step S60.

  Step S60 is executed when it is determined in either step S40 or step S50 that the driving operation requires attention. In this step S60, it is determined whether or not the elapsed time after the recognition result is output from the recognition processing unit 21 has exceeded a preset allowable standby time.

  If the determination in step S60 is affirmative, it is a case where a considerable amount of time has elapsed since the occupant spoke, and it is not preferable not to output the recognition result indefinitely even in a situation that requires attention to driving operation. The recognition result proceeds to step S30, and the recognition result is output.

  On the other hand, when step S60 is negative determination, it progresses to step S70 and waits for the fixed time set to the time shorter than the said allowable standby time. Then, it returns to above-mentioned step S40, and it is judged again whether the condition which requires attention for driving | operation operation was eliminated.

  As described above, according to the present embodiment described above, the determination unit 22 determines whether or not it is a situation that requires attention from the vehicle running state, and if it is determined that the situation requires attention, the recognition result is displayed. The output time is delayed.

  More specifically, the determination unit 22 determines whether or not the driving operation requires attention from a traveling-related signal indicating the traveling state of the vehicle (S40 in FIG. 2), and outputs from the external condition acquisition device 52. Whether or not the driving operation requires attention is also determined from the situation information (S50 in FIG. 2). When it is determined that the driving operation requires attention, the recognition result is output after it is determined that the driving operation does not require attention. Therefore, it is possible to prevent the attention to driving operation from becoming insufficient by outputting the recognition result. In addition, since the time when the recognition result is output is a time when the driving operation has a relatively large margin, even when the recognition result is erroneous recognition, the canceling operation can be performed particularly easily.

  Furthermore, if the recognition result is a priority command, the recognition result is output immediately, regardless of whether or not the driving operation requires attention. Can be prevented.

  In addition, when the elapsed time after the voice is recognized exceeds the allowable standby time, the recognition result is output even if it is determined that the driving operation requires attention. Therefore, the inconvenience that the recognition result cannot be output for a long period can be avoided.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the above-mentioned embodiment, The following embodiment is also contained in the technical scope of this invention, and also the summary other than the following is also included. Various modifications can be made without departing from the scope.

  For example, in the above-described embodiment, the devices from which the recognition result is output from the output control unit 23 are the navigation ECU 30 and the output device 31, but the output result may be output to an audio device or an air conditioner.

  In the above-described embodiment, the recognition result is output when it is determined that the allowable standby time has elapsed even if it is determined that the driving operation requires attention. In order to enable operation, the permissible standby time may not be provided, and the recognition result may not be output until it is determined that the driving operation is not in a situation.

It is a block diagram which shows the structure of the navigation apparatus 10 provided with the function as a speech recognition apparatus of this invention. It is a figure which shows the process of the output control part 23 and the determination part 22 of the speech recognition processing apparatus 20 of FIG. 1 as a flowchart.

Explanation of symbols

10: Navigation device, 11: Sound input device, 12: Voice detection device, 13: Position information detection device, 20: Voice recognition processing device, 21: Recognition processing unit, 22: Determination unit, 23: Output control unit, 30: Navigation ECU 31: Output device 50: Position signal sensor 51: Rudder angle sensor 52: External situation acquisition device

Claims (7)

A speech processing unit that receives speech uttered by a user and executes recognition processing, and a recognition processing unit that extracts the input speech as character data; and an output control unit that outputs a recognition result in the recognition processing unit to a predetermined device. A voice recognition device comprising:
It further comprises a determination means for sequentially determining whether or not the driving operation requires attention,
When the output control means does not determine that the timing for outputting the recognition result is a situation that requires attention for driving operation based on the determination that the determination means determines that the situation requires attention for driving operation. A speech recognition apparatus characterized by being slower than the above. In claim 1,
The output control means outputs the recognition result after the determination means determines that the situation requiring attention for driving operation is not determined by the determination means as the situation requiring attention for driving operation. A speech recognition apparatus characterized by: In claim 2,
The output control means, when the elapsed time from the recognition of speech by the recognition processing means exceeds a preset allowable waiting time, the recognition result regardless of the determination result in the determination means. A speech recognition apparatus characterized by outputting. The speech recognition apparatus according to any one of claims 1 to 3,
When the output control unit determines that the character data extracted by the recognition processing unit is set in advance as the content to be output quickly, the output control unit performs the driving operation regardless of the determination result by the determination unit. A speech recognition apparatus that outputs the recognition result at the same time as when it is not determined that the situation requires attention. The speech recognition apparatus according to any one of claims 1 to 4,
The speech recognition apparatus according to claim 1, wherein the determination unit determines whether or not the driving operation requires attention based on a travel-related signal indicating a travel state of the vehicle. The speech recognition apparatus according to claim 5.
The determination means acquires a signal indicating a steering angle as the travel-related signal, and determines that it is a situation that requires attention for driving operation based on being able to determine that the steering operation is being performed from the signal. A featured voice recognition device. The speech recognition apparatus according to any one of claims 1 to 6,
Whether the determination means acquires situation information of a road on which the vehicle is traveling from an external situation acquisition device mounted on the vehicle, and whether or not the driving operation requires attention based on the obtained information A speech recognition apparatus characterized by determining whether or not.

Images