JPWO2020016967A1 - Voice recognition device, in-vehicle navigation device, automatic voice dialogue device, and voice recognition method - Google Patents

Abstract

The voice recognition device (100) performs voice recognition based on the voice signal acquisition unit (111) that acquires the voice signal from the voice input unit (13) and the voice signal acquired by the voice signal acquisition unit (111), and recognizes the voice. It was determined whether or not the voice signal acquired by the voice recognition unit (112) and the voice signal acquisition unit (111) that output the result contained an ultrasonic signal, and it was determined that the voice signal contained an ultrasonic signal. In this case, the voice recognition control unit (113) is provided to control the recognition result based on the voice signal so as not to be output from the voice recognition unit (112).

Description

The present invention relates to a voice recognition device, an in-vehicle navigation device, an automatic voice dialogue device, and a voice recognition method.

As the accuracy of the voice recognition technology increases, the voice recognition technology is applied to electronic devices and the like, and the electronic devices and the like are controlled based on the voice emitted by the operator.
For example, Patent Document 1 describes a voice collecting means for collecting voices emitted by an occupant, a voice transmitting means for transmitting the collected voices to the equipment outside the vehicle, and an object created in the equipment outside the vehicle based on the transmitted voice. An in-vehicle navigation device for vehicles that includes a destination information receiving means for receiving location information from equipment outside the vehicle and provides guidance based on the received destination information, and voice recognition for voice collected by the voice collecting means. Is provided with a destination extraction means for extracting a destination from the voice collected by the voice collecting means, and after the voice is collected by the voice collecting means, until the destination information is received by the destination information receiving means. Discloses an in-vehicle navigation device for vehicles that provides guidance based on a destination extracted by a destination extraction means.

Japanese Unexamined Patent Publication No. 2008-2566559

However, in voice recognition, not only the voice emitted by the operator but also the voice having a frequency outside the human audible range emitted from a device that generates ultrasonic waves such as a parametric speaker is input. It may be recognized.
Ultrasound has frequencies outside the human audible range and is usually inaudible to humans. Further, the ultrasonic wave can have high directivity. For this reason, when an ultrasonic wave is emitted toward a voice input unit of an electronic device or the like to which voice recognition technology is applied, even a person around the voice input unit inputs the voice to the voice input unit. There is a problem that electronic devices and the like are controlled by the input ultrasonic signal without noticing it.

The present invention is for solving the above-mentioned problems, and an object of the present invention is to provide a voice recognition device capable of suppressing the recognition result output of voice recognition by ultrasonic waves.

The voice recognition device according to the present invention includes a voice signal acquisition unit that acquires a voice signal from a voice input unit, and a voice recognition unit that performs voice recognition based on the voice signal acquired by the voice signal acquisition unit and outputs a recognition result. , It is determined whether or not the audio signal acquired by the audio signal acquisition unit contains an ultrasonic signal, and when it is determined that the audio signal includes an ultrasonic signal, the recognition result based on the audio signal is obtained by the audio recognition unit. It is equipped with a voice recognition control unit that controls not to output from.

According to the present invention, it is possible to suppress the recognition result output of voice recognition by ultrasonic waves.

FIG. 1 is a block diagram showing a main part of an in-vehicle navigation device to which the voice recognition device according to the first embodiment is applied. 2A and 2B are diagrams showing an example of the hardware configuration of the main part of the voice recognition device according to the first embodiment. FIG. 3 is a flowchart illustrating an example of processing of the voice recognition device according to the first embodiment. FIG. 4 is a block diagram showing a main part of an in-vehicle navigation device to which the voice recognition device according to the modified example of the first embodiment is applied. FIG. 5 is a block diagram showing a main part of an automatic voice dialogue device to which the voice recognition device according to the second embodiment is applied.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

Embodiment 1.
The voice recognition device 100 according to the first embodiment will be described below, as an example, as being applied to the vehicle-mounted navigation device 10.
FIG. 1 is a block diagram showing a main part of an in-vehicle navigation device 10 to which the voice recognition device 100 according to the first embodiment is applied.

The vehicle 1 includes an in-vehicle navigation device 10, a navigation signal receiver 11, a map database 12, a voice input unit 13, a display device 14, and a voice output device 15.

The navigation signal receiver 11 is a receiving device that receives a navigation signal such as a GPS signal from a navigation satellite.

The map database 12 is a storage device that stores map information in which information related to a road map is recorded.

The voice input unit 13 is, for example, a microphone that converts the acquired sound wave into a voice signal and outputs the converted voice signal to the voice recognition device 100 described later.

The display device 14 is, for example, a display that displays guidance image information for providing route guidance to a destination output by the vehicle-mounted navigation device 10 described later.

The voice output device 15 is, for example, a speaker that outputs guidance voice for providing route guidance to a destination output by the vehicle-mounted navigation device 10 described later.

The in-vehicle navigation device 10 includes a voice recognition device 100, a navigation signal acquisition unit 101, a map information acquisition unit 102, a navigation control unit 103, a display output unit 104, and a voice output unit 105.

The navigation signal acquisition unit 101 acquires the navigation signal received by the navigation signal receiver 11.

The map information acquisition unit 102 acquires map information from the map database 12. The map database 12 is not always installed in the own vehicle as long as the map information acquisition unit 102 can acquire the map information. For example, the map information acquisition unit 102 may acquire map information from the map database 12 existing on the public network via a public network such as the Internet or a public line.

The navigation control unit 103 determines a point on the road on which the own vehicle travels, that is, a traveling position of the own vehicle, based on the navigation signal acquired by the navigation signal acquisition unit 101 and the map information acquired by the map information acquisition unit 102. Identify. The navigation control unit 103 generates travel position information indicating the specified travel position.
The navigation control unit 103 sets a destination based on the recognition result of voice recognition by the voice recognition device 100, which will be described later, and determines a travel route from the travel position of the own vehicle to the destination point. The navigation control unit 103 generates route guidance information based on the determined travel route.

The display output unit 104 generates guidance image information for performing route guidance based on the map information acquired via the navigation control unit 103 and the traveling position information and the route guidance information generated by the navigation control unit 103. , The guide image information is output to the display device 14.

The voice output unit 105 generates guidance voice information for performing route guidance based on the route guidance information generated by the navigation control unit 103, and outputs the guidance voice information to the voice output device 15.

That is, the in-vehicle navigation device 10 determines the traveling route to the set destination based on the navigation signal acquired from the navigation signal receiver 11 and the map information acquired from the map database 12, and provides route guidance. The information to be performed is output to the display device 14 and the audio output device 15.

The voice recognition device 100 includes a voice signal acquisition unit 111, a voice recognition unit 112, a voice recognition control unit 113, and a notification output unit 114.

The voice signal acquisition unit 111 acquires a voice signal from the voice input unit 13.
The voice signal acquisition unit 111 outputs the acquired voice signal to the voice recognition unit 112 and the voice recognition control unit 113.
The voice signal acquisition unit 111 adds a time stamp when the voice signal acquisition unit 111 acquires the voice signal, and outputs the voice signal to which the time stamp is added to the voice recognition unit 112 and the voice recognition control unit 113 as voice information. You may.

The voice recognition unit 112 performs voice recognition based on the voice signal acquired by the voice signal acquisition unit 111, and outputs the recognition result.
For example, the voice recognition unit 112 outputs the recognition result to the navigation control unit 103, and the navigation control unit 103 sets the destination based on the recognition result acquired from the voice recognition unit 112. Since the voice recognition process performed by the voice recognition unit 112 based on the voice signal can be performed by applying a well-known voice recognition technique, detailed description thereof will be omitted.

The voice recognition control unit 113 determines whether or not the voice signal acquired by the voice signal acquisition unit 111 includes an ultrasonic signal. When it is determined that the voice signal includes an ultrasonic signal, the voice recognition control unit 113 controls so that the voice recognition unit 112 does not output the recognition result based on the voice signal to the navigation control unit 103.
Specifically, in the process of determining whether or not the audio signal included in the audio signal by the speech recognition control unit 113 is performed, for example, the speech signal is spectrally analyzed by discrete Fourier transform, and a signal having a frequency higher than a predetermined frequency is used. Judge by the presence or absence. More specifically, for example, when the voice recognition control unit 113 determines that the voice signal includes an ultrasonic signal, the voice recognition control unit 113 controls the voice recognition unit 112 so as not to perform voice recognition, based on the voice signal. The recognition result is controlled so as not to be output from the voice recognition unit 112 to the navigation control unit 103. The predetermined frequency is not limited to 20,000 hertz, and may be a frequency lower than 20,000 hertz, for example, 10,000 hertz, as long as it is near the upper limit of the frequency that humans can hear.

Further, the process of determining whether or not the audio signal performed by the audio recognition control unit 113 includes an ultrasonic signal may be determined by whether or not a signal having a frequency higher than a predetermined frequency has a predetermined amplitude or more. The predetermined amplitude is, for example, the lower limit value of the amplitude required when the voice recognition unit 112 performs the voice recognition process.
Further, when the voice recognition control unit 113 determines that the voice signal includes an ultrasonic signal, for example, the voice recognition control unit 113 obtains a recognition result based on the voice signal during the period in which the voice signal is determined to contain the ultrasonic signal. By controlling so that the voice recognition unit 112 does not output to the navigation control unit 103, it may be controlled so that the recognition result based on the voice signal is not output from the voice recognition unit 112 to the navigation control unit 103. More specifically, for example, referring to the time stamp added by the voice signal acquisition unit 111, the voice recognition control unit 113 determines the start and end of the period in which the voice signal includes the ultrasonic signal. The indicated information is output to the voice recognition unit 112. More specifically, when the voice recognition control unit 113 determines that the voice signal includes an ultrasonic signal, the time when the ultrasonic signal is included in the voice signal, that is, the voice signal includes the ultrasonic signal. Information indicating the start of the period determined to be performed is immediately output to the voice recognition unit 112. After that, when the voice recognition control unit 113 determines that the voice signal does not include the ultrasonic signal, the time when the ultrasonic signal in the voice signal is no longer included, that is, the voice signal contains the ultrasonic signal. Information indicating the end of the determination period is output to the voice recognition unit 112. The voice recognition unit 112 navigates the recognition result of voice recognition during the period based on the information indicating the start and end of the period in which the voice signal output by the voice recognition control unit 113 is determined to include the ultrasonic signal. It is discarded without being output to the control unit 103.

The notification output unit 114 indicates that when the voice recognition control unit 113 controls the voice recognition unit 112 so that the navigation control unit 103 does not output the recognition result based on the voice signal, the notification output unit 114 controls not to output the recognition result. Generate information and output the generated notification information.
More specifically, for example, the notification output unit 114 controls the voice recognition control unit 112 so that the voice recognition control unit 113 does not output the recognition result based on the voice signal to the navigation control unit 103. Information to the effect that the recognition result is not output is acquired from 113. The notification output unit 114 generates notification information indicating that the recognition result is not output, based on the information that the recognition result acquired from the voice recognition control unit 113 is controlled so as not to be output, and is generated, for example. The notification information is output to the navigation control unit 103. The navigation control unit 103 causes the notification information output by the notification output unit 114 to be output from the display device 14 or the voice output device 15 via the display output unit 104 or the voice output unit 105. The navigation control unit 103 may output the notification information from both the display device 14 and the voice output device 15. The devices that output the notification information are not limited to the display device 14 and the voice output device 15 as long as the notification information output by the notification output unit 114 can be notified to the uttering operator or the like. For example, the navigation control unit 103 may turn on a lamp (not shown) such as a light emitting diode based on the notification information output by the notification output unit 114.

The notification output unit 114 is not an essential configuration in the voice recognition device 100, and can be added or deleted from the voice recognition device 100 as appropriate.
That is, the main part of the voice recognition device 100 may be composed of a voice signal acquisition unit 111, a voice recognition unit 112, and a voice recognition control unit 113.

2A and 2B are diagrams showing an example of the hardware configuration of the main part of the voice recognition device 100 according to the first embodiment.
The hardware configuration of the main part of the voice recognition device 100 according to the first embodiment will be described with reference to FIGS. 2A and 2B.

As shown in FIG. 2A, the voice recognition device 100 is composed of a computer, which has a processor 201 and a memory 202. The memory 202 stores a program for causing the computer to function as a voice signal acquisition unit 111, a voice recognition unit 112, a voice recognition control unit 113, and a notification output unit 114. When the processor 201 reads and executes the program stored in the memory 202, the functions of the voice signal acquisition unit 111, the voice recognition unit 112, the voice recognition control unit 113, and the notification output unit 114 are realized.

Further, as shown in FIG. 2B, the voice recognition device 100 may be configured by the processing circuit 203. In this case, the functions of the voice signal acquisition unit 111, the voice recognition unit 112, the voice recognition control unit 113, and the notification output unit 114 may be realized by the processing circuit 203.

Further, the voice recognition device 100 may be composed of a processor 201, a memory 202, and a processing circuit 203 (not shown). In this case, some of the functions of the voice signal acquisition unit 111, the voice recognition unit 112, the voice recognition control unit 113, and the notification output unit 114 are realized by the processor 201 and the memory 202, and the remaining functions are processed. It may be realized by the circuit 203.

The processor 201 uses, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a microprocessor, a microcontroller, or a DSP (Digital Signal Processor).

The memory 202 uses, for example, a semiconductor memory or a magnetic disk. More specifically, the memory 202 includes a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), and an EEPROM (Electrically Memory). State Drive) or HDD (Hard Disk Drive) or the like is used.

The processing circuit 203 includes, for example, an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), an FPGA (Field-Programmable Gate Array), a System-System (System) System, a System LSI. Is used.

The operation of the voice recognition device 100 according to the first embodiment will be described with reference to FIG.
FIG. 3 is a flowchart illustrating an example of processing of the voice recognition device 100 according to the first embodiment.
The voice recognition device 100 repeatedly executes the process shown in the flowchart shown in FIG.

First, in step ST301, the voice signal acquisition unit 111 acquires a voice signal from the voice input unit 13.
The voice signal acquisition unit 111 sequentially performs the processing of step ST301 by background processing, and the voice recognition device 100 sequentially performs the processing after step ST302 on the voice signal acquired by the voice signal acquisition unit 111. You can do it.

Next, the voice recognition control unit 113 determines whether or not the voice signal acquired by the voice signal acquisition unit 111 includes an ultrasonic signal (step ST302).

When it is determined in step ST302 that the voice signal does not include the ultrasonic signal (step ST302: NO), in step ST303, the voice recognition unit 112 performs voice recognition based on the voice signal, and the recognition result. Is output.
After the process of step ST303, the voice recognition device 100 ends the process shown in the flowchart shown in FIG. After completing the process shown in the flowchart, the voice recognition device 100 returns to step ST301 and repeatedly executes the process shown in the flowchart.

When it is determined in step ST302 that the voice signal includes an ultrasonic signal (step ST302: YES), in step ST304, the voice recognition control unit 113 outputs the recognition result based on the voice signal to the voice recognition unit 112. Is controlled so as not to be output to the navigation control unit 103.

After step ST304, in step ST305, the notification output unit 114 generates notification information indicating that the recognition result is not output, and outputs the generated notification information.
After the process of step ST304, the voice recognition device 100 ends the process shown in the flowchart shown in FIG. After completing the process shown in the flowchart, the voice recognition device 100 returns to step ST301 and repeatedly executes the process shown in the flowchart.

As described above, the voice recognition device 100 performs voice recognition based on the voice signal acquisition unit 111 that acquires the voice signal from the voice input unit 13 and the voice signal acquired by the voice signal acquisition unit 111, and outputs the recognition result. It is determined whether or not the voice signal acquired by the voice recognition unit 112 and the voice signal acquisition unit 111 includes an ultrasonic signal, and when it is determined that the voice signal includes an ultrasonic signal, it is based on the voice signal. A voice recognition control unit 113 for controlling the recognition result so as not to be output from the voice recognition unit 112 is provided.
With this configuration, the voice recognition device 100 can suppress the recognition result output of voice recognition by ultrasonic waves.

Further, when the voice recognition control unit 113 controls the voice recognition unit 112 so as not to output the recognition result based on the voice signal, the notification information indicating that the recognition result is not output is generated and the generated notification information is generated. By providing the notification output unit 114 that outputs the above, the voice recognition device 100 can notify the uttering operator or the like that the recognition result is not output because the voice signal includes the ultrasonic signal.

The voice recognition device 100a according to the modified example of the first embodiment will be described with reference to FIG.
FIG. 4 is a block diagram showing a main part of the vehicle-mounted navigation device 10 to which the voice recognition device 100a according to the modified example of the first embodiment is applied.
In FIG. 4, the same reference numerals are given to the same configurations as those shown in FIG. 1, and the description thereof will be omitted.

The voice recognition device 100 according to the first embodiment shown in FIG. 1 differs from the voice recognition device 100a according to the modified example of the first embodiment in the following points.

The voice recognition unit 112 of the voice recognition device 100 according to the first embodiment directly acquires the voice signal acquired by the voice signal acquisition unit 111 from the voice signal acquisition unit 111, whereas the modified example of the first embodiment The voice recognition unit 112a of the voice recognition device 100a according to the above means acquires the voice signal acquired by the voice signal acquisition unit 111a via the voice recognition control unit 113a.
Further, when the voice recognition control unit 113 of the voice recognition device 100 according to the first embodiment determines that the voice signal includes an ultrasonic signal, the voice recognition unit 112 is prevented from performing voice recognition or the voice signal. By controlling so that the voice recognition unit 112 does not output the recognition result based on the voice signal during the period in which it is determined that the ultrasonic signal is included in the navigation control unit 103, the recognition result based on the voice signal is obtained. Was controlled so as not to be output from the voice recognition unit 112a to the navigation control unit 103. On the other hand, when the voice recognition control unit 113a of the voice recognition device 100a according to the modified example of the first embodiment determines that the voice signal includes an ultrasonic signal, the voice recognition control unit 112a outputs the voice signal to the voice recognition unit 112a. By controlling so that the voice recognition unit 112a cannot acquire the voice signal for voice recognition, the recognition result based on the voice signal is transmitted from the voice recognition unit 112a to the navigation control unit 103. It controls not to output.

Since the functions in each configuration of the voice recognition device 100a according to the modified example of the first embodiment are the same as the functions in each configuration of the voice recognition device 100 according to the first embodiment except for the above-mentioned functions, the description thereof is omitted. To do.
Further, since the hardware configuration of the voice recognition device 100a according to the modified example of the first embodiment is the same as the hardware configuration of the voice recognition device 100 according to the first embodiment, the description thereof will be omitted. That is, each function of the voice signal acquisition unit 111a, the voice recognition unit 112a, the voice recognition control unit 113a, and the notification output unit 114 may be realized by the processor 201 and the memory 202, or the processing circuit 203. It may be realized by.

Further, since the processing flow of the voice recognition device 100a according to the modified example of the first embodiment is the same as the processing flow of the voice recognition device 100 according to the first embodiment, the description thereof will be omitted. That is, the processes in the voice signal acquisition unit 111, the voice recognition unit 112, the voice recognition control unit 113, and the notification output unit 114 in the flowchart shown in FIG. 3 are the voice signal acquisition unit 111a, the voice recognition unit 112a, and the voice recognition control, respectively. It is processed by the unit 113a and the notification output unit 114.

With this configuration, the voice recognition device 100a according to the modified example of the first embodiment can suppress the recognition result output of voice recognition by ultrasonic waves.

In the first embodiment and the modified example of the first embodiment, the vehicle-mounted navigation device 10 shows an example in which the destination is set based on the recognition results acquired from the voice recognition devices 100 and 100a. The operation of the navigation device 10 based on the recognition results acquired from the voice recognition devices 100 and 100a is not limited to setting the destination. For example, the vehicle-mounted navigation device 10 may reset the route and set the enlargement or reduction display of the guidance image information based on the recognition result acquired from the voice recognition devices 100 and 100a. Further, for example, when the in-vehicle navigation device 10 has the function of the in-vehicle audio device, the in-vehicle navigation device 10 reproduces music information or the like based on the recognition result acquired from the voice recognition devices 100, 100a. You may control to do so.

Embodiment 2.
The voice recognition device 100 according to the second embodiment will be described below, assuming that it is applied to the automatic voice dialogue device 50 as an example.

FIG. 5 is a block diagram showing a main part of the automatic voice dialogue device 50 to which the voice recognition device 100 according to the second embodiment is applied.
In FIG. 5, the same reference numerals are given to the same configurations as those shown in FIG. 1, and the description thereof will be omitted.

The automatic voice dialogue device 50 will be described later.

The example sentence database 16 is a storage device in which example sentence information for searching an example sentence corresponding to the recognition result is stored based on the recognition result acquired from the voice recognition device 100 by the automatic voice dialogue device 50 described later.

The voice input unit 17 is, for example, a microphone that converts the acquired sound wave into a voice signal and outputs the converted voice signal to the voice recognition device 100 described later.

The voice output device 18 is, for example, a speaker that outputs a voice signal output by the automatic voice dialogue device 50 described later.

The display device 19 is, for example, a display that displays the image information output by the automatic voice dialogue device 50 described later.

An automatic voice dialogue system is composed of an automatic voice dialogue device 50, an example sentence database 16, a voice input unit 17, a voice output device 18, and a display device 19.

The automatic voice dialogue device 50 includes a voice recognition device 100, a matching unit 152, an answer creation unit 153, a voice generation unit 154, an answer voice output unit 155, and a display output unit 156.

The matching unit 152 searches the example sentence database 16 in which the example sentence information is stored for an example sentence corresponding to the recognition result based on the recognition result acquired from the voice recognition device 100 described later.
More specifically, for example, when the recognition result acquired from the voice recognition device 100 is the character string "What is it now?", The character string "What time is it now?" Corresponding to the character string is used in the example sentence database. Search from 16.

The answer creation unit 153 generates a character string of the answer corresponding to the recognition result based on the result searched by the matching unit 152.
More specifically, for example, when the result of the search by the matching unit 152 is the character string "what time is it now?", The answer corresponding to the character string is, for example, the character "1:15 pm". Generate a column.

The voice generation unit 154 converts the character string generated by the answer creation unit 153 into a voice signal and outputs it to the answer voice output unit 155, which will be described later.

The answer voice output unit 155 outputs the voice signal output by the voice generation unit 154 to a voice output device 18 such as a speaker.

The display output unit 156 generates, for example, image information indicating the state of the automatic voice dialogue device 50 based on the result of the matching unit 152 searching the example sentence database 16 for the character string corresponding to the recognition result, and the generated image. The information is output to the display device 19. More specifically, for example, when the matching unit 152 searches the example sentence database 16 for a character string corresponding to the recognition result and the example sentence information corresponding to the character string does not exist in the example sentence database 16, the display output unit 156 , Image information indicating that the voice recognition has failed is generated, and the generated image information is output to the display device 19.

Since each configuration of the voice recognition device 100 and the voice recognition device 100 is the same as that described in the first embodiment, the description thereof will be omitted.

The voice signal acquisition unit 111 in the voice recognition device 100 according to the second embodiment acquires a voice signal from the voice input unit 17.

Further, the notification output unit 114 in the voice recognition device 100 according to the second embodiment is controlled so as not to output the recognition result based on the information that the recognition result acquired from the voice recognition control unit 113 is controlled not to be output. Notification information indicating that is generated, and the generated notification information is output to, for example, the matching unit 152. The matching unit 152 outputs the notification information output by the notification output unit 114 from the display device 19 or the voice output device 18 via the display output unit 156 or the response voice output unit 155. The matching unit 152 may output the notification information from both the display device 19 and the voice output device 18. If it is possible to notify the operator or the like who has uttered that the recognition result is not output because the voice signal includes an ultrasonic signal, the devices that output the notification information are limited to the display device 19 and the voice output device 18. It's not a thing. For example, the matching unit 152 may turn on a lamp (not shown) such as a light emitting diode based on the notification information output by the notification output unit 114.

Note that the notification output unit 114 according to the second embodiment is not an essential configuration in the voice recognition device 100 as in the first embodiment, and can be added or deleted from the voice recognition device 100 as appropriate.
That is, the main part of the voice recognition device 100 according to the second embodiment may be composed of a voice signal acquisition unit 111, a voice recognition unit 112, and a voice recognition control unit 113.

Since the hardware configuration of the voice recognition device 100 according to the second embodiment is the same as the hardware configuration of the voice recognition device 100 according to the first embodiment, the description thereof will be omitted.

Since the processing flow of the voice recognition device 100 according to the second embodiment is the same as the processing flow of the voice recognition device 100 according to the first embodiment, the description thereof will be omitted.

As described above, the automatic voice dialogue device 50 is not limited to a simple dialogue such as answering a question time, for example, and purchases a product via the Internet, for example, based on the recognition result obtained from the voice recognition device 100. There are things that carry out commercial transactions such as. Since the conventional automatic voice dialogue device recognizes the voice even when the ultrasonic wave is received, for example, a user such as the owner of the automatic voice dialogue device by the ultrasonic wave emitted by a malicious third party. However, there was a problem that unintended commercial transactions were carried out.

However, when the automatic voice dialogue device 50 to which the voice recognition device 100 according to the second embodiment is applied determines that the voice signal includes an ultrasonic signal, it is controlled so as not to output a recognition result based on the voice signal. Therefore, it is possible to suppress unintended commercial transactions by the user.

The automatic voice dialogue device 50 according to the second embodiment may be the one to which the voice recognition device 100a described in the modified example of the first embodiment is applied.

In the embodiments described so far, the voice recognition devices 100 and 100a have shown an example in which the voice recognition units 112 and 112a are included in the voice recognition devices 100 and 100a, but the present invention is not limited to this. For example, the voice recognition devices 100 and 100a have a configuration (not shown) for connecting to a public network such as the Internet or a public line, and the voice recognition devices 100 and 100a have a voice recognition unit 112 existing on the public network. , 112a is transmitted to a voice recognition server (not shown) via the configuration, the voice recognition server outputs a recognition result based on the voice signal, and the voice recognition devices 100 and 100a have the configuration. The recognition result output by the voice recognition server may be acquired via.

Further, in the embodiments described so far, in the voice recognition devices 100 and 100a, the voice recognition units 112 and the voice recognition control unit 113 receive the voice signals acquired by the voice signal acquisition units 111 and 111a from the voice input units 13 and 17. An example of output to is shown, but this is not the case. For example, the voice signal acquisition units 111 and 111a output the voice signal acquired from the voice input units 13 and 17 to the voice recognition unit 112 and receive ultrasonic waves arranged in the vicinity of the voice input units 13 and 17. The ultrasonic signal acquired from the ultrasonic input unit (not shown) may be output to the voice recognition control unit 113. Here, the ultrasonic input unit is, for example, an ultrasonic microphone that receives ultrasonic waves.

It should be noted that, within the scope of the present invention, any combination of the embodiments can be freely combined, any component of the embodiment can be modified, or any component can be omitted in each embodiment. ..

The voice recognition device according to the present invention can be applied to a device in which a user performs an input operation by voice.

1 vehicle, 10 in-vehicle navigation device, 11 navigation signal receiver, 12 map database, 13,17 voice input unit, 14,19 display device, 15,18 voice output device, 16 example sentence database, 50 automatic voice dialogue device, 100 , 100a voice recognition device, 101 navigation signal acquisition unit, 102 map information acquisition unit, 103 navigation control unit, 104,156 display output unit, 105 voice output unit, 111,111a voice signal acquisition unit, 112,112a voice recognition unit, 113, 113a Voice recognition control unit, 114 Notification output unit, 152 Matching unit, 153 Answer creation unit, 154 Voice generation unit, 155 Answer voice output unit, 201 processor, 202 memory, 203 processing circuit.

Claims (7)

An audio signal acquisition unit that acquires an audio signal from the audio input unit,
A voice recognition unit that performs voice recognition based on the voice signal acquired by the voice signal acquisition unit and outputs the recognition result.
When it is determined whether or not the voice signal acquired by the voice signal acquisition unit includes an ultrasonic signal and it is determined that the voice signal includes the ultrasonic signal, the recognition result based on the voice signal A voice recognition control unit that controls not to output the sound from the voice recognition unit,
A voice recognition device characterized by being equipped with. The voice recognition device according to claim 1, wherein when the voice recognition control unit determines that the voice signal includes the ultrasonic signal, the voice recognition control unit controls the voice recognition unit so as not to perform the voice recognition. .. When the ultrasonic signal contained in the voice signal has a predetermined amplitude or more, the voice recognition control unit determines that the voice signal includes the ultrasonic signal, and the voice signal includes the ultrasonic signal. The voice recognition device according to claim 1, wherein the recognition result based on the voice signal during a period determined to be included is controlled so as not to be output from the voice recognition unit. When the voice recognition control unit controls the voice recognition unit so as not to output the recognition result based on the voice signal, the notification information indicating that the recognition result is not output is generated and generated. The voice recognition device according to claim 1, further comprising a notification output unit that outputs notification information. An in-vehicle navigation device comprising the voice recognition device according to any one of claims 1 to 4, and operating based on the recognition result output from the voice recognition device. An automatic voice dialogue device comprising the voice recognition device according to any one of claims 1 to 4, and operating based on the recognition result output from the voice recognition device. The audio signal acquisition unit acquires the audio signal from the audio input unit,
The voice recognition unit outputs the recognition result of performing voice recognition based on the voice signal acquired by the voice signal acquisition unit.
When the voice recognition control unit determines whether or not the voice signal acquired by the voice signal acquisition unit contains an ultrasonic signal, and determines that the voice signal includes the ultrasonic signal, the voice Control so that the recognition result based on the signal is not output from the voice recognition unit,
A voice recognition method characterized by.

Images