JP6281202B2 - Response control system and center - Google Patents

Description

  The present invention relates to a technique for automatically responding to voice data input by a user.

  2. Description of the Related Art Conventionally, there is a response system that performs voice recognition processing on voice data input from a user and provides the user with various information according to the recognition processing result (for example, Patent Document 1). The response system disclosed in Patent Document 1 includes an on-vehicle device and an information center that is provided outside the vehicle and performs wireless communication with the on-vehicle device.

  In the response system disclosed in Patent Literature 1, the vehicle-mounted device acquires the user's voice data and transmits the voice data to the information center. On the other hand, the information center performs voice recognition processing on the voice data, and returns response voice data corresponding to the recognition processing result to the vehicle-mounted device. Then, the vehicle-mounted device that has received the response voice data returns the voice according to the response voice data.

  On the other hand, response systems that provide various kinds of information in response to voice input by users by wireless communication between an information center managed by a mobile phone company and the mobile phone have become widespread. In addition, not only a mobile phone but a response system that automatically responds to a voice input by a user in various scenes and environments is being used.

JP 2004-348658 A

  In recent years, a technique for operating an in-vehicle device and a mobile phone in cooperation with each other has also been developed. Thus, when operating an onboard equipment and a mobile telephone in cooperation, it becomes the composition which can use a plurality of response systems via an onboard equipment.

  However, in such a configuration, the user must input a voice after selecting a response system according to the purpose from a plurality of response systems by operating the vehicle-mounted device. In addition, when a different response system is to be used after selecting the response system, an operation for switching the response system to be used must be performed.

The present invention has been made based on this circumstance, and the object of the present invention is to provide a response system that makes a response when a plurality of response systems that respond to a user's voice input are available. response control system that allows eliminating the need to select to provide a contact and center.

The invention of the response control system for achieving the object includes an audio acquisition unit (18A) that acquires input audio by a user as input audio data via a microphone (12), and input audio data acquired by the audio acquisition unit. The voice response processing is performed on the first response system (3) that generates the first response voice data that is a response to the input voice data based on the result of the voice recognition process, and the voice acquisition unit acquires A second response system that performs voice recognition processing on the input voice data and generates second response voice data that is voice data that is a response to the input voice data based on the result of the voice recognition processing ( and 4) the corresponding list storage unit that stores a correlation list (35A) where the first response system that describes the contents of the input audio data to respond (35), the first response An answering system determination unit (32A) for determining which one of the first and second response systems should respond to the input voice data based on the result of the voice recognition processing by the system; Answer output unit (18D) for outputting response voice data generated by the answer side response system, which is a response system that is determined to respond to the input voice data in the answer side system determination unit, to the speaker (14). The answering system determination unit determines that the first response system should respond to the input voice data when the result of the voice recognition processing is associated with the correspondence list. On the other hand, if the result of the speech recognition process is not associated with the correspondence list, the second response system should respond to the input speech data. Wherein the determining.

  In the above configuration, the answering system determination unit should respond to the input voice data among the first and second response systems based on the result of the voice recognition processing by the first response system. Determine whether. Then, the answer output unit causes the speaker to output response voice data from the response system that has been determined to be answered by the answer determination unit.

  According to such a configuration, since the answering system determination unit automatically selects a response system according to the contents of the input voice data, the user has to select a response system that asks for an answer to the question every time a question is asked. Can be omitted, and the convenience of the user can be improved.

  The center invention bears the function as the first response system and includes the answering system determination unit.

It is a block diagram which shows an example of a schematic structure of the response control system 100 concerning this embodiment. 3 is a block diagram illustrating an example of a schematic configuration of a first center 3. FIG. 3 is a block diagram illustrating an example of a schematic configuration of a control unit 18. FIG. It is a flowchart which shows the flow of the response switching process which the control part 18 implements. It is a flowchart which shows the flow of the 1st center response process which the 1st center side control part 32 implements. 3 is an example of a display screen of the display device 13. It is a block diagram which shows an example of a schematic structure of the response control system 100A of the modification 4.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS. FIG. 1 is a diagram illustrating an example of a schematic configuration of a response control system 100 according to the present embodiment. As shown in FIG. 1, the response control system 100 includes a navigation device 1, a mobile phone 2, a first center 3, and a second center 4. The navigation device 1 and the first center 3, the navigation device 1 and the mobile phone 2, and the mobile phone 2 and the second center 4 each perform data transmission / reception using a known wireless communication technology.

  In the present embodiment, the first center 3 and the second center 4 each operate as a response system, and the navigation device 1 operates as a user interface for using each response system (the vehicle-mounted device described in the background art section and claims). To do. The first center 3 corresponds to the first response system described in the claims, and the second center 4 corresponds to the second response system described in the claims. Hereinafter, a vehicle equipped with the navigation device 1 is referred to as a host vehicle.

  The first center 3 is an information center of an automobile company as an example, and is a voice for responding to questions related to navigation device operation, traffic jam information, and own vehicle operation, and using functions of the navigation device 1. Corresponds to an instruction by input (this is an instruction command). As an outline of the operation of the first center 3, the content of the question of the user is analyzed by performing a voice recognition process based on the voice data transmitted from the navigation device 1. Then, it determines whether the center should respond to the user's question, generates response voice data that is a response to the user's question, and returns it to the navigation device 1. For example, the first center 3 tells the location of the nearest convenience store from the user's current location in response to the user's question “Where is the convenience store?”. Hereinafter, the configuration of the first center 3 will be described in more detail with reference to FIG.

  As shown in FIG. 2, the first center 3 includes a first center side communication unit 31, a first center side control unit 32, a voice recognition unit 33, a voice recognition database (hereinafter, the database is abbreviated as DB) 34, a first center. A side memory 35 and a speech synthesizer 36 are provided. The first center side communication unit 31, the first center side control unit 32, the voice recognition unit 33, the voice recognition DB 34, the first center side memory 35, and the voice synthesis unit 36 each have a bus 37 compliant with, for example, a known communication standard. Are connected so that they can communicate with each other.

  The first center side communication unit 31 has a function of performing various signal processing such as modulation / demodulation for data transmission / reception, and communicates with the navigation apparatus 1 via a network such as a mobile phone network or the Internet network. To implement. The first center side communication unit 31 outputs data sent from the navigation device 1 to the first center side control unit 32 and transmits data input from the first center side control unit 32 to the navigation device 1. To do. The data sent from the navigation device 1 to the first center 3 includes current location information of the navigation device 1 as well as voice data spoken by the user. It is assumed that the current location information is sent sequentially (for example, every 100 milliseconds).

  In the speech recognition DB 34, as data necessary for speech recognition processing, for example, an acoustic model in which acoustic features of small units (phonemes) of human utterances are described, a recognition dictionary that associates acoustic features of phonemes with words, words A language model that expresses the connection relationship between the two is stored. Note that the speech recognition DB 34 in the present embodiment is a large-scale database corresponding to, for example, 1000 to tens of thousands of words. The voice recognition DB 34 also stores voice data used by the voice synthesizer 36 to generate response voice data.

  The voice recognition unit 33 performs voice recognition processing using various data stored in the voice recognition DB 34 from the input voice data input from the first center side control unit 32. Since the voice recognition process may use a known technique, a description thereof is omitted here. The result of the voice recognition process is output to the first center side control unit 32. The voice synthesizing unit 36 generates response voice data by synthesizing the voice waveform data stored in the voice recognition DB 34 based on an instruction from the first center side control unit 32. The generated response voice data is output to the first center side control unit 32. The response voice data generated in the first center 3 corresponds to the first response voice data described in the claims.

  The first center memory 35 is a mass storage device such as a writable HDD (Hard Disk Drive). The first center side memory 35 stores a correspondence list 35A and response data 35B. In the correspondence list 35A, a list of questions and command commands to be handled by the center is registered. The response data 35B is data for responding to questions and command commands from the user.

  For example, in the response data 35B, an answer corresponding to each question registered in the correspondence list 35A and a pattern of a control signal for controlling the operation of the navigation device 1 corresponding to the command command are registered. . Based on the response data 35B, the first center-side control unit 32 returns response voice data to the user's question, or returns a control signal for controlling the operation of the navigation device 1 to the navigation device 1. .

  In this embodiment, the response data 35B for generating response voice data is a group of commands that specify voice waveform data necessary for generating response voice data, but is not limited thereto. Voice data registered in advance as an answer corresponding to a question registered in the correspondence list 35A may be used. In this case, since the voice data may be returned to the navigation device 1 as response voice data, the first center 3 does not need to include the voice synthesizer 36.

  In the present embodiment, an HDD is used as the first center-side memory 35, but other known storage media such as a DVD and a flash memory may be used. The first center-side memory 35 corresponds to a correspondence list storage unit described in claims.

  The first center-side control unit 32 is configured as a computer, and includes a well-known CPU, a nonvolatile memory such as a ROM and an EEPROM, a volatile memory such as a RAM, an I / O, and a bus line that connects these configurations ( All are not shown). The nonvolatile memory stores a program for executing various processes. The first center side control unit 32 controls processing performed by the first center side communication unit 31, the voice recognition unit 33, and the voice synthesis unit 36.

  For example, when the first center-side control unit 32 acquires the voice data sent from the navigation device 1 (this is input voice data) from the first center-side communication unit 31, the first center-side control unit 32 recognizes the voice data. Output to the unit 33, and causes the voice recognition unit 33 to perform voice recognition processing.

  Based on the result of the speech recognition processing acquired from the speech recognition unit 33, the response data 35B stored in the first center side memory 35 is referred to, and the response speech data corresponding to the input speech data is speech synthesized. It is made to generate in part 36. When the response voice data is acquired from the voice synthesis unit 36, the response voice data is transmitted from the first center side communication unit 31 to the navigation device 1.

  Moreover, the 1st center side control part 32 is provided with the answer side determination part 32A and the update part 32B as a functional block, as shown in FIG. The updating unit 32B performs step S219 of the flowchart shown in FIG. A detailed description of the updating unit 32B will be described later.

  The answering side determination unit 32A determines whether or not the own center should respond to the input voice data from the result of the voice recognition process acquired from the voice recognition unit 33 and the correspondence list 35A. For example, if the user's question is in the range registered in the correspondence list 35A as a result of the voice recognition process, it is determined that the center should respond to the input voice data.

  When the user's question is a question outside the range registered in the correspondence list 35A, it is determined that the center should not respond to the input voice data. The determination result is transmitted to the navigation device 1 as a determination result signal prior to transmission of the response voice data. The answer side determination unit 32A corresponds to the answer side system determination unit described in the claims.

  Even if the answering side determination unit 32A determines that the center should not answer, the first center side control unit 32 cooperates with the speech synthesis unit 36 and the like. It may be configured to generate response voice data such as “Did it?” And return it to the navigation device 1. The processing when the answering side determination unit 32A determines that the center should not answer may be designed as appropriate, but some response voice data is generated and returned to the navigation device 1.

  In addition, even when a control signal for controlling the operation of the navigation device 1 is transmitted, response voice data that allows the user to recognize what operation is to be performed is also transmitted. However, as will be described later, when the content of the input voice data is a request to output a response from another center, the response from the other center is not sent without returning the response voice data. May be transmitted as a control signal.

  Therefore, in the present embodiment, the response voice data is returned when the input voice data is acquired, unless the content of the input voice data is a request to output an answer from another center. To do. Of course, as another configuration, when the content of the input voice data is a command command, the response voice data may not be returned.

  The mobile phone 2 is a well-known mobile phone, and communicates with the second center 4 via the network and also communicates with the second communication unit 17 included in the navigation device 1. For example, the mobile phone 2 converts the signal received from the second communication unit 17 into a signal according to the communication standard between the mobile phone 2 and the second center 4 and transmits the signal to the second center 4. In addition, the signal received from the second center 4 is converted into a signal according to the communication standard between the mobile phone 2 and the second communication unit 17 and transmitted to the second communication unit 17.

  The second center 4 is an information center of a mobile phone company, for example, and provides various services to the user of the mobile phone 2 using the mobile phone network managed by the mobile phone company. The second center 4 has the same configuration as the first center 3 except that the answer side determination unit 32A is not provided. That is, based on the voice data transmitted from the navigation device 1, the user's question content is analyzed by performing voice recognition processing. Then, response voice data serving as a response to the user's question is generated and returned to the navigation device 1. However, since the second center 4 does not have a function corresponding to the answer side determination unit 32A, it is not determined whether or not the center should respond to the user's question.

  Taking the case where a schedule management function is assumed as a service provided by a mobile phone company as an example, for example, the second center 4 responds to a user's question “What is today's schedule?” It will tell you the schedule for the day. In addition, the second center 4 acquires data (such as schedule information for the day) that is the basis for generating the response voice data by the mobile phone 2 and the second center 4 performing various data communications. It is good also as composition to do. The response voice data generated by the second center 4 corresponds to the second response voice data described in the claims.

  The navigation device 1 is mounted on a vehicle, and has a function of performing route guidance similar to that of a general navigation device. In addition, for example, the navigation device 1 is connected via the first center 3 or the mobile phone 2. It has a function of communicating with the second center 4. As shown in FIG. 1, the navigation device 1 includes a talk switch (hereinafter referred to as talk SW) 11, a microphone (hereinafter referred to as microphone) 12, a display device 13, a speaker 14, a memory 15, a first communication unit 16, and a second communication unit. 17 and a control unit 18.

  The talk SW 11 is for instructing the user (driver) to start voice input, and is provided at a position where the user can easily operate, for example, near the side surface of the steering column cover or the shift lever. As an example, the talk SW 11 is a so-called click-type switch. When the talk SW 11 is turned on by a user operation (ie, clicked), an on signal is output to the control unit 18.

  When an ON signal is input from the talk SW 11, the control unit 18 performs processing for acquiring audio data and starts connection with the first center 3 and the second center 4. The user can input the spoken voice to the navigation device 1 by starting to speak within a certain time (for example, within 1.5 seconds) after turning on the talk SW 11.

  Note that the timing for ending the voice input when starting to speak within a certain time after pressing the talk SW 11 is when the power level of the voice signal input from the microphone 12 is below a certain threshold, as will be described later. And

  The microphone 12 is, for example, a small omnidirectional microphone, collects surrounding sounds such as voices and noises spoken by the user, converts them into electrical voice signals, and outputs them to the control unit 18. The microphone 12 is provided at a position where the user's voice can be easily picked up, such as an upper surface portion of the steering column cover or a sun visor on the driver's seat side.

  The display device 13 displays text and images based on the input from the control unit 18 and notifies the user of various information. The display device 13 is arranged, for example, in the center of the instrument panel or in a combination meter provided in front of the driver's seat. The display device 13 is capable of full color display, for example, and can be configured using a liquid crystal display, an organic EL display, a plasma display, or the like.

  The speaker 14 converts the electrical sound signal input from the control unit 18 into sound (including simple sound) and outputs the sound. The memory 15 is a storage device that stores various data, and mainly stores response voice data acquired from the first center 3 and the second center 4 via various communication units 16 and 17 and user voice data. To do. The memory 15 may be configured using a known storage medium, and may be an HDD or a removable memory (for example, an SD card) having a relatively small storage capacity.

  The first communication unit 16 includes a transmission / reception antenna (not shown), and performs communication with the first center 3 via a communication network. The first communication unit 16 can employ various devices such as an in-vehicle communication module such as DCM (Data Communication Module) used for telematics communication, for example. The first communication unit 16 demodulates the signal received from the first center 3 and inputs the demodulated signal to the control unit 18, and modulates the data input from the control unit 18 and transmits the data to the first center 3.

  The second communication unit 17 includes a transmission / reception antenna (not shown), and exchanges information with the mobile phone 2 by performing communication (hereinafter referred to as BT communication) in accordance with the Bluetooth (registered trademark) standard. . In the present embodiment, the configuration in which the communication between the mobile phone 2 and the second communication unit 17 is performed by BT communication is shown, but the configuration is not necessarily limited thereto. For example, it may be configured to perform wireless communication according to short-range wireless communication standards such as ZigBee (registered trademark) or wireless LAN standards such as IEEE 802.11, or may be configured to perform wired communication such as USB communication. The second communication unit 17 only needs to have a function such as modulation / demodulation according to the communication standard with the mobile phone 2.

  The control unit 18 is configured as a normal computer, and includes a well-known CPU, a non-volatile memory such as a ROM and an EEPROM, a volatile memory such as a RAM, an I / O, and a bus line connecting these configurations (all are (Not shown). The nonvolatile memory stores a program for executing various processes. As shown in FIG. 3, the control unit 18 includes a voice acquisition unit 18A, a response temporary storage unit 18B, a response side center setting unit 18C, a response output unit 18D, and a response feedback unit as functional blocks for executing various processes. 18E, an agent display control unit 18F, and another response request determination unit 18G.

  Based on the ON signal from the talk SW 11, the sound acquisition unit 18 </ b> A performs a process of acquiring sound data obtained by removing noise components from the sound signal input from the microphone 12. For example, when the ON signal is input, the voice acquisition unit 18A enters a standby state in which the voice signal input from the microphone 12 can be converted into voice data. When a state in which no sound is input continues for a certain period of time (for example, 1.5 seconds) after entering the standby state, the conversion is automatically disabled. If it is determined that the voice is input within a certain time, the voice signal input from the microphone 12 until it is determined that the voice input is completed is converted into voice data.

  Whether or not the voice is input and whether or not the voice input has ended may be determined by using a known technique, for example, by determining whether or not the power level of the voice signal is equal to or higher than a predetermined threshold. Of course, in such a configuration, when an audio signal having a power level equal to or higher than the threshold value is input, it is determined that the audio is input. A publicly known technique may be used as a method for acquiring these audio data. The voice data acquired by the voice acquisition unit 18A is converted into a digital packet signal using a known technique such as VoIP (Voice over Internet Protocol) and transmitted to the first center 3 via the first communication unit 16. .

  The answer temporary storage unit 18 </ b> B temporarily stores the response voice data received from the first center 3 and the second center 4 in the memory 15. It is assumed that the answer temporary storage unit 18B stores these audio data until at least, for example, NO is determined in step S115 in FIG. The answer temporary storage unit 18B corresponds to the temporary storage unit described in the claims.

  In response to the determination result signal sent from the first center 3, the answer-side center setting unit 18C sends response voice data acquired from either the first center 3 or the second center 4 to the question input by the user. To output from the speaker 14 as an answer to the above. Then, the center that answers the question input by the user is set as the answering center, and the center that is not set as the answering center is set as the semi-answering center.

  More specifically, when the determination result signal indicates that the first center 3 should respond, the response obtained from the first center 3 by adopting the first center 3 as the answering center. The answer output unit 18D is instructed to output audio data. If the determination result signal indicates that the first center 3 should not respond, the second center 4 is adopted as the answering center and the response voice data acquired from the second center 4 is output. Instruct the answer output unit 18D as follows.

  In the present embodiment, the answering center setting unit 18C is configured to set each of the first center 3 and the second center 4 as the answering center and the semi-answering center. The setting unit 18C is an optional element for carrying out the invention. The answer-side center corresponds to the answer-side response system described in the claims, and the semi-answer-side center corresponds to the semi-answer-side response system described in the claims.

  The answer output unit 18D converts the response voice data acquired from either the first center 3 or the second center 4 into a voice signal based on an instruction from the answer side center setting unit 18C or a user operation, and outputs the voice signal to the speaker 14. The speaker 14 is made to output as sound. The response feedback unit 18E, the agent display control unit 18F, and the separate response request determination unit 18G will be described later.

(Response switching process)
Here, the flow of the response switching process which the control part 18 of the navigation apparatus 1 implements is demonstrated using the flowchart shown in FIG. The flowchart in FIG. 4 is started when, for example, the ignition switch of the host vehicle is turned on and power is supplied to the navigation device 1.

  First, in step S101, it is determined whether or not an ON signal is input from the talk SW11. If an ON signal is input from the talk SW 11, step S101 is YES, and the process proceeds to step S103. If no ON signal is input from the talk SW 11, step S101 is NO and the process returns to step S101. That is, it waits until the talk SW 11 is pressed by the user.

  In step S103, the control unit 18 connects to the first center 3, makes the response system by the first center 3 available, and also connects to the second center 4 via the mobile phone 2, The response system by the center 4 is made available. That is, each response system is in a state of waiting for reception of voice data transmitted from the navigation device 1.

  When each response system becomes available, the agent display control unit 18F causes the display device 13 to simultaneously display agents A and B corresponding to each response system, as shown in FIG. . Agent A in FIG. 6 is an image of an agent corresponding to the response system by the first center 3, and Agent B is an image of an agent corresponding to the response system by the second center 4. An agent is a character such as a fictional character or anthropomorphized animal.

  As a technique for displaying an agent corresponding to a certain response system, for example, a technique disclosed in JP-A-2006-195578 may be applied. However, Patent Document 1 does not display a plurality of agents corresponding to each of a plurality of response systems. The agent display control unit 18F according to the present embodiment generates an agent image corresponding to each response system using a known technique such as Patent Document 1, and displays the images of the plurality of agents on one screen. Thus, it is assumed that the images are combined or superimposed.

  When the process in step S103 ends, the process proceeds to step S105. Note that the voice acquisition unit 18A acquires voice data from the microphone 12 while S103 is being performed.

  In step S105, it is determined whether or not a user's voice is input from the microphone 12. That is, if no voice is input from the user within a predetermined time after the talk SW 11 is pressed in step S101, step S105 is NO and the process returns to step S101. On the other hand, when a voice is input from the user, step S105 becomes YES and the process proceeds to step S107. In step S107, the voice data acquired by the voice acquisition unit 18A is transmitted to the first center 3 and the second center 4 using the VoIP technology or the like, and the process proceeds to step S109.

  In step S109, the determination result signal sent from the first center 3 is acquired via the first communication unit 16, and the process proceeds to step S111. When the determination result signal is received, the answer-side center setting unit 18C sets the first center 3 and the second center 4 as the answer-side center or the semi-answer-side center, respectively, based on the determination result signal.

  In step S111, response voice data transmitted from the first center 3 and the second center 4 is acquired. Here, the response voice data from the semi-answer side center is temporarily saved in the memory 15 by the answer temporary saving unit 18B. In the present embodiment, the process proceeds from step S111 to step S113 when the response voice data from the answering center is acquired. However, after the response voice data is acquired from both the centers 3 and 4, the process proceeds to step S113. It may be configured.

  In step S113, the answer output unit 18D causes the answer side center setting unit 18C to output the response voice data from the answer side center to the speaker 14 by voice. In addition, the answering voice data of the answering center is output from the speaker 14, and the agent corresponding to the answering center is displayed relatively large as shown in FIGS. 6B and 6C. Display the image as the agent is speaking. When step S113 is performed, the process proceeds to step S115.

  In addition, when the text data corresponding to the content of the response voice data can be acquired together with the response voice data from each center, the text may be displayed near the agent. The visual effect for displaying an image as the agent is speaking may be designed as appropriate.

  In step S115, the separate answer request determination unit 18G determines whether an operation for requesting another answer from the user has been received. For example, when an operation for requesting another answer is accepted, the user touches the area where the agent of the semi-answer side center is displayed on the touch panel (not shown) stacked on the display device 13. May be detected.

  In addition, the user may request the controller 18 for another answer by voice-inputting the name of the agent corresponding to the semi-answer side center. However, in this case, it is necessary to register the names of agents corresponding to the response systems of both centers in the correspondence list 35A. Then, the first center side control unit 32 transmits to the navigation device 1 a control signal for outputting response voice data of the semi-answer side center by detecting that the name of the agent of the semi-answer side center is called. .

  If an operation for requesting another answer is not received from the user even after a lapse of a certain time (for example, 2 seconds) after the transition to step S115, an operation for requesting another answer is not performed from the user. It is determined that

  If an operation for requesting another answer is received from the user, step S115 becomes YES and the process moves to step S117. If the user does not request another answer, step S115 is NO and the process proceeds to step S119. In step S117, the response voice data from the semi-answer side center is read from the memory 15 and output from the speaker 14 as voice. Also, the response voice data of the semi-answer side center is output from the speaker 14 and the agent corresponding to the semi-answer side center is displayed relatively large as shown in FIGS. 6B and 6C. Display an image as if talking. When step S117 is performed, the process proceeds to step S119.

  In step S119, an answer feedback process is performed, and the process returns to the standby state in step S101. In this answer feedback processing, the contents of the judgment result signal, that is, whether or not the judgment result of the answer side judgment unit 32A is correct is transmitted to the first center 3. For example, if an operation for requesting another answer is not performed by the user in step S115, a signal indicating that the answer side determination unit 32A has made a correct determination is transmitted to the first center 3. On the other hand, when an operation for requesting another answer is received from the user in step S115, a signal indicating that the answer side determination unit 32A has made an incorrect determination is transmitted to the first center 3.

  In the present embodiment, once step S103 is performed, the connection with the first center 3 and the second center 4 is always maintained and the display device 13 is also configured as shown in FIG. Is displayed. When maintaining the connection, a signal for confirming whether the connection can be maintained at a constant period (for example, every 200 milliseconds) may be transmitted and received. Then, when an ON signal is input in a state where the connection is disconnected, step S103 may be performed again. Of course, as another configuration, the connection with the first center 3 and the second center 4 may be disconnected every time step S119 is performed.

(First center response process)
Next, the first center response process performed by the first center-side control unit 32 in response to each process in the response switching process performed by the navigation device 1 will be described with reference to the flowchart shown in FIG. The flowchart shown in FIG. 5 is started when a connection request signal is received from the navigation device 1 (step S103 in FIG. 4).

  First, in step S201, a response signal is returned to the connection request signal from the navigation device 1 to establish a connection between the navigation device 1 and the first center 3. When the connection between the navigation device 1 and the first center 3 is established, the process proceeds to step S203. In step S203, it is determined whether or not input voice data has been acquired from the navigation device 1. When the input voice data is acquired, step S203 becomes YES and the process proceeds to step S205. If the input voice data has not been acquired, step S203 is NO and the process returns to step S203. That is, the first center side control unit 32 is in a standby state until the input voice data is acquired.

  In step S205, the input voice data is output to the voice recognition unit 33, and a known voice recognition process is performed. When the result of the voice recognition process is acquired from the voice recognition unit 33, the process proceeds to step S207. In step S207, the answer-side determination unit 32A determines whether or not the center should respond to the input voice data from the result of the voice recognition process acquired from the voice recognition unit 33 and the correspondence list 35A. If it is determined that the center should respond to the input voice data, step S209 is YES and the process moves to step S211. If it is determined that the center should not respond to the input voice data, step S209 is NO and the process proceeds to step S213.

  In step S211, a determination result signal in which the center is set as the answering center is returned to the navigation device 1, and the process proceeds to step S215. In step S213, a determination result signal in which another center is set as the answering center is returned to the navigation device 1, and the process proceeds to step S215.

  In step S215, based on the result of the voice recognition process, the response data 35B stored in the first center memory 35 is referred to, and the response voice data corresponding to the content of the input voice data is sent to the voice synthesizer 36. Generate. When response voice data is acquired from the voice synthesizer 36, the process proceeds to step S217. In step S217, the response voice data is transmitted from the first center communication unit 31 to the navigation device 1, and the process proceeds to step S219.

  In step S219, the update unit 32B performs feedback processing. In this feedback processing, first, the result of whether or not the determination result of the answer side determination unit 32A is correct is acquired from the navigation device 1 (step S119 in FIG. 4). When the content that the determination of the answer side determination unit 32A is correct is acquired from the navigation device 1, the determination after the next time may be determined in the same manner as this time. Further, when the content that the determination of the answer side determination unit 32A is incorrect is acquired from the navigation device 1, the correspondence list 35A is updated so as to make a determination different from this time for the subsequent determination.

  The response process performed by the second center in response to the response switching process by the navigation device 1 is the same as the first center response process except that steps S207 to S213 in FIG. 5 are not performed. The detailed description in is omitted.

  A series of operations in response to user operations in the response control system 100 described above will be described. First, when the user presses down the talk SW 11, agents A and B corresponding to the response systems of the first center 3 and the second center 4 are displayed on the display device 13 as shown in FIG. S103). For example, when a voice signal corresponding to an utterance “Where is the convenience store?” Is input from the user, the navigation device 1 converts the voice signal into voice data, and the first center 3 and the second center. 4 (step S107).

  When the first center 3 acquires the voice data (that is, input voice data), the voice recognition process is performed (step S205), and the answer determination process is performed (step S207). In this example, the content of the input voice data is a question about the location of a convenience store. As a result of the voice recognition process, it is determined that it is a road guidance question, and it is determined that the center should respond. (Step S209 YES) And the determination result signal which set the self center (namely, 1st center 3) as an answer side center is output (step S211).

  The answering center setting unit 18C of the navigation device 1 receives the determination result signal, sets the first center 3 as the answering center, sets the second center 4 as the semi-answering center, and sets each center 3 , 4 waits for response voice data to be sent. Thereafter, when the response voice data is acquired from each of the centers 3 and 4, the response voice data from the first center 3 is output from the speaker 14, and the response temporary storage unit 18B stores the response voice data from the second center 4 in the memory 15. (Step S113).

  While the response voice data from the first center 3 is being output from the speaker 14, the agent A corresponding to the first center 3 is displayed relatively large as shown in FIG. Display as A speaks.

  Thereafter, when an operation for requesting another answer is received from the user within a predetermined time (YES in step S115), response voice data from the second center 4 is output from the speaker 14 (step S117). And the navigation apparatus 1 implements an answer feedback process, and complete | finishes a series of response switching processes with respect to this user's utterance "where is a convenience store?" Although the example in which the first center 3 is the answering center has been described above, the same processing may be performed even when the second center 4 is the answering center.

(Summary of this embodiment)
In the above configuration, which one of the first and second centers the response side determination unit 32A should respond to the input voice data based on the result of the voice recognition processing by the first center 3. The determination result is transmitted to the navigation device 1. In the navigation device 1, the answer-side center setting unit 18C sets the answer-side center and the semi-answer-side center based on the determination result, and the answer output unit causes the speaker to output response voice data from the answer-side center. .

  According to such a configuration, the answering side determination unit 32A automatically selects a center (that is, a response system) that should respond to the user's voice input according to the contents of the input voice data. Therefore, it is possible to save the trouble of selecting a response system for performing the operation and to improve the convenience for the user.

  Further, in the present embodiment, by providing the answer side determination unit 32A in the center outside the vehicle (here, the first center 3), answer feedback processing from more users can be accepted. Accordingly, the answering side determination unit 32A can perform more accurate answer determination processing.

  Further, when the response temporary storage unit 18B temporarily stores the response voice data from the semi-answer side center in the memory 15 and receives a predetermined operation input (such as touch of an agent image) from the user in step S115, the semi-answer Outputs the response voice from the side center. As a result, even if the determination result of the answering side determination unit 32A is inappropriate, it is possible to promptly hear answers from other centers without inputting the same audio content again.

  The response feedback process and the feedback process are implemented. Thereby, the user's reaction to the determination result of the answer side determination unit 32A (whether or not the content of the response from the answer side center is satisfied) can be reflected in the determination from the next time, and the accuracy of the answer determination process can be improved. Can be improved.

  Further, the agent corresponding to the center where the agent display control unit displays the agents A and B corresponding to the respective centers on the display device 13 and generates the response voice data output from the speaker 14 by the answer output unit 18D. Display as if speaking. Thus, the user can recognize at a glance which center corresponds to his / her voice input.

  In the present embodiment, the navigation device 1 is used as an interface when using a plurality of response systems. However, the present invention is not limited to this. An in-vehicle device that does not have a navigation function may be used as an interface when using a plurality of response systems.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the above-mentioned embodiment, The various modifications described below are also included in the technical scope of this invention, and also in addition to the following However, various modifications can be made without departing from the scope of the invention.

(Modification 1)
In the above-described embodiment, the answer side determination unit 32A is provided in the first center 3, but of course, the answer side determination unit 32A may be provided in the second center 4. In addition, although an information center of an automobile company is assumed as the first center 3 and an information center of a mobile phone company is assumed as the second center 4, an information center other than these may be used.

(Modification 2)
In the above-described embodiment, the communication with the second center 4 is performed via the mobile phone 2, but is not limited thereto. A configuration in which the navigation device 1 and the second center 4 communicate without using the mobile phone 2 may be employed. Furthermore, instead of the mobile phone 2, another mobile terminal having a wireless communication function (for example, a known tablet terminal) may be used. Note that the memory 15 may be an arbitrary element.

(Modification 3)
In the above-described embodiment, the first center 3 is provided with the answer side determination unit 32A that determines whether the first center 3 answers or the second center 4 answers, but the present invention is not limited to this. For example, the navigation device 1 may be provided with functions corresponding to the correspondence list 35A and the answer side determination unit 32A.

  The voice recognition result used for the answer determination process may be obtained from the first center 3 or the second center 4, or the navigation device 1 may be provided with functions corresponding to the voice recognition unit 33 and the voice recognition DB 34. . According to the configuration of the first modification, since the answering center can be determined in the navigation device 1, the configuration of the response control system can be further simplified.

(Modification 4)
In addition, although the above demonstrated the structure with which the some response system which responds to a user's input audio | voice data is equipped with the center outside a vehicle, it does not restrict to this. For example, as shown in FIG. 7, the navigation device 1 may be provided with a response system 19 corresponding to the first center 3. In this case, since communication is performed only with the second center 4, the communication cost can be suppressed as compared with the above-described embodiment.

(Modification 5)
In general, since the voice recognition process requires a relatively high processing capacity for the CPU, if the navigation apparatus 1 is provided with a function such as a voice recognition process as in the fourth modification, the navigation apparatus 1 has a high performance CPU. There is a concern that the navigation device 1 becomes expensive. Therefore, a configuration may be adopted in which the voice recognition processing is shared between the navigation device 1 and the external centers 3 and 4 using a known DSR (Distributed Speech Recognition) technique.

  For example, the navigation apparatus 1 is equipped with a function for performing only acoustic analysis on the user's voice data, and the feature amount obtained by the acoustic analysis is transmitted to the external centers 3 and 4. The external centers 3 and 4 execute voice recognition processing based on the feature amount sent from the navigation device 1, and return response voice data to the navigation device 1.

(Modification 6)
In the above-described embodiment, the answer feedback process in step S119 and the feedback process in step S219 are implemented as a learning method for improving the accuracy of the determination by the answer side determination unit 32A. Not exclusively.

  For example, the center-side control unit 32 may determine whether the user is satisfied with the answer based on whether the user repeatedly inputs the same or highly similar question. That is, as a result of the voice recognition process, the center-side control unit 32 determines that the same or highly similar question is asked again within a certain time (for example, 1 minute). B) It is determined that the user is not satisfied with the answer output from the speaker 14. In this case, the answering side determination unit 32A sets the center set as the semi-answering center at the first time as the answering center and answers the center set as the semi-answering center at the first time. What is necessary is just to correct so that it may set to a side center.

  It should be noted that whether or not the content of the input voice data is the same as or high in similarity with the previously input voice data is determined based on the words ( Evaluation may be made based on the degree of matching (including similar words of the word).

  However, just because the user is not satisfied with the response by the first response voice data does not mean that the answer center is inappropriate. The user may be requesting different answers from the same answering center.

  Therefore, the answering side determination unit 32A calculates the probability of the determination result based on the degree of coincidence between the result of the voice recognition process and the correspondence list 35A. If the certainty is equal to or greater than a predetermined threshold value, the response list 35A may be corrected to generate more appropriate response voice data without correcting the correspondence list 35A.

  The coincidence degree between the result of the speech recognition process and the correspondence list 35A is paraphrased as an evaluation of the closeness of the meaning between the correspondence list, the user's input voice and the question registered in the correspondence list.

  For example, in response to the first question “What is a nearby gas station?”, After the first center 3 presents the location of a nearby A company gas station, the question “What is the B company gas station?” Assume that it is entered.

  As the contents of the question to be answered by the first center 3, “gas station location” is registered in the correspondence list, and the first input voice and the second input voice ask the location of the gas station. Therefore, it is assumed that the certainty of the determination result that the first center 3 should respond is equal to or greater than a predetermined threshold value.

  In such a case, the response data may be corrected so as to preferentially guide the gas station of Company B from the next time while maintaining the determination result of the answering center. In this case, each center is preferably provided with response data 35B reflecting the preference of each user.

(Modification 7)
The first center 3 provided with the answer side determination unit 32A may capture the response voice data of the second center 4. The first center 3 analyzes the response voice data of the second center 4 and reflects it in the answer determination process of the first center 3 and the response data 35B, thereby improving the accuracy of the answer to the user's question. Can do.

  In particular, when the first center 3 and the second center 4 are carrying out similar services (such as searching for nearby restaurants), the user's preferences and the like can be obtained by capturing response voice data of the second center 4. Is expected to reflect.

  In this case, the navigation device 1 temporarily stores the response voice data acquired from the second center 4 in the answer temporary storage unit 18B and transmits the response voice data to the first center 3 via the first communication unit 16. In addition, it is preferable to perform the process which takes in response audio | voice data of a 2nd center independently of a 1st center response process.

(Modification 8)
Moreover, when the content of the question is relatively ambiguous as a result of the voice recognition processing, the intention of the question may be estimated based on the vehicle situation. For example, input voice data such as “I'm hungry” when driving, such as slow driving, frequent braking, sudden turn of a blinker, or when the current vehicle position is outside the user's normal range of action If it is obtained, it is estimated that a restaurant or the like is being searched for, and it is determined that route guidance is necessary.

  Then, response voice data such as “There is a restaurant ahead. Do you want to set a destination?” Is returned. Vehicle information (vehicle speed, brake, turn signal, vehicle position, etc.) for estimating the vehicle status is transmitted to the center of the automobile company through the navigation device 1. In addition, it is assumed that the determination as to whether the object is being searched or lost is performed at the center of the automobile company. Note that the navigation apparatus 1 may determine the vehicle status based on the vehicle information, and transmit the determination result to the center of the automobile company.

  According to such a configuration, even when the content of the user's voice input is ambiguous, a more appropriate response can be implemented by estimating the user's situation from the vehicle information. Become.

(Modification 9)
In the embodiment described above, the list of questions and command commands that should be handled by the first center 3 is registered in the correspondence list 35A, and the list of questions and command commands that should be handled by other centers is registered. Although it was set as the structure which is not, it is not restricted to this.

  That is, the correspondence list 35A may include a list of questions and command commands that the second center 4 should handle in addition to a list of questions and command commands that the first center 3 should handle. As a result, when the content of the user's question is registered in the correspondence list 35A that the second center 4 should be a question, the answer side determination unit 32A should not simply respond to the center 3 itself. Not only that, it is possible to determine that the second center 4 should respond to the user's question.

  Accordingly, the answering side determination unit 32A has a likelihood that the center will respond for each center based on the degree of coincidence between the result of the voice recognition processing and the content of the correspondence list 35A for each center ( (Likelihood) may be calculated, and the higher likelihood may be set as the answer center. The answering side determination unit 32A assumes that the likelihood increases as the degree of coincidence between the result of the speech recognition processing and the content of the correspondence list 35A for each center increases.

(Modification 10)
In the modification 9, the correspondence list 35A is registered with a list of questions and command commands that should be handled by each center. However, the present invention is not limited to this. For example, in the correspondence list 35A, words or thin words that are strongly related to the service performed by the center are defined for each center, and the answering side determination unit 32A determines the words included in the result of the speech recognition process. You may judge the answer side center by evaluating the relationship with each center.

  For example, when words such as “road”, “destination”, “road”, and “traffic jam” are included, the first center 3 is set as the answering center, and “phone” “(person noun)” “ If a word such as “schedule” is included, the second center 4 may be set as the answering center. When evaluating the degree of coincidence between the question to be handled by the center and the input voice data, weighting for each word may be applied.

(Modification 11)
Moreover, although the above example demonstrated the structure which can use two response systems (namely, center) as a response system which a user can utilize, it is not restricted to this. There may be three or more centers available to the user. That is, there may be a plurality of centers corresponding to the second response system with respect to one center corresponding to the first response system recited in the claims.

  In this case, in the correspondence list 35A, the contents of the input voice that each center should respond to are registered for each center. Then, the answering side determination unit 32A, based on the result of the voice recognition process and the correspondence list, which response system among the plurality of response systems available to the user responds to the input voice data. What is necessary is just to determine.

  For example, the answering side determination unit 32A calculates the likelihood for each center based on the degree of coincidence between the result of the voice recognition process and the content of the correspondence list 35A for each center. Then, it is only necessary to preferentially output the voice response data from the center with the highest likelihood.

DESCRIPTION OF SYMBOLS 100 Response control system, 1 Navigation apparatus (vehicle equipment), 12 Microphone, 14 Speaker, 18 Control part, 18A Voice acquisition part, 18B Answer temporary storage part, 18C Answer side center setting part, 18D Answer output part, 18E Answer feedback part , 18F agent display control unit, 18G separate response request determination unit, 2 mobile phone, 3 first center (first response system), 32 first center side control unit, 32A answer side determination unit, 32B update unit, 33 voice Recognizing unit, 34 voice recognition database, 35 first center side memory, 35A correspondence list, 35B response data, 4... Second center (second response system)

Claims (8)

An audio acquisition unit (18A) that acquires input audio by the user as input audio data via the microphone (12);
A voice recognition process is performed on the input voice data acquired by the voice acquisition unit, and first response voice data serving as a response to the input voice data is generated based on a result of the voice recognition process. Response system (3)
Voice recognition processing is performed on the input voice data acquired by the voice acquisition unit, and based on the result of the voice recognition processing, second response voice data that is voice data serving as a response to the input voice data is obtained. A second response system (4) to generate;
A correspondence list storage unit (35) for storing a correspondence list (35A) describing the contents of the input voice data to which the first response system should respond;
An answerer that determines which one of the first and second response systems should respond to the input voice data based on the result of the voice recognition processing by the first response system. A system determination unit (32A);
An answer output unit for causing the speaker (14) to output the answer voice data generated by the answer side answer system, which is the answer system determined to respond to the input voice data in the answer side system judgment unit ( and 18D), with a,
The answering system determination unit
When the result of the voice recognition processing is associated with the correspondence list, the first response system determines that the first response system should respond to the input voice data,
If the result of the speech recognition process is not associated with the correspondence list, the response control system determines that the second response system should respond to the input speech data . In claim 1 ,
A user who requests to output the response voice data generated by the semi-answer side response system, which is a response system that is not determined to respond to the input voice data, by the answering system determination unit to the speaker Another response request determination unit (18G) for determining whether or not the operation has been accepted,
When the separate response request determination unit determines that the user operation has been received by the semi-response side response system, the response output unit causes the speaker to output response audio data of the semi-response side response system A response control system characterized by In claim 2 ,
A temporary storage unit (18B) for temporarily storing the response voice data acquired from the semi-response side response system until it is determined that the user operation is not accepted by the separate response request determination unit;
When it is determined that the user operation is received by the separate response request determination unit, the response audio data of the semi-response side response system stored in the temporary storage unit is output by voice to the speaker. Response control system. In claim 3 ,
When it is determined that the user operation has been received by the separate response request determination unit, the response system that is the semi-response side response system this time is the next time or later with respect to the result of the speech recognition processing on the input speech data. A response control system comprising an update unit (32B) for updating the contents of the correspondence list so as to be the answer side response system. In any one of Claim 2 to 4 ,
When the input voice data determined to be the same or highly similar content from the result of the voice recognition processing by the first response system is input again within a predetermined time,
The response system that was the quasi-answer side response system last time is the answer side response system,
The response control system updates the contents of the correspondence list so that the response system that was the semi-answer side response system last time is the answer side response system from the next time. In any one of Claim 1 to 5 ,
The response control system includes a display device (13),
An agent display control unit (18F) that simultaneously displays on the display device an agent image corresponding to each of the first and second response systems;
The agent display control unit displays the agent image corresponding to the response system that has generated the response voice data output to the speaker by the answer output unit so that the agent image is speaking. system. In any one of Claim 1 to 6 ,
The response control system includes an on-vehicle device (1) mounted on a vehicle, a first center (3) provided outside the vehicle to perform wireless communication with the on-vehicle device, and an on-vehicle device provided outside the vehicle. A second center (4) for performing wireless communication with the vessel,
The vehicle-mounted device includes the voice acquisition unit and the answer output unit,
The first center is responsible for the function as the first response system and includes the answering system determination unit.
The response control system, wherein the second center has a function as the second response system. The center provided with the function of the said 1st center of Claim 7 .

Images