JP2016133378A - Car navigation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent execution of a command that is likely to identify personal information on an owner of a vehicle and to thereby prevent the personal information from being stolen in manipulating a car navigation device even if the owner has the vehicle stolen.SOLUTION: A car navigation device 20 according to the present invention includes: a first determination unit 121 that determines whether a command corresponding to input first voice data is associated with a function that can identify personal information on an owner; and a second determination unit 141 that collates the first voice data with second voice data registered in advance and that determines whether or not the two voice data match each other, and is configured so that if the command is associated with the function that can identify the personal information of the owner and if the first voice data does not match the second voice data as a result of collation by the second determination unit 141, the command is not executed.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a car navigation device capable of preventing personal information from being stolen.

  In recent years, theft of automobiles has increased, and voice recognition has been used for security functions for preventing theft. In Patent Document 1, voice recognition is used for the purpose of preventing malfunction of an alarm device provided in a vehicle, voice data registered in advance by an owner is stored, and the registered voice data and voice data input from a microphone are stored. A device is disclosed in which the alarm device is controlled by checking the above. In Patent Document 2, in order to improve security due to loss or theft of a key, a release command such as a door lock is registered in advance as voice data, and when voice data is input, the voice data matches the release command. A device is disclosed in which the door lock is released if the two are determined to match. However, in these cases, if not only the key but also the entire car is stolen, the necessity for voice data input is lost and the effectiveness of security is lacking.

JP 2009-25882 A JP 7-44189 A

  Car navigation devices often contain information that identifies the location of the home, the telephone number of the home, and other individual owners. If the car is stolen, the car navigation device is operated, and the information for identifying the individual owner can be easily viewed, it may fall into a situation where the car cannot be stolen alone.

  The present invention has been made in consideration of the above circumstances, and the purpose of the present invention is to use the personal information of the owner in the operation of the car navigation device even if the car equipped with the car navigation device is stolen. It is an object of the present invention to provide a car navigation device capable of preventing an identifiable command from being executed and thereby preventing personal information from being stolen.

  The present invention relates to an operation in which information for identifying an owner can be viewed by an operation of a car navigation device. If the speaker recognition function does not identify a registered speaker, the car navigation device By providing a guard that cannot execute this function, damage can be minimized.

  In the car navigation device according to the first aspect of the present invention, the first command for determining whether or not the command corresponding to the input first voice data relates to a function capable of specifying the owner's personal information. The determination unit (121) includes a second determination unit (141) that collates the first audio data with the second audio data registered in advance and determines whether or not they match. If the command is related to a function that can identify the personal information of the owner, and the first voice data and the second voice data do not match as a result of the collation in the second determination unit, It is characterized by not executing the command.

  According to this configuration, for example, even when a car theft operator operates the car navigation device, a command related to a function that can identify the owner's personal information is not executed, and thus the owner's personal information is leaked. Can be avoided.

  The car navigation device according to the invention described in claim 2 includes a first input unit (23, 26) capable of manually inputting a command, a second input unit (15) capable of inputting voice data, A first determination unit (121) for determining whether or not the command input from the first input unit relates to a function capable of specifying the personal information of the owner; A second determination unit (141) that collates the registered second audio data and determines whether or not they match. When the first determination unit determines that the command input from the first input unit is related to a function that can identify the personal information of the owner, the first audio data is transmitted to the second input unit. And a process of collating the first voice data with the second voice data in the second determination unit, and the first voice data as a result of the collation in the second determination unit If the second audio data does not match, the command is not executed.

  According to this configuration, for example, even when a vehicle theft operates a car navigation device, a command related to a function that can identify the owner's personal information is not executed, and thus the owner's personal information is leaked. Can be avoided.

It is a block diagram which shows an example of schematic structure of the car navigation apparatus which concerns on embodiment. It is a block diagram which shows an example of schematic structure of the speech recognition part which concerns on embodiment, a speech recognition control part, and a speaker recognition part. It is a flowchart which shows an example in the case of performing a series of operation including the speaker recognition process which concerns on embodiment by voice operation. It is a flowchart which shows an example in the case of performing a series of operation including the speaker recognition process which concerns on embodiment by manual operation. It is a flowchart which shows an example in the case of performing ON / OFF switching process of the speaker recognition process which concerns on embodiment. In an embodiment, it is a figure showing an example of operation in the case of performing a command “go home” by voice operation. In an embodiment, it is a figure showing an example of processing operation in the case of "setting home as a destination" by manual operation. In an embodiment, it is a figure showing an example of processing operation of ON / OFF switching processing of a speaker recognition function. It is a figure which shows the processing operation example of speaker recognition function application in embodiment.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of a car navigation device 20 having a speaker recognition function according to the embodiment. The car navigation device 20 is a so-called car navigation system that is mounted on, for example, a vehicle and used for route guidance to a destination. The car navigation device 20 includes a control circuit 28. The car navigation device 20 further includes a position detector 21, a data input device 22, an operation switch group 23, a communication device 24, an external memory 25, a display device 26, and a remote control sensor 27, which are connected to the control circuit 28. It is connected. The car navigation device 20 further includes a voice recognition unit 10, and the voice recognition unit 10 is connected to the control circuit 28. The car navigation device 20 further includes a speaker 14, a microphone 15, and a push-to-talk (PTT) switch 16 (hereinafter referred to as a PTT switch 16), which are included in the voice recognition unit 10. It is connected to the. The voice recognition unit 10 and the control circuit 28 are configured as, for example, an ordinary computer, and include a well-known CPU, ROM, RAM, I / O, a bus line for connecting these configurations, and the like. . Each component of the navigation device 20 does not necessarily need to be physically configured as a separate body, and may be configured as various software that operates by a single CPU, for example.

  The position detector 21 includes a gyroscope 21a, a distance sensor 21b, and a GPS receiver 21c. The gyroscope 21a is a measuring instrument that calculates the rotation direction and direction of an automobile by detecting the angle and angular velocity of an object, and may use any of a mechanical type, a fluid type, and an optical type. The distance sensor 21b is a measuring instrument that measures the moving distance of the automobile. The GPS receiver 21c detects the position of the vehicle based on radio waves from the satellite. Each of these sensors has an error of a different nature, so that a plurality of sensors are configured to be used while interpolating each other. Depending on the accuracy, a part of the above may be used, and a steering rotation sensor, a wheel sensor for each rolling wheel, and the like may be additionally used.

  The data input unit 22 is a dictionary used when performing recognition processing in the speech recognition unit 10 in addition to so-called map matching data for improving the accuracy of position detection, various data for car navigation including map data and landmark data. A device for inputting data. As a storage medium in which the data input device 22 stores these data, it is generally considered that a hard disk or a DVD is used because of the amount of data, but other media such as a CD-ROM may be used. When a DVD is used as the data storage medium, the data input device 22 is a DVD player.

  The display device 26 is a color display device. On the screen of the display device 26, the vehicle current position mark input from the position detector 21, the map data input from the map data input device 22, and further on the map. Additional data such as a guide route to be displayed and a mark of a set point can be displayed in an overlapping manner. Further, the display device 26 has a touch panel function, and can display a menu screen for displaying a plurality of options, a command input screen for displaying a plurality of options when an option is selected.

The communication device 24 is for communicating with a contact specified by the set contact communication information, and is constituted by a mobile communication device such as a mobile phone.
The car navigation apparatus 20 also has a so-called route guidance function. Input to the car navigation apparatus 20 is performed from the remote control sensor 27 via a remote control terminal 27a (hereinafter referred to as a remote controller 27a). It is also possible to input using the operation switch group 23. Input of various commands from the remote controller 27a and the operation switch group 23 is performed by manual operation. When the position of the destination is input, the optimum route from the current position to the destination can be automatically selected to form and display a guidance route. A technique such as the Dijkstra method is known as a method for automatically setting an optimum route. For example, a touch switch or a mechanical switch integrated with the display device 26 is used as the operation switch group 23 and is used for inputting various commands. Each function in the car navigation device 20 is realized by software processing in the control circuit 28.

  The voice recognition unit 10 is a device for inputting various commands to the car navigation device 20 by voice (voice data) emitted by a person. The speech recognition unit 10 includes a speech synthesis unit 11, a speech recognition unit 12, a speech recognition control unit 13 (hereinafter simply referred to as a control unit 13), and a speaker recognition unit 17. The speaker 14 is connected to the voice synthesizer 11. The microphone 15 is connected to the voice recognition unit 12. The PTT switch 16 is connected to the control unit 13.

  The voice recognition unit 12 performs word recognition processing on the input voice data (command) according to an instruction from the control unit 13, and outputs the recognition result to the control unit 13. Audio data is input from the microphone 15. The voice recognition unit 12 collates the voice data input from the microphone 15 using stored dictionary data (a voice recognition dictionary unit 107 described later). For example, the plurality of comparison target pattern candidates stored in the dictionary data are compared with the voice data, and the comparison target pattern having a high degree of coincidence is output to the control unit 13.

  The recognition of the word series in the input voice data is performed as follows, for example. First, the input voice data is acoustically analyzed by comparing with the acoustic model of each voice data, and an acoustic feature quantity (for example, cepstrum) is extracted. By this acoustic analysis, acoustic feature amount time-series data is obtained. Then, this time series data is divided into several sections by a well-known HMM (Hidden Markov Model), DP (Dynamic Programming) matching method, or neural network, and each section is stored as dictionary data. Find out which word in a word corresponds to. As described above, it is possible to recognize the input voice data as a word and thereby recognize an operation command based on the voice data.

  Based on the recognition result in the voice recognition unit 12, the control unit 13 determines whether or not the voice data is to be output to the speaker recognition unit 17. If the voice data input from the microphone 15 corresponds to an operation command that can specify personal information based on the recognition result in the voice recognition unit 12, it is determined that the voice data is output to the speaker recognition unit 17. The For example, this corresponds to the case where the voice data corresponds to an operation command such as “go home” or “call home”.

  On the other hand, when it does not correspond to the operation command that can specify the personal information, it is determined that the voice data is not output to the speaker recognition unit 17 (non-target). For example, this corresponds to the case where the voice data corresponds to an operation command such as “go to Tokyo Station” or “go to the closest parking lot for the time from here”. If it does not correspond to an operation command that can specify personal information, the function of the car navigation device 20 is executed as it is.

  If it is determined that the voice data is to be output to the speaker recognition unit 17, the speaker recognition unit 17 performs speaker recognition processing and outputs the recognition result to the control unit 13. In the speaker recognition process, it is determined whether or not the input voice data matches the voice data registered in advance. The voice data registered in advance is, for example, voice data of an owner or an owner's family (hereinafter referred to as “owner etc.”). That is, in the speaker recognition process, it is determined whether the input voice data matches the voice of the owner or the like and the speaker is specified.

  In the speaker recognition processing in the speaker recognition unit 17, if no speaker is specified, it is determined that the function is to be rejected (command function is not executed), and a response voice output instruction to that effect is given to the speech synthesizer 11. To finish the process. The voice synthesizer 11 synthesizes a voice based on a response voice output instruction from the voice dialogue controller 13 using a voice waveform stored in the waveform database. The synthesized voice is output to the speaker 14 and the voice is output from the speaker 14. For example, a voice saying “The operation is not available at this time” is output. Here, the function execution rejection means that the execution of the operation command input by the voice data is stopped and the function of the car navigation device 20 corresponding to the command is not executed. By rejecting the function execution of the car navigation device 20, it is possible to avoid leakage of personal information to a person other than the owner (theft).

Further, as one mode of rejecting the function execution, for example, the car navigation device 20 may be configured to perform the function in the same state as when the owner's personal information is not registered. For example, when the home is set as the destination with the command “return to home”, the operation is performed in a state where the home is not yet registered. In this case, it is not a response that does not execute the function, such as “Currently, that operation is not available”, but a response such as “Your home is not registered. . In this way, by making the same response as when the owner's personal information is not registered, it becomes the same movement as normal operation. I can't. Moreover, the same effect is acquired even if it returns the navigation apparatus 20 to a factory shipment state as one aspect of function execution rejection.
As a result, for example, when a car is stolen, the probability of a secondary damage such as the reverse of the feeling of the theft and further damage to the car is reduced. In addition, it is possible to hide the information regarding the position of the home, which is the original purpose.

  In addition, following such processing, when an operation for restarting the car navigation device 20 is performed, the start of the car navigation device 20 may be stopped (not started). By doing so, the car navigation device 20 cannot be activated by a thief, so that personal information can be more reliably prevented from leaking. In such a case, the car navigation device 20 cannot be restarted unless the identity verification is performed by inputting the password set by the owner.

  When a speaker is specified in the speaker recognition process in the speaker recognition unit 17, that is, when the input voice data matches the voice of the owner or the like, the operation of the owner of the car navigation device 20 is performed. Therefore, the normal operation is performed. In this case, a response voice output instruction is given to the voice synthesizer 11. Further, necessary processing is performed on the control circuit 28 that executes processing of the car navigation device 20 in accordance with the command based on the input voice data. For example, a destination or command necessary for the navigation processing is notified, and processing for instructing to set the destination or execute the command is performed. As a result of such processing, if this voice recognition unit 10 is used, the destination of navigation to the car navigation apparatus can be instructed by voice input even if the user does not manually operate the operation switch group 23 or the remote controller 27a. It becomes possible.

  When the control unit 13 determines that the command corresponding to the input voice data is not an operation command that can specify personal information, the control unit 13 determines that the voice data is not output to the speaker recognition unit 17 and In step 20, the normal operation described above is performed.

  In the voice command input operation according to the present embodiment, it is assumed that the user inputs voice data through the microphone 15 while pressing the PTT switch 16. The control unit 13 monitors the timing when the PTT switch 16 is pressed, the timing when it is returned, and the time during which the pressed state continues. When the PTT switch 16 is pressed, the execution of processing in the voice recognition unit 12 is instructed. On the other hand, when the PTT switch 16 is not pressed, the processing is not executed. Therefore, the voice data input via the microphone 15 while the PTT switch 16 is being pressed is output to the voice recognition unit 12. With such a configuration, the car navigation device 20 can execute various processes such as route setting, route guidance, facility search, and facility display when a user inputs a command.

  Next, the configuration of the voice recognition unit 12, the control unit 13, and the speaker recognition unit 17 will be described in more detail. As shown in FIG. 2, the speech recognition unit 12 includes a speech extraction unit 101, a speech recognition collation unit 103 (hereinafter referred to as collation unit 103), a speech recognition result output unit 105 (hereinafter referred to as output unit 105), a speech A recognition dictionary unit 107 (hereinafter referred to as the dictionary unit 107) is provided. The dictionary unit 107 includes, for example, a command correspondence dictionary 201, an address correspondence dictionary 203, a music correspondence dictionary 205, a telephone directory correspondence dictionary 207, and the like. These dictionaries store data that can be collated with voice data and associated with commands, addresses, and the like.

  The control unit 13 includes a speaker recognition target determination unit 121 (hereinafter referred to as determination unit 121), a function execution process determination unit 123 (hereinafter referred to as determination unit 123), and an audio output content determination unit 125 (hereinafter referred to as determination unit 125). Called). The speaker recognition unit 17 includes a speaker recognition verification unit 141 (hereinafter referred to as a verification unit 141) and registered speaker data 143.

  The user inputs sound for performing an operation desired by the user from the microphone 15. The voice extraction unit 101 of the voice recognition unit 12 converts the voice data input from the microphone 15 into, for example, a digital signal, and extracts voice data that can be collated with data stored in the dictionary. The collation unit 103 collates the voice data input from the voice extraction unit 101 with the data stored in the dictionary (201 to 207...), And what kind of words the input voice data is. Recognize For example, when the input voice data is “go home”, it is recognized as a word corresponding to a command that guides a route to the home in the command correspondence dictionary 201. The collation unit 103 outputs the recognition result to the output unit 105. The output unit 105 outputs the recognition result input from the collation unit 103 to the determination unit 121 of the control unit 13. Note that the audio data input from the microphone 15 is also output from the output unit 105 to the determination unit 121.

  Based on the recognition result input from the output unit 105, the determination unit 121 of the control unit 13 determines whether the input voice data is a target for outputting the voice data to the speaker recognition unit 17. . When the input voice data relates to a command that can specify personal information, it is determined that the voice data is to be output to the speaker recognition unit 17. In this case, the determination unit 121 outputs the voice data to the verification unit 141 of the speaker recognition unit 17. In the speaker recognition unit 17, the voice data input from the control unit 13 and the registered speaker data 143 are collated by the speaker collation unit 141, and a match or mismatch is determined.

  As a result of the collation in the collation unit 141, when a speaker is specified (match), that is, when it is determined that the voice data is input by the owner or the like, the collation unit 141 determines the collation result and determines the control unit 13 Output to the unit 123. The determination unit 123 determines a process to be executed in the car navigation device 20 based on the input collation result, and the determination result is output to the control circuit 28.

  If the speaker is not specified as a result of the verification in the verification unit 141 (mismatch), that is, if the voice data is not input by the owner or the like, the verification unit 141 displays the verification result in the control unit 13 Output to function execution rejection determination unit 127. The function execution rejection determination unit 127 determines that the function execution is rejected based on the verification result in the verification unit 141, and outputs the determination result to the determination unit 125. The determination unit 125 determines the content to be output as a sound based on the determination result, and outputs the content to the speech synthesis unit 11.

  In the determination unit 121 of the control unit 13, when the input voice data is not related to a command capable of specifying personal information, it is determined that the voice data is not output to the speaker recognition unit 17 (asymmetric). In this case, the determination result is output to the mechanism execution process determination unit 123 and the determination unit 125. The determination unit 125 determines the content to be output as a sound based on the determination result, and outputs the content to the speech synthesis unit 11. The determination unit 123 determines a process to be executed in the car navigation device 20 based on the input collation result, and the determination result is output to the control circuit 28.

  Next, processing for executing speaker recognition will be described with reference to FIGS. 3, 4, and 5. FIG. Moreover, FIG.1 and FIG.2 is referred suitably. FIG. 3 is a flowchart showing processing by voice operation according to the present embodiment. First, when voice data is input to the voice extraction unit 101 via the microphone 15 in step S10, the collation unit 103 uses the dictionary (201 to 207) of the dictionary unit 107 to perform recognition processing in step S20. Is executed. In step S <b> 30, the recognition result is output from the matching unit 103 to the output unit 105.

  Next, in S40, the determination unit 121 determines whether or not the speaker is a speaker recognition target. If it is determined that the speaker is to be recognized (“YES” in step S40), the speaker recognition process in the matching unit 141 is executed using the registered speaker data 143 in step S50, and the process proceeds to step S60. Thus, it is determined whether or not a speaker has been identified, that is, a match or a mismatch.

  If it is not determined as a speaker recognition target in step S40, that is, if it is determined as asymmetric (“NO” in step S40), a function execution process is performed in step S80, and the voice operation is terminated. To do.

  Next, in step S60, it is determined whether or not the speaker has been specified by the collation unit 141. If it is determined that the speaker has been specified (match) ("YES" in step S60), step S80 is performed. Then, the function execution process is performed by the determination unit 123, and the determination unit 125 determines to end the voice operation.

  If it is determined in step S60 that the speaker has not been identified by the verification unit 141 (no match) ("NO" in step S60), the determination unit 127 rejects the function execution in step S70. Is determined. In this case, the determination result is output to the determination unit 125, and the voice operation is terminated in the determination unit 125.

  If a message for suddenly terminating the voice operation is output here, the thief is informed that the guard by the speaker recognition function is put on. Therefore, the control may be performed with a motion that is not different from the normal operation. For example, when setting the home as the destination with the command “return to home”, the navigation function is not executed so that the position of the home is not specified, but the function is executed with the home not yet registered. Behave as expected. In this case, for example, the response is not “Your operation is not currently available” but “Your home is not registered. Please register your home before executing”. By doing so, it becomes impossible to specify that the guard by the speaker recognition function is put because it becomes the same movement as normal operation, and it is possible to prevent the information about the position of the home from being revealed. Become. In this way, a series of voice operation processing including speaker recognition processing is completed.

  FIG. 4 is a flowchart showing processing by manual operation. When the screen transition is made by manual operation in the first step S110, it is determined in step S120 whether or not it is a speaker recognition target. The manual operation is performed by, for example, a user input to the operation switch group 23 (manual input unit) or an operation button (manual input unit) displayed on the display device 26 having a touch panel function. If the input command is determined to be a speaker recognition target (“YES” in step S120), speaker recognition processing is executed in step S130. Next, in step S140, it is determined whether or not the speaker is specified by the collation unit 141 (match or mismatch). The process of determining whether or not it is a speaker recognition target in step S120 is performed here by the control circuit 28 (speaker recognition target determination unit). If it is not determined as a speaker recognition target in step S120 (“NO” in step S120), the determination unit 123 determines to execute the function in step S160, and the manual operation is terminated. The navigation function is executed by the control circuit 28.

  Next, when it is determined in S140 that the speaker has been identified (matched) ("YES" in step S140), a function execution process is performed in step S160, and the manual operation is terminated. The execution of the navigation function is performed by the control circuit 28. If it is determined in step S140 that the speaker has not been identified (mismatch) (“NO” in step S140), the determination unit 127 determines that the function execution is rejected in step S150. Done and the manual operation is terminated. In addition, the determination unit 127 outputs the determination result to the determination unit 125, and based on this, the determination unit 125 responds, for example, “Currently, the operation cannot be used”. In addition, instead of this response, the operation may be performed in such a way that the function is executed in a state where the home has not yet been registered, and the response “No home has been registered. Please execute after registering the home” may be used. . In this way, a series of manual operation processing including speaker recognition processing is completed. In the process in step S140, the speaker recognition process is performed by voice input. Alternatively, the speaker recognition process may be performed by, for example, inputting a password and checking the password. In this way, for example, in the case of an operation by the owner himself / herself, the normal operation can be restored by inputting the password, so that convenience is improved.

  FIG. 5 is a flowchart showing the operation of the speaker recognition ON / OFF switching process. When the screen transitions to the setting screen for the speaker recognition ON / OFF switching process in the first step S210, the determination unit 121 executes the speaker recognition process in step S220. Based on the processed content, in step S230, it is determined whether or not the speaker is specified by the collation unit 141. If it is determined in step S230 that the speaker has been identified (matched) (“YES” in step S230), in step S250, the determination unit 123 determines to execute the function, and the speaker The recognition ON / OFF switching process ends. In step S230, if it is determined that the speaker is not specified (mismatch) by collation unit 141 ("NO" in step S230), function determination is rejected by determination unit 127 in S240. A determination is made, and the speaker recognition ON / OFF switching process ends. In this way, the speaker recognition ON / OFF switching process is completed.

  As described above, by incorporating the speaker recognition function into the voice operation and the manual operation, it is possible to prevent the third party from easily operating ON / OFF switching of the guard function of the car navigation device 20. . As a result, even if the car is stolen, important information included in the car navigation device 20 (for example, information that can identify personal information, such as a home position or a home phone number) (Including information) can be prevented from being stolen, and damage can be minimized.

  Further, by enabling the speaker recognition ON / OFF switching process, the guard function of the car navigation device 20 can be stopped and operated as necessary, and convenience is improved. For example, even when the user is away from the car for a while, it may be troublesome if the guard function is enabled. Therefore, if the guard function can be stopped and operated according to the desire of the owner, convenience is improved and usability is improved.

  Next, processing operation examples in the car navigation device 20 to which the present invention is applied will be described with reference to FIGS. FIGS. 6A to 6C show an operation example when a “return to home” command is executed by voice operation. FIG. 6A shows a case where the guard by the speaker recognition function is OFF, that is, normal voice operation processing. In this case, in response to a voice command “go home” from the user, the home is set as the destination as instructed by the user. Therefore, even if a command is input by the theft when the car is stolen, a command that can identify personal information is executed, and the location of the owner's home is revealed to the theft, for example. It will end up.

  FIG. 6B shows a processing operation example when the guard by the speaker recognition function is ON and the registered speaker user (owner) utters a voice command “go home”. In this case, the registered speaker is identified by the speaker recognition function in the speaker recognition / collation unit 141, and thereafter, the same processing operation example as shown in FIG.

  FIG. 6C shows an example of processing operation when the guard by the speaker recognition function is ON and a user (theft person) who is not a registered speaker user utters a voice command “go home”. For example, when a car is stolen. In this case, since the command input person is not identified as the registered speaker (owner) in the speaker recognition function in the speaker recognition verification unit 141 (mismatch), the function execution by the voice command “return to home” is rejected. In this example, the voice operation is forcibly terminated. As a result, even if the car is stolen, the command that can specify the personal information cannot be executed, so that the personal information of the owner is protected.

  7A to 7C show processing operation examples in the case of “setting home as a destination” by manual operation. The manual operation is performed by an input to the operation switch group 23 by the user, for example. FIG. 7A is a diagram illustrating normal manual operation processing when the guard by the speaker recognition function is OFF. In this case, in response to the user's manual operation of “setting home as the destination”, the home is set as the destination as instructed by the user. Therefore, even if a command is input by the theft when the car is stolen, a command that can identify personal information is executed, and the location of the owner's home is revealed to the theft, for example. It will end up.

  FIG. 7B is a diagram illustrating a processing operation example when the guard by the speaker recognition function is ON and the registered user (owner) performs a manual operation of “setting home as a destination”. In this case, with the speaker recognition function, the user is identified (matched) as a registered speaker (owner), and thereafter, the same processing operation example as in FIG. Here, speaker recognition processing by voice input is performed, but instead, speaker recognition processing may be performed by, for example, password input and verification.

  FIG. 7C shows an example of processing operation when the guard by the speaker recognition function is ON and a user (theft person) who is not a registered speaker user performs a manual operation of “setting home as a destination”. FIG. For example, when a car is stolen. In this case, since the speaker recognition function does not identify the registered speaker (mismatch), the function execution by the manual operation of “set home as the destination” is rejected, and the manual operation is forcibly terminated. It has become. As a result, even if the car is stolen, the command that can specify the personal information cannot be executed, so that the personal information of the owner is protected. Here, speaker recognition processing by voice input is performed, but instead, speaker recognition processing may be performed by, for example, password input and verification.

  FIGS. 8A to 8B show processing operation examples of the speaker recognition function ON / OFF switching processing. FIG. 8A is a diagram illustrating an operation when the speaker recognition function is ON and the registered speaker user performs “speaker recognition ON / OFF switching processing”. The “speaker recognition ON / OFF switching process” is, for example, input to the operation switch group 23 (manual input unit) or a “setting” button displayed on the display device 26 (manual input unit) having a touch panel function. And the input to the “Switch ON / OFF of speaker recognition” button displayed subsequently. When the “speaker recognition ON / OFF switching” button is pressed, the process proceeds to speaker recognition processing. In the speaker recognition process, as described above, the user's voice is input via the speaker 14 and processed by the voice recognition unit 12, the voice recognition control unit 13, and the speaker recognition unit 17. In this case, the speaker is identified as a registered speaker by the speaker recognition function, and the screen changes to a “speaker recognition ON / OFF switching” setting screen as instructed by the user. Here, speaker recognition processing by voice input is performed, but instead, speaker recognition processing may be performed by, for example, password input and verification.

  FIG. 8B is a diagram illustrating a processing operation example when a speaker recognition function is ON and a user who is not a registered speaker user performs “speaker recognition ON / OFF switching processing”. For example, when a car is stolen. In this case, since the speaker recognition function does not identify the registered speaker (mismatch), the screen transition to the “Speaker recognition ON / OFF switching” setting screen is rejected, and the operation is forcibly terminated. It has become. Here, speaker recognition processing by voice input is performed, but instead, speaker recognition processing may be performed by, for example, password input and verification.

  In the above-described embodiment, the “speaker recognition ON / OFF switching process” is performed by input to the operation switch group 23 (manual input unit) or the display device 26 (manual input unit) having a touch panel function. However, the present invention is not limited to this. For example, “speaker recognition ON / OFF switching process” may be performed by voice input from the speaker 14.

  FIGS. 9A and 9B are diagrams illustrating another processing operation example when guarding by a speaker recognition function is applied. FIG. 9A is a diagram showing processing when a voice operation is performed and the speaker recognition function is ON and a user (theft person) who is not a registered speaker user utters a voice command “go home”. . For example, it is a processing operation example when a car is stolen. In this case, the speaker recognition function does not identify the registered speaker (mismatch). Therefore, instead of simply rejecting the execution of the function by the voice command “return to home”, the operation is executed in a state where the home is not yet registered. For example, the response is “Your home is not registered. Please register your home before executing”.

  FIG. 9B shows processing when the speaker recognition function is turned on by manual operation and a user who is not a registered speaker user performs a manual operation of “setting home as a destination”. For example, when a car is stolen. In this case, the speaker recognition function does not identify the registered speaker (mismatch). Therefore, as in FIG. 9A, the function execution by manual operation is not simply rejected, but the function is executed in a state where the home is not yet registered. As described above, even if a theft is used, it can be determined that a guard to which the speaker recognition function is applied cannot be determined.

  The present invention is not limited to the above-described embodiments, and can be applied to various embodiments without departing from the scope of the invention. For example, the configuration of the car navigation device 20 and the series of processes shown in FIGS. 3 to 9 are given as examples, and can be appropriately modified and implemented without departing from the scope of the invention.

  If the matching level (reliability) in matching between the registered speaker data 143 and the input voice data in the matching unit 141 of the speaker recognition unit 17 is set, and the matching level (reliability) is considerably low, It may be determined that the theft is speaking (using). For example, the setting state of the car navigation device 20 may be controlled to be forcibly returned to the factory shipment state (initial state). In this case, since the home address of the owner or the like is not registered, the response of the car navigation device 20 is, for example, “Home is not registered. Please execute after registering the home” It becomes a response. Thus, by making the response the same as when the owner's personal information is not registered, the movement is the same as the normal operation. Therefore, since it is not realized that the guard is provided with the speaker recognition function for the theft, the damage can be minimized.

  Further, the command “return to home” has been described as an example of the command relating to the processing that can specify personal information, but the present invention is not limited to this. You may enable it to change (add, delete) the command used as the object which provides the above-mentioned guard. For example, when the car navigation device 20 is linked to a smartphone (portable device), it is possible to add not only the “return home” command but also the “call home” command so that it can be guarded. .

  By doing in this way, the command regarding the process which can specify various personal information can be made into the object of a guard function. Therefore, even when a car theft operator operates the car navigation device, various commands related to the function that can identify the owner's personal information are not executed, so that the owner's personal information is leaked. It can be avoided more fully.

    In the drawing, 10 is a speech recognition unit, 11 is a speech synthesis unit, 12 is a speech recognition unit, 13 is a speech recognition control unit, 14 is a speaker (speech output unit), and 15 is a microphone (speech input unit, second input unit). ), 16 a PTT switch, 17 a speaker recognition unit, 20 a car navigation device, 21 a position detector, 21 a a gyroscope, 21 b a distance sensor, 21 c a GPS receiver, 22 a data input device, 23 Operation switch group (manual input unit, first input unit), 24 is a communication device, 25 is an external memory, 26 is a display device (manual input unit, first input unit), 27 is a remote control sensor, 27a is a remote control, 28 is a control circuit (speaker recognition target determination unit, first determination unit), 101 is a speech extraction unit, 103 is a speech recognition collation unit, 105 is a speech recognition result output unit, 107 is a speech recognition dictionary unit, and 121 is a speech Recognition target determination unit (first determination unit), 123 is a mechanism execution process determination unit, 125 is a voice output content determination unit, 127 is a function execution rejection determination unit, and 141 is a speaker recognition verification unit (second determination unit). 143, registered speaker data, 201, a command correspondence dictionary, 203, an address correspondence dictionary, 205, a music correspondence dictionary, and 207, a telephone directory correspondence dictionary.

Claims (10)

A first determination unit (121) for determining whether or not a command corresponding to the input first audio data relates to a function capable of specifying the owner's personal information;
A second determination unit (141) that collates the first audio data with pre-registered second audio data and determines whether or not they match,
The command relates to a function capable of identifying the owner's personal information,
The car navigation device (20), wherein the command is not executed when the first voice data and the second voice data do not match as a result of the collation in the second determination unit. A first input unit (23, 26) capable of manually inputting a command;
A second input unit (15) capable of inputting voice data;
A first determination unit (121) for determining whether or not the command input from the first input unit relates to a function capable of specifying the owner's personal information;
A second determination unit (141) that compares the input voice data with the second voice data registered in advance and determines whether or not they match,
In the first determination unit, when it is determined that the command input from the first input unit is related to a function capable of specifying the owner's personal information,
A process of inputting first audio data to the second input unit;
In the second determination unit, the first voice data and the process of collating the second voice data are performed,
The car navigation device characterized in that the command is not executed when the first voice data and the second voice data do not match as a result of the collation in the second determination unit.   The car navigation device according to claim 1, wherein the process in the first determination unit and the process in the second determination unit can be switched on and off. The car navigation device according to claim 3, wherein when the ON / OFF switching is operated, the process in the first determination unit and the process in the second determination unit are performed.   The car navigation device according to any one of claims 1 to 4, wherein a command to be processed in the first determination unit can be changed (added and deleted). The car navigation device according to any one of claims 1 to 5, wherein the processing in the first determination unit is performed by password verification.   If the first audio data and the second audio data do not match as a result of collation in the second determination unit, the activation is stopped when the car navigation device is restarted. The car navigation device according to any one of claims 1 to 6.   The car navigation device according to claim 7, wherein the car navigation device cannot be restarted unless the person is verified by inputting a password.   If the first voice data and the second voice data do not match as a result of collation in the second determination unit, the owner's personal information is registered instead of not executing the command. The car navigation device according to any one of claims 1 to 8, wherein the same response as that in a case where the vehicle is not connected is performed.   In the collation of the first audio data and the second audio data in the second determination unit (141), a step is provided for the degree of coincidence between the two, and when the degree of coincidence is low, the setting state of the car navigation device is set. The car navigation device according to any one of claims 1 to 9, wherein the vehicle navigation device is forcibly returned to a factory-shipped state.

Images