JP2007057805A - Information processing apparatus for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow an information processing apparatus for a vehicle, that performs information processing in response to a voice, to be reacted with only with the voice of a driver for processing, that the driver should instructs to be performed, while reducing cost. <P>SOLUTION: The voice (acoustic feature value) of the driver is previously stored. Then only when a voice input from a microphone 35 is a voice for processing for the driver and the voice matches the previously stored voice (acoustic feature value) of the driver, the processing corresponding to the input voice is performed. None of hardware such as a CCD camera need not be used to decide whether the voice input from the microphone 35 is the voice of the driver, so that the cost is not raised. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a vehicle information processing apparatus that executes information processing of a vehicle by voice input from a user.

2. Description of the Related Art Conventionally, there is an information processing apparatus for a vehicle that performs information processing by voice instead of manual operation of a remote control or a switch when driving or the like (for example, Patent Document 1). However, in the conventional vehicle information processing apparatus, malfunction may occur when a plurality of occupants simultaneously emit sound. Therefore, in Patent Document 2, the seat position of the person who has input the sound is specified from a plurality of microphones that input the sound from a plurality of positions in the vehicle and a CCD camera for determining the presence or absence of a passenger in the vehicle seat.
Japanese Patent Laid-Open No. 11-288296 JP 2004-354930 A

  By the way, in the vehicle information processing apparatus that performs information processing by the above-described voice, there is information processing that the driver should instruct execution, such as information processing related to vehicle travel. On the other hand, there is information processing that does not hinder even if an occupant other than the driver instructs execution, such as information processing related to audio. In this respect, since it is possible to determine whether or not the input voice is the driver's voice by using the technique of Patent Document 2, of the information processing that can be executed by voice, the process that the driver should instruct execution is It is also possible to react only to the driver's voice.

  However, in Patent Document 2, since it is necessary to use a plurality of microphones and a CCD camera in order to specify the seat position of the person who has input the voice, the cost increases.

  The present invention has been made in view of the above problems, and in a vehicle information processing apparatus that performs information processing by voice, the driver's voice is used for processing that the driver should instruct execution while reducing costs. The only purpose is to react.

  In order to achieve the above object, an information processing apparatus for a vehicle according to claim 1 includes a voice input means for inputting voice, a process in which the voice input by the voice input means is to be instructed by the driver to execute, First determination means for determining whether or not the occupant is instructed to execute a process capable of instructing execution, including a process capable of instructing execution, and a voice print storing a driver's voice print Storage means; second determination means for determining whether or not the voice input by the voice input means is the voice of the driver based on the voiceprint stored in the voiceprint storage means; and the first When the determination unit determines that the voice input by the voice input unit is a voice instructing execution of a process that can be instructed to execute by the voice, the process to be instructed by the driver to be executed For, characterized by comprising an execution means for executing only the process to the audio of the second of said driver based on the determination result of the determining means.

  According to this, since it is not necessary to provide a plurality of microphones and a CCD camera as in Patent Document 2, it is possible to determine whether or not the input sound is the driver's sound, so that the driver can execute while reducing costs. The processing to be instructed can react only to the driver's voice.

  The vehicle information processing apparatus according to claim 2 is characterized in that the voiceprint storage means stores the voice input by the voice input means at the start of driving as the voiceprint of the driver. According to this, since the driver's voiceprint at the time of every driving | running | working is memorize | stored, it is effective when applied to the vehicle which a several person drives especially.

  The vehicle information processing apparatus according to claim 3, further comprising a canceling unit that stops storing the driver's voiceprint when starting the subsequent driving when the driver's voiceprint is already stored in the voiceprint storage unit. To do. For example, when the driver at each driving is almost determined by a specific person, it is not necessary to memorize the driver's voiceprint at each driving. On the contrary, if the driver's voiceprint is memorized at every driving, it is expected that it will be a burden on the driver. Claim 3 takes this into consideration.

  The vehicle information processing apparatus according to claim 4 is characterized in that the voiceprint storage means includes a switch for storing the voice inputted by the voice input means as the voiceprint of the driver. By providing the switch, a voice print can be stored at the driver's preference. This is particularly effective when the driver changes after starting driving.

  Embodiments to which the present invention is applied will be described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of a multimedia system as an embodiment. This multimedia system is installed in a vehicle and comprehensively controls or processes information such as map display using map data and navigation such as driving guidance, as well as image display and audio devices using other media. It is configured as a system like this.

  Specifically, the multimedia ECU 10, the display device 15 with a touch switch, the air conditioner ECU 51, the TV tuner 52, and the audio device 53 are connected to each other via the communication line 40.

  The multimedia ECU 10 is connected to a position detector 4, a map data input device 6, and an operation switch group 8 so that data can be input / output to / from the voice recognition device 30. Has been. In addition, data can be input / output with the display device 15 with a touch switch described above. Note that the multimedia ECU 10 includes a navigation control unit 10a and a multimedia control unit 10b, all of which include a well-known CPU, ROM, RAM, I / O, and a bus line that connects them. It is configured as a computer. The navigation control unit 10a is a main body that executes navigation-related processing, and the multimedia control unit 10b is processing related to other media, specifically, the air conditioner ECU 51, the TV tuner 52, and the audio device 53 described above. It becomes the subject that executes.

  Each of the position detectors 4 includes a known gyroscope 18, a distance sensor 20, and a GPS receiver 22 for GPS (Global Positioning System) that detects the position of the vehicle based on radio waves from satellites. . Each of these sensors 18, 20, and 22 has an error of a different nature, and is configured to be used while being interpolated by a plurality of sensors. Depending on the accuracy, a part of the above may be used, and further, a steering rotation sensor, a wheel sensor of each rolling wheel, or the like may be used.

  The map data input device 6 is a device for inputting various data including so-called map matching data, map data, and landmark data for improving the accuracy of position detection. As a medium, a CD-ROM or DVD is generally used because of its data amount, but a medium such as a memory card may be used.

  The touch switch display device 15 includes a touch switch 12 that blocks a predetermined number of infrared rays running vertically and horizontally when the screen is touched with a finger, identifies a pressed area, and performs a switch function, and controls the entire display device. A display ECU 13 and an LCD monitor 14. The LCD monitor 14 can display in color, and on its screen, the vehicle current position mark input from the position detector 4, the map data input from the map data input device 6, and the guidance displayed on the map are displayed. Additional data such as a route and landmarks for setting points to be described later can be displayed in an overlapping manner. This is a method of use as a navigation device. For example, a television image of a channel selected by the TV tuner 52 can be displayed.

  In the case of the present embodiment, the operation switch group 8 is a mechanical switch disposed around the display device 15 with a touch switch (or integrally with the housing of the display device) and is mainly used. Used for media selection. Specifically, it is a switch for selecting a desired one from various media such as navigation, television, audio, air conditioner, and CD.

  The voice recognition device 30 is used to instruct the destination or the like similarly when the user inputs the voice, while the operation switch group 8 is used to instruct the destination or the like by manual operation. It is a device for making it possible. The configuration of the speech recognition device 30 will be further described with reference to FIG.

  The voice recognition device 30 includes a voice recognition unit 31, a dialogue control unit 32, a voice synthesis unit 33, a voice input unit 34, a microphone 35, a PTT (Push-To-Talk) switch 36, and a speaker 37. And a PTT switch control unit 38.

  The voice recognition unit 31 performs input voice recognition processing on the voice data input from the voice input unit 34 according to an instruction from the dialogue control unit 32, and returns the recognition result to the dialogue control unit 32. That is, the speech data acquired from the speech input unit 34 is collated using the stored dictionary data, and the dialogue control unit 32 sets a higher comparison pattern having a higher degree of matching than a plurality of comparison pattern candidates. Output to. The recognition of the word sequence in the input speech is performed by sequentially analyzing the speech data input from the speech input unit 34 to extract the acoustic feature amount (for example, cepstrum), and the acoustic feature amount obtained by this acoustic analysis. Get series data. Then, the time series data is divided into several sections by a known DP matching method, HMM (Hidden Markov Model), or a neural network, and it is determined which word is stored as dictionary data for each section. . Furthermore, the voice recognition unit 31 outputs an acoustic feature amount corresponding to the word together with the word recognized by the dialogue control unit 32.

  From the recognition result and the internal state managed by the dialog control unit 32, the interactive control unit 32 instructs the speech synthesizer 33 to generate a response voice and performs processing of the system itself, for example, for navigation processing. A process for instructing to execute the setting process by notifying the necessary destination is executed. At this time, the dialogue control unit 32 instructs the execution of only a predetermined process among the processes for instructing the multimedia ECU 10 to execute. Specifically, instructions related to navigation processing (for example, changing the scale of the map screen or setting the destination) are instructed to be executed only for the audio from the driver, and other processing (for example, audio volume or air conditioner (Temperature setting, etc.) is also instructed to be executed for voices from passengers other than the driver. The processing at this time is a characteristic part of the present invention, and will be described later using a flowchart.

  Further, in order to instruct execution only for the voice from the driver as described above, the dialogue control unit 32 also determines whether or not the voice data input from the voice input unit 34 is the voice of the driver. Specifically, the dialogue control unit 32 can store a driver's voice print (acoustic feature amount), and the acoustic feature amount input from the speech recognition unit 31 and the acoustic feature amount of the driver. The voiceprint determination unit 59 determines whether or not the voice data input from the voice input unit 34 is the driver's voice.

  The above processing is post-determination processing. As a result, if this voice recognition device 30 is used, it is possible to specify a destination for navigation processing by voice input without operating the operation switch group 8. It becomes.

  The voice input unit 34 converts the surrounding voice captured by the microphone 35 into digital data and outputs the digital data to the voice recognition unit 31. Note that the voice input unit 34 outputs only voices having a predetermined threshold value or more to the voice recognition unit 31 so that noise input from the microphone 35 is not output to the voice recognition unit 31. In addition, the PTT control unit 38 monitors whether or not the PTT switch 36 is pressed, and if the PTT switch 36 is pressed, outputs the fact to the voice input unit 34 and the dialogue control unit 32. . Then, the dialogue control unit 32 stores the acoustic feature quantity input from the voice recognition unit 31 at this time as the acoustic feature quantity of the driver.

  Returning to the description of the configuration in FIG. 1, the air conditioner ECU 51 controls the operation of the air conditioner.

  The TV tuner 52 is a device for selecting and receiving a television broadcast signal, and can allocate a broadcast signal having a predetermined frequency to a so-called preset memory. Then, if the occupant designates the preset memory number, the television broadcast signal of the corresponding broadcasting station can be received. The received television image can be displayed on the LCD monitor 14 of the display device 15.

  The audio device 53 is a device that plays music media, receives a radio signal of a predetermined frequency, and outputs it to a speaker. As for the radio signal of the predetermined frequency, the frequency corresponding to the radio signal that the occupant wants to receive each time can be designated, or the frequency corresponding to each radio station can be assigned to the preset memory in advance as in the TV tuner 52 described above. .

  The air conditioner ECU 51, the TV tuner 52, and the audio device 53 perform various operations by manual operation of the occupant, and in the present embodiment, the air conditioner ECU 51, the TV tuner 52, and the audio device 53 operate by the occupant's voice. ing. For example, when the occupant says “Set the temperature of the air conditioner to xx ° C.”, the voice recognition device 30 recognizes the voice, and based on the recognition result, the multimedia control unit 10b, which will be described later, The ECU 51 is instructed to set the set temperature to xx ° C.

  A supplementary description will be given of the navigation control unit 10a in the multimedia ECU 10. When using the navigation function, for example, when the driver selects a route information display process for displaying the guidance route on the LCD monitor 14 by the operation switch group 8 from the menu displayed on the LCD monitor 14, or When a similar instruction is given from the dialogue control unit 32 to the navigation control unit 10a by inputting a desired menu through the microphone 35 via the voice recognition device 30, the following processing is performed. That is, when the driver inputs a destination by voice or operation of the operation switch group 8 based on the map on the LCD monitor 14, the current location of the vehicle is obtained based on the hygiene data obtained from the GPS receiver 22. The cost is calculated between the current location and the current location by the Dijkstra method, and the shortest route from the current location to the destination is obtained as a guidance route. Then, the guidance route is displayed overlaid on the road map on the LCD monitor 14, and an appropriate route is guided to the driver, such as an enlarged display of the intersection and voice guidance about the intersection to be bent. Such calculation processing and guidance processing for obtaining a guidance route are generally well-known processing. The voice guidance condition setting and the language (for example, Japanese or English) to be displayed on the screen can be freely set. Furthermore, the user can arbitrarily register the destination and other points that are convenient to register.

  As described above, the multimedia system according to the present embodiment reacts only to the driver's voice when the voice is input from the microphone 35 and the voice instructs the execution of the processing related to navigation. In the case of instructing execution of processing relating to the air conditioner ECU 51, the TV tuner 52, and the audio device 53, it also responds to the voices of passengers other than the driver. For this purpose, it is necessary to determine whether or not the voice input from the microphone 35 is the voice of the driver. For this judgment, the voice of the driver (acoustic feature value) is stored in advance. The processing at this time will be described with reference to the flowchart of FIG. This process is performed by the voice recognition device 30.

  First, in step S10, it is determined whether or not the PTT switch 36 is turned on. This determination is made by the PTT switch control unit 38. Here, when the PTT switch 36 is not turned on, the subsequent processing is not performed. That is, the process of storing the driver's voice (acoustic feature value) is not performed.

  On the other hand, if the PTT switch 36 is on, it is next determined in step S11 whether or not a voice is input from the microphone 35. Specifically, the PTT switch control unit 38 outputs a signal indicating that the PTT switch 35 is turned on to the dialogue control unit 32, and the dialogue control unit 32 sends the signal from the speaker 37 to the driver via the voice synthesis unit 33. Encourage them to utter certain words. The predetermined word may be any word as long as an acoustic feature amount can be obtained, such as the name of the person. In addition, the PTT switch control unit 38 also outputs a signal indicating that the PTT switch 36 is turned on to the voice input unit 34, and whether the voice input unit 34 receives a voice input from the microphone 35 based on the output signal. Determine whether or not. Note that the PTT switch control unit 38 may determine whether or not a voice is input from the microphone 35. In this case, the PTT switch control unit 38 may refer to the voice input unit 34. Further, whether or not voice input is made from the microphone 35 is determined based on whether or not it is recognized that voice input is made within a predetermined time. Here, when no sound is input from the microphone 35, the subsequent processing is not performed. It should be noted that at this time, without promptly leaving the flowchart, the user is prompted to utter a predetermined word again, or returns to step S10 again to make a determination of step S11 a predetermined number of times. It is also possible to leave the flowchart.

  On the other hand, when a voice is input from the microphone 35, in step S12, the voice is stored as a driver voice (acoustic feature value). Specifically, as described above, the dialogue control unit 32 stores the voice (acoustic feature amount) input from the voice recognition unit 31. Here, the stored voice (acoustic feature value) becomes the driver's voice (acoustic feature value).

  As described above, when the PTT switch 36 is turned on, the driver's voice (acoustic feature value) is stored. This is for distinguishing from the voice for instructing normal navigation processing and the like.

  Next, a process for executing a navigation-related process or the like by voice on the assumption that the driver's voice (acoustic feature value) is stored will be described with reference to the flowchart of FIG. This process is performed by the voice recognition device 30 and the multimedia ECU 10.

  First, in step S20, it is determined whether or not sound is input from the microphone 35. Here, when no sound is input from the microphone 35, the subsequent processing is not performed. On the other hand, when voice is input from the microphone 35, in step S21, the voice recognition unit 31 performs processing for recognizing the content of the input voice. Specifically, the acquired voice data is collated using stored dictionary data. Then, the upper comparison target pattern determined based on the collation result is output to the dialogue control unit 32 as a recognition result. At this time, an acoustic feature quantity is also output to the dialogue control unit 32 in order to determine whether or not the voice input from the microphone 35 is the voice of the driver.

  Next, in step S22, it is determined whether or not the voice input from the microphone 35 is the voice of the driver. Specifically, the voiceprint determination unit 59 in the dialogue control unit 32 collates the acoustic feature amount input from the speech recognition unit 31 with the acoustic feature amount of the driver stored in the dialogue control unit 32. . If it is determined that the voice input from the microphone 35 is the voice of the driver, the process proceeds to step S23.

  In step S23, it is determined whether or not the voice input from the microphone 35 is a voice instructing processing limited to the driver. In the present embodiment, the sound for all the processes that can be instructed to be executed by sound is used as the sound for the driver process. That is, the driver can instruct the execution of all of the processing related to navigation and the processing for the air conditioner ECU 51, the TV tuner 52, and the audio device 53 by voice. Note that whether or not the voice input from the microphone 35 is a voice for the driver's processing is determined based on the signal input from the voice recognition unit 31. Here, when it is determined that the sound input from the microphone 35 is the sound for the driver process, in step S24, the process to be executed is talked back. This is performed by the dialogue control unit 32 using the speaker 37 via the speech synthesis unit 33. As a result, the driver can confirm in advance whether or not it matches the processing desired by the driver. Thereafter, in step S25, processing corresponding to the input voice is executed. This is made possible by notifying the multimedia ECU 10 of the processing content executed by the dialogue control unit 32, and the multimedia ECU 10 instructs the corresponding device to execute the processing based on the notification content. On the other hand, when it is determined that the voice input from the microphone 35 is not the voice for the driver process, the subsequent process is not performed. Note that the case where the voice is not for the driver process is, for example, for a process for which an execution instruction cannot be given by voice, or for a process for which an execution instruction can be given by voice, even if the voice is given for an appropriate voice. There are cases where there is no such thing.

  If it is determined in step S22 that the sound input from the microphone 35 is not the driver's sound, the process proceeds to step S26. In step S <b> 26, it is determined whether or not the voice input from the microphone 35 is a voice instructing a general process that can be instructed by an occupant other than the driver. Here, the general process that can be instructed by an occupant other than the driver is a process other than the process related to navigation. Specifically, the process performed by the multimedia control unit 10b in the multimedia ECU 10 (air conditioner ECU 51, This refers to a process in which an execution instruction can be given by voice in the TV tuner 52 and the audio device 53). For example, changing the set temperature of the air conditioner, selecting a TV broadcast radio station, adjusting the volume, and the like. Here, when it is determined that the sound input from the microphone 35 is for a general process that can be executed even by an occupant other than the driver, the process to be executed in step S24 is talkbacked in the same manner as described above. Then, the process is executed in step S25. On the other hand, when it is determined that the voice input from the microphone 35 is not for general processing that can be executed by an occupant other than the driver, the subsequent processing is not performed. In addition, in the case where it is not for general processing that can be instructed even by an occupant other than the driver, in addition to the case similar to the above (determination of step S23), in the case of voice for driver processing (voice for navigation-related processing) Etc.

  As described above, in the multimedia system according to the present embodiment, various processes can be executed by the passenger's voice. At this time, the driver's voice (acoustic feature amount) is stored in advance for the navigation-related processing. Therefore, it reacts only to the driver's voice. As a result, the driver can travel with peace of mind. Further, since it is not used for hardware such as a plurality of microphones and a CCD camera as in Patent Document 2 to determine whether or not the sound is a driver's voice, the cost is not increased. Further, by using the PTT switch to store the driver's voice (acoustic feature quantity), the driver can store the voice (acoustic feature quantity) at a desired time. This is particularly effective when the driver changes after starting driving.

The vehicle information processing apparatus according to the present invention is not limited to the above-described embodiment, and may be variously modified without departing from the spirit thereof.
(Modification)

  In the above embodiment, in order to store the driver's voice (acoustic feature), the voice input when the PTT switch is turned on is stored as the driver's voice (acoustic feature) ( (See FIG. 3). This is for distinguishing from voice for instructing normal navigation processing or the like as described above. However, the present invention is not limited to this, and a voice input from the microphone 35 at the start of driving may be stored as a driver voice (acoustic feature value) without using a PTT switch. For example, when the engine of the vehicle is started, the driver is urged to utter predetermined words such as his / her name. The voice input along with it is stored as the driver's voice (acoustic feature value). Thus, since the driver's voice (acoustic feature value) is stored at the start of each driving, it is particularly effective when applied to a vehicle driven by a plurality of people.

  In addition, it can be assumed that there are some drivers who find it cumbersome to memorize the driver's voice (acoustic feature) at the start of every driving, so the driver's voice (acoustic feature) is already stored. In such a case, the process of storing the driver's voice (acoustic feature value) may be stopped at the start of subsequent driving.

It is a block diagram which shows schematic structure of the multimedia system of embodiment. It is a block diagram which shows schematic structure of a speech recognition apparatus. It is a flowchart which shows the process which memorize | stores the audio | voice (acoustic feature-value) of a driver. It is a flowchart which shows the process which performs the process regarding navigation, etc. by an audio | voice.

Explanation of symbols

4 Position detector 6 Map data input device 8 Operation switch group 10 Multimedia ECU
DESCRIPTION OF SYMBOLS 15 Display apparatus 30 Voice recognition apparatus 51 Air-conditioner ECU
52 TV tuner 53 Audio device

Claims (4)

Voice input means for inputting voice;
The voice input by the voice input means is a voice that instructs execution of a process that can be instructed to execute, including a process that should be instructed by the driver and a process that can be instructed to the passenger other than the driver. First determination means for determining whether or not
Voiceprint storage means for storing the driver's voiceprint;
Second determination means for determining whether or not the voice input by the voice input means is the voice of the driver based on the voice print stored in the voice print storage means;
When the first determination unit determines that the voice input by the voice input unit is a voice instructing execution of a process that can be instructed to execute by the voice, An information processing apparatus for a vehicle, comprising: execution means for executing processing only for the voice of the driver based on a determination result of the second determination means.   The vehicle information processing apparatus according to claim 1, wherein the voiceprint storage unit stores the voice input by the voice input unit at the start of driving as the voiceprint of the driver.   3. The vehicle information processing according to claim 2, further comprising: a canceling unit that stops storing the driver's voiceprint at the start of subsequent driving when the driver's voiceprint is already stored in the voiceprint storage unit. apparatus.   The vehicle information processing apparatus according to claim 1, wherein the voiceprint storage unit includes a switch that stores the voice input by the voice input unit as the voiceprint of the driver.

Images