WO2019106811A1 - Query response system and query response method - Google Patents
Query response system and query response method Download PDFInfo
- Publication number
- WO2019106811A1 WO2019106811A1 PCT/JP2017/043192 JP2017043192W WO2019106811A1 WO 2019106811 A1 WO2019106811 A1 WO 2019106811A1 JP 2017043192 W JP2017043192 W JP 2017043192W WO 2019106811 A1 WO2019106811 A1 WO 2019106811A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- query response
- query
- user
- computer
- response device
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/44—Program or device authentication
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
Definitions
- the present invention relates to a query response system and a query response method, and more particularly to a query response system and a query response method for responding to queries from a user in a secure manner.
- a query response system that recognizes a voice (utterance) uttered by a user and returns a voice response corresponding to the utterance.
- Amazon (R) Echo analyzes speech from a user as an AI (Artificial Intelligence) speaker, and returns an appropriate voice response.
- AI Artificial Intelligence
- Such an AI speaker learns how to respond to the user's speech and returns an appropriate response as the use is repeated.
- An AI speaker can work in conjunction with an application that provides a particular service.
- the AI speaker selects an application based on a specific keyword in the utterance from the user (that is, implementing a process of returning an appropriate response to the user's utterance).
- the selected application executes a process of returning a response according to the utterance, and the AI speaker outputs the response in voice.
- the AI speaker When cooperating with the application described above, the AI speaker needs to communicate with the computer device on which the application is implemented. Usually, when performing such communication, wireless communication such as WiFi is used. In the prior art, when performing wireless communication, the user is required to utter an authentication code. For example, when the user utters “Turn on TV", the AI speaker responds such as "Tell me PIN code” etc., and the user utters the PIN code, whereby the authentication is performed.
- Authentication via the authentication code requires the user to utter the authentication code every time, which is inconvenient. In addition, since the authentication code is issued by voice, the authentication code may be leaked to a third party.
- the biometric authentication of the uttered voice is a voice biometric authentication method that determines whether the user is a valid person based on the voice. Voice biometric authentication is less accurate and less reliable than fingerprints and irises.
- Two-pass authentication is an authentication method employed in online banking and the like.
- a transaction is established by creating transaction data at a first computer device of two interconnected computer devices and approving at a second computer device.
- Such a scheme can ensure security, but requires operation by a user via a computer device and lacks convenience.
- the AI speaker as long as information is exchanged by voice, it is required to simplify human operations. This is useful, for example, when driving a car, when working in a kitchen, or when it is required to simplify the operation through a computer device.
- Patent Document 1 discloses a technology for executing a bank transaction by voice according to voice from a customer.
- the user is identified by voice input of information such as an account number from a customer and a service code. Further, in the technology according to Patent Document 1, the identification is performed by comparing the voiceprint of the customer registered in advance with the voiceprint of the voice data uttered by the customer himself.
- Patent Document 1 uses (1) authentication through an authentication code by speech and (2) biometric authentication of speech. This technology does not solve the above-mentioned problems at all.
- the present invention has been made in view of such problems, and an object thereof is to provide a query response system that returns a response to a query from a user in a security secured manner.
- a query response system is a computer system comprising a server computer and a query response device, wherein the query response device receives an input and a query based on the input.
- a telegram is generated and transmitted to the server computer, the server computer transmits a connection confirmation request telegram to the client computer, receives a connection confirmation reply telegram including the first identifier from the client computer, and the first identifier
- a connection between the client computer and the query response device is established by identifying the query response device and determining whether the first identifier and the stored second identifier match. And the first identifier matches the second identifier.
- the query response device In response to determining that generates a response message corresponding to the query message sent to the query response device, wherein the query response device, and outputs the response message.
- the query response system it is possible to secure security when returning a response to a query from a user.
- FIG. 1 is a diagram showing an example of the overall configuration of a computer system that constitutes a query response system according to an embodiment of the present invention.
- the query response system according to the present invention is described as an example in which a response is returned in the form of voice (that is, voice output) in response to an utterance from the user Do.
- the query response system comprises a server computer 1, a query response device 2, and a client computer 3.
- the server computer 1 and the query response device 2 and the client computer 3 are connected via the network 4 (a public network such as the Internet), and the connection method may be either wired or wireless.
- the query response device 2 and the client computer 3 are connected by a connection method in accordance with a short distance wireless communication standard such as Bluetooth (registered trademark).
- the server computer 1 is a server computer device that provides application services to users.
- the server computer 1 executes an application program corresponding to the application service to be implemented in response to a query (speech) from the user through the query response device 2, and generates a response message corresponding to the speech.
- the response message is sent to the query response device 2.
- the server computer 1 may be implemented by a single device or may be implemented by a plurality of devices. Also, the server computer 1 may be implemented by a plurality of devices each implementing a different service. In this case, the query response device 2 recognizes an utterance from the user (for example, based on a specific keyword in the utterance), and selects a server computer 1 implementing an application service corresponding to the utterance.
- the query response device 2 is a speaker device with a learning function that responds with speech in response to speech from the user.
- the query response device 2 connects with the client computer 3 (via Bluetooth connection) by detecting that the client computer 3 is within a certain range. Then, when the user utters, the voice is recognized, a query message corresponding to the utterance is transmitted to the server computer 1, a response message corresponding to the query message is received, converted to voice, and voice output is performed. That is, the query response device 2 selects the server computer 1 which returns a response corresponding to the speech from the user, and functions as an intermediary device which outputs the response from the server computer 1 as voice.
- the query response device 2 may also be implemented by a single device or by multiple devices.
- the client computer 3 is a computer device used to authenticate a user.
- the client computer 3 transmits connection information to the server computer 1.
- the server computer 1 confirms the connection between the client computer 3 and the query response device 2 to authenticate the user.
- the client computer 3 is assumed to be a mobile phone including a smartphone, but is not limited to such a device, and it is a computer device having a short distance wireless communication function (in particular, a Bluetooth connection function)
- the equipment is not limited.
- it may be a wearable device (smart watch and smart glass) worn by a user on clothes or the like.
- the server computer 1 includes a control unit 11, a memory 12, a storage unit 13, and a communication unit 14. These components are mutually connected via a bus.
- the controller 11 may be implemented by a processor or the like, and may include a central processing unit (CPU), a graphic processing unit (GPU), and / or one or more processor cores.
- the server computer 1 implements a DBMS (Database Management System), and the control device 11 executes programs implementing an OS (Operating System), application programs implementing an application service, and programs implementing a DBMS.
- the memory 12 functions as a volatile storage area, and may include, for example, random access memory (RAM), dynamic RAM, or cache memory.
- the control device 11 reads the program described above into the memory 12 and executes the program.
- the storage device 13 is a storage device built in the server computer 1 and may include, for example, a hard disk drive, a solid state drive, an optical disk, and a flash drive.
- the storage device 13 stores the above-described program and a database table 13a such as a supply destination data table described later.
- the communication device 14 is implemented by, for example, a network interface card (for example, a LAN card) and transmits and receives data between the query response device 2 and the client computer 3 via the network 4.
- the query response device 2 includes a control unit 21, a memory 22, a storage unit 23, a communication unit 24, an input driver 25, a microphone 26, an output driver 27, and a speaker 28, whose components are mutually coupled via a bus. It is done.
- the control device 21 is implemented by a CPU or the like as in the server computer 1.
- the control device 21 implements a program for implementing the OS, a program for implementing a codec that encodes audio signals into digital data and decodes the digital data into audio signals (codec program), and a program for implementing a learning function (learning program) Are read into the memory 22 and executed.
- the learning function may be implemented using known techniques, such as neural network models.
- the storage device 23 is a built-in storage device including a hard disk drive and the like, and stores learning data accumulated according to a neural network model or the like.
- the communication device 24 includes a wireless LAN card, a Bluetooth adapter, and the like. The communication device 24 transmits and receives data to and from the server computer 1 via the network 4, and transmits and receives data to and from the client computer 3 via a Bluetooth connection.
- the microphone (input device) 26 receives an utterance by the user via the input driver 25.
- the received speech is encoded by the control device 21 (codec) and converted into a query message.
- the speaker (output device) 28 outputs the voice signal converted from the response message received from the server computer 1 as voice.
- the response message is decoded by the control device 21 (codec) and transmitted to the speaker 28 via the output driver 27.
- the client computer 3 has a communication device and a control device (not shown).
- the communication device transmits / receives data to / from the server computer 1 via the network 4 and transmits / receives data to / from the client computer 3 via the Bluetooth connection.
- the control device executes a client application program corresponding to the OS and an application program executed by the server computer 1.
- the client application program is downloaded from the server computer 1 in advance.
- the user data table 300 is a data table used for authenticating the user with respect to the inquiry about the balance among the database tables 13 a stored in the storage device 13 of the server computer 1.
- the user data table 300 includes data items “user ID”, “client computer notification ID”, “used device name”, and “valid flag”. Each data record is generated for each user.
- the user ID a unique identifier is set for each user in an application service (hereinafter, "application service") implemented by the server computer 1.
- the client computer notification ID is an ID by which the server computer 1 identifies a transmission destination for transmitting a connection confirmation request message described later to the client computer.
- the name (for example, model number etc.) of the query response device 2 which a target user uses is set as a utilization device name. That is, the target user and the query response device 2 to be used are stored in association with each other.
- the validity flag is set to a value indicating whether the service responding to the inquiry about the balance is valid, and is, for example, "1: valid" or "2: invalid".
- the data record of the target user in the user data table 300 is generated (added), for example, when the target user registers in the application service.
- a unique user ID is assigned and logs in with a predetermined authentication (such as a password).
- the user ID assigned at this time is set in the target data record of the user data table 300.
- connection confirmation request message for performing connection confirmation is transmitted, for example, using a push notification service of the OS mounted on the server computer 1.
- the push notification service uses, for example, “APN (Apple Push Notification Service)” when the client computer 3 implements iOS (registered trademark) and “GCM (Google Google) when Android (registered trademark) is installed. Use Cloud Messaging). Therefore, the ID assigned when the user registers in the push notification service is set in the target data record of the user data table 300 (client computer notification ID).
- the above-described push notification service is merely an example, and is not limited to the above example. Any method by which the server computer 1 can transmit a connection confirmation request message to the client computer 3 may be used.
- pairing needs to be performed between two devices (devices compatible with Bluetooth connection).
- the two devices are within the communicable range, the presence of the other device is detected mutually and the device name of the object is recognized.
- pairing is completed, and pairing information is stored in both devices (such as the device name of the other device).
- the client computer 3 and the query response device 2 are paired in advance, and the device name of the query response device 2 is set to the use device name of the user data table 300 (that is, the client computer 3 and the query response device
- the pairing information with 2 is registered in the user data table 300).
- Bluetooth connection two devices emit radio waves in the 2.4 GHz band defined by IEEE (Institute of Electrical and Electronics Engineers) 802.15.1 and communicate in a range of several meters to several tens of meters. That is, communication between the client computer 3 and the query response device 2 is effective only when both are in the range of several meters to several tens of meters.
- IEEE Institute of Electrical and Electronics Engineers
- the account data table 400 is a data table storing account information for each user among the database tables 13a stored in the storage device 13 of the server computer 1 necessary for implementing the above-described service.
- the account data table 400 includes data items "user ID”, “account number”, “financial institution code”, “branch code”, “account type”, “nominee”, “balance”, And “limit amount”.
- Each data record is generated for each user who uses the Internet banking service described above. That is, each user who uses the application service according to the present embodiment is generated as a pair with the data record of the user data table 300 described in FIG.
- the limit amount is set to a predetermined value when limiting the response to the inquiry about the balance described later.
- the user A utters the query response system to inquire about its balance, and a response relating to the balance is audibly returned. It is assumed that the account information of the user A is registered in the account data table 400.
- communication between the client computer 3 and the query response device 2 is performed by Bluetooth connection, pairing is already performed between the two, user ID of the user A, use device name, and client computer notification It is assumed that the ID is registered in the user data table 300. In such a state, when the client computer 3 enters a range in which a Bluetooth connection can be made with the voice compatible device, the connection state is automatically established.
- the process executed by the server computer 1 in the process described in FIG. 5 is implemented by the control device 11 of the server computer 1 executing an application program.
- the processing executed by the query response device 2 is implemented by the control device 21 of the query response device 2 executing a codec program, a learning program, and the like.
- the processing executed by the client computer 3 is implemented by the control device of the client computer 3 executing the client application program.
- step S501 the communication device 24 of the query response device 2 detects the presence of the client computer 3 (or the client computer 3 detects the presence of the query response device 2).
- the process of this detection is executed by the query response device 2 detecting the radio wave emitted by the client computer 3.
- the query response device 2 detects the presence of the client computer 3 in step S501, a connection between the two is established, and the client computer 3 can recognize the device name of the query response device 2.
- the user A utters, for example, "Please tell me the balance of the account” to the query response device 2 (step S502).
- the microphone 26 of the query response device 2 receives the speech in step S502 and converts it into an audio signal
- the microphone 26 transmits the speech signal to the control device 21.
- the control device 21 converts the voice signal into digital data, and analyzes the keyword included in the digital data (that is, analyzes the speech of the user A) (step S503).
- the analysis of keywords is performed by extracting keywords in the speech converted to digital data.
- the keyword extraction is performed using known feature word extraction algorithms such as, for example, TF-IDF (Term Frequency-Inverse Document Frequency) and SVM (Support Vector Machine).
- an application service corresponding to the analyzed keyword is specified according to learning data such as a neural network (step S504).
- learning data such as a neural network
- data in which a keyword such as “balance” and the Internet banking service are associated is stored in the learning data.
- the control device 21 specifies the server computer 1 providing the internet banking service.
- the control device 21 generates a query message, and the communication device 24 transmits it to the server computer 1 identified in step S504 (step S505).
- the query message includes at least information indicating the type of request by the user A of the client computer 3 acquired in step S501.
- the type of request is stored in the learning data in association with a predetermined keyword. For example, the keyword “balance” is associated with an inquiry request regarding the balance.
- the control device 11 acquires a target data record from the user data table 300 (step S506). Then, the control device 11 generates a connection confirmation request message for confirming whether a connection is established between the client computer 3 and the query response device 2. Then, the communication device 14 transmits a connection confirmation request message with the client computer notification ID included in the user data record acquired in step S506 as the destination (that is, the client computer 3) (step S507).
- the connection confirmation request message is transmitted using the above-described push notification service or the like.
- connection confirmation response message includes at least the device name of the query response device 2 acquired in step S501.
- the client application program needs to be executed on the client computer 3.
- the OS of the client computer 3 may start the client application program.
- the target application can be identified by referring to the port number of the connection confirmation request message (the port number is associated with the port number assigned to the application program of the server computer 1) ).
- step S506 the control device 11 determines that the device name (first identifier) included in the connection confirmation response message is in step S506. It is determined whether or not the used device name (second identifier) included in the user data table acquired in step S2 matches (step S509). In this process, if the used device name included in the user data record matches the device name included in the connection confirmation response message, connection is established between the client computer 3 and the query response device 2 corresponding to the target device name. Can be confirmed. That is, it can be confirmed that the speech from the user A is a user having the client computer 3 connected to the target query response device 2, and consequently, the identity of the user A can be confirmed.
- the server computer 1 may notify the target user to urge the connection. Specifically, the communication device 14 of the server computer 1 sends a predetermined message to the query response device 2 and / or the client computer 3. The query response device 2 and / or the client computer 3 outputs the transmitted message in a predetermined format such as audio or visual display.
- the control device 11 executes processing in accordance with the type of request included in the query message transmitted in step S505.
- the type of request is an inquiry request regarding balance
- the control device 11 acquires the target data record from the account data table 400 based on the user ID included in the data record acquired in step S506. (Step S510).
- the control device 11 When acquiring the data record, the control device 11 generates a query response message based on the acquired data record, and the communication device 14 transmits it to the query response device 2 (step S511).
- the query response message includes at least the balance included in the data record acquired in step S510.
- the control device 21 converts the query response message into an audio signal (step S512).
- the converted audio signal is output as audio through the speaker 28 via the output driver 27 (step S513).
- the voice output is "The balance is 1,029,389 yen.”
- step S510 If the device names do not match in step S509, that is, the identity of the user A can not be confirmed, and a message notifying that effect is notified, the process of step S510 is performed after a predetermined time has passed since the notification. An error response is returned without being executed. In this case, in step S511, an error response message indicating that the identity verification could not be made is generated, and that effect is voice-outputted through the processing in steps S512 and S513. For example, the voice output is "I can not answer because I could not confirm my identity.”
- the error response described above may also be applied when the balance included in the data record of the account data table 400 acquired in step S510 exceeds the limit amount. This makes it possible to limit the response to the inquiry when the user has a balance exceeding a certain amount.
- the server computer 1 notifies the client computer 3 that there is an inquiry about the balance from the user A at any time after the process of step S 505 (that is, any after the user A has made an inquiry about the balance). It may be sent (eg, in the form of an email). This process allows the user A to confirm that an attempt has been made by a third party to make an inquiry regarding a balance inquiry.
- a notification indicating that the user A has responded to the inquiry about the balance is transmitted to the client computer 3 at any time after the process of step S 509 (that is, any one after the user A's identity verification is successful). It is also good. Furthermore, if identity verification fails in the process of step S509, a notification may be sent to the client computer 3 indicating that identity verification could not be performed in response to the inquiry regarding the balance from the user A, in any of subsequent processes. . With any of the processes, the user A can confirm that the identification of the third party's inquiry about the balance is successful or unsuccessful.
- the user A can invalidate the function of the inquiry about the balance described above through the operation of the client computer 3.
- the valid flag included in the target data record of the user data table 300 through the operation of the client computer 3 is updated to “2: invalid”.
- the control device 11 can return an error response to the query response device 2 by setting the value set in the valid flag in the processing after step S508 described above.
- the query response system according to the first embodiment of the present invention has been described.
- user authentication can be performed without requiring an utterance or input of a PIN code or the like, and security can be secured in exchange of secret information such as a balance.
- the response is returned via the Bluetooth connection between the client computer 3 and the query response device 2, at least the client computer 3 is not returned from the query response device 2 as a response.
- the Bluetooth connection between the client computer 3 and the query response device 2 is a condition, even if one of the devices is stolen etc., it is possible to prevent eavesdropping by a third party unless both devices are aligned. can do.
- security can be further enhanced by implementing the client computer 3 with a computer device with a lock function.
- the client computer 3 is implemented by a wearable device
- the query response system according to the present invention can be used without requiring an operation by the user.
- the wearable device when the user wears the device, the lock is released (by the internal sensor detecting the contact state between the user and the device), the user performs an operation to release the lock. It is because it is not necessary.
- the connection between the client computer 3 and the query response device 2 is described as being established according to the Bluetooth communication standard, but the present invention is not limited to such a format. Any communication method may be adopted which performs a procedure equivalent to pairing similar to Bluetooth. In this case, at the time of the first inter-device connection, both sides hold information identifying the other device, and the subsequent communication is that the communication is encrypted by the secret key when both devices are within the predetermined range. Become. The same applies to the second embodiment described below.
- a communication method of near field wireless connection is used. It is desirable to do. For example, near-field wireless communication using ZigBee or infrared light is used.
- FIG. 6 shows that there are a plurality of identification targets identified based on the keyword in the utterance from the user, and each identification target is identified based on the corresponding keyword. That is, based on one or more keywords included in the utterance, a plurality of different identification targets such as the application service and the user are determined.
- the identification target includes the application service provided to the user, that is, the application service is identified based on the first keyword in the speech from the user (in the present embodiment, the Internet banking service). Further, the identification target includes the type of the request by the user, and the type of the request is identified based on the second keyword in the utterance from the user (in the present embodiment, the inquiry about the balance). Furthermore, the identification target includes the inquired user, that is, the inquired user is identified (authenticated) based on the third keyword in the speech from the user.
- the keyword “ ⁇ bank” in this utterance is the first keyword, and is associated with the Internet banking service in the present embodiment. That is, the query response device 2 selects the server computer 1 that provides the target internet banking service, in response to the determination that the “soo bank” is included in the speech from the user.
- the association (first correspondence relationship) between the first keyword and the application service is stored in learning data.
- the keyword "balance" in the utterance is the second keyword, and is associated with the processing request for the balance related inquiry in the present embodiment. That is, the query response device 2 generates a query message including information indicating that it is an inquiry request regarding the balance, in response to the determination that the “speech” is included in the utterance from the user.
- the association (second correspondence relationship) between the second keyword and the processing request is stored in learning data.
- the keyword "Smith” in the utterance is the third keyword, and is associated with the inquired user in the present embodiment. This keyword is included in the query message sent from the query response device 2 and the server computer 1 authenticates the user based on this keyword.
- the second embodiment has a portion in which the processing overlaps with the first embodiment, so only the different portion will be described.
- the processes of the steps shown in Table 1 below correspond to the processes of the steps described in FIG.
- step S702 the user A utters, for example, "Smith, tell me the account balance of the bank OO.”
- the control device 21 of the query response device 2 specifies the corresponding application service from the keyword analyzed in step S703 according to the first correspondence stored in the learning data. That is, in response to having determined that the first keyword " ⁇ bank" is included in the speech from the user A, the server computer 1 providing the internet banking service is specified.
- step S705 the control device 21 determines the type of request from the user A according to the second correspondence relationship stored in the learning data, from the keyword analyzed in step S703. Then, a query request including information indicating the type of the request and the keyword analyzed in step S703 is generated. The query message is transmitted to the server computer 1 by the communication device 24.
- step S706 the control device 11 of the server computer 1 acquires the target data record from the user data table 300.
- the data structure of the user data table 800 in the second embodiment will be described. An example will be described.
- the user data table 800 includes an authentication keyword in addition to the data items included in the user data table described in FIG. 3.
- the authentication keyword is set by the user to an arbitrary keyword value.
- step S709 in addition to the device name included in the data record acquired in step S706, it is determined whether the keyword set in the authentication keyword is included in the query message transmitted in step S705. . That is, it is determined whether the speech from the user A includes a preset keyword.
- the user A can be authenticated by setting a predetermined keyword in advance and including the keyword in the utterance.
- the determination of the third keyword match is executed in the server computer 1, but the present invention is not limited to such a format.
- the query response device 2 may determine whether the authentication keyword set in the dialog from the user is included.
- the correspondence between the device identifier such as the BD address obtained from the client computer 3 and the keyword is stored in the learning data of the query response device 2, and the keyword included in the dialog from the user is determined according to the learning data.
- the query response system according to the second embodiment of the present invention has been described.
- secret information can be exchanged in a system in which security is further ensured.
- user authentication can be performed without requiring an operation by the user.
- the response to the speech from the user may be limited by determining whether the client computer 3 is within a predetermined position range.
- the specific example of the process in this case is shown below.
- the connection confirmation response message transmitted from the client computer 3 in step S 507 (in the second embodiment, step S 707) described in FIG. 5 includes the position information of the client computer 3.
- the client computer 3 needs to have a GPS receiver, and the position information is generated based on the GPS signal received from the satellite by the GPS receiver.
- position information indicating a predetermined range is stored in the above-described user data table 300 (in the second embodiment, the user data table 800). Then, the server computer 1 determines whether the position information included in the connection confirmation response message transmitted from the client computer 3 is within the range of the position information included in the target data record of the user data table 300.
- the response to the utterance is not output in voice only within the range designated by the user. This is because, for example, even if both the query response device 2 and the client computer 3 are stolen etc., it is restricted to use only at the designated position, so that a third party eavesdrops on the balance. It can prevent the act.
- the query response device 2 returns a response in the form of voice in response to a query from a user uttered (that is, voice).
- the query from the user may include input from the user on any medium other than voice.
- the returned response may also include the output of the response on any medium other than voice.
- the query response device 2 will be implemented as a multimedia response device that accepts input on media other than voice and outputs a response.
- the input from the user may be, for example, pressing of a physical button provided on the query response device 2, touch operation on a touch screen (touch and slide, etc.), tactile operation on the entire query response device 2, and the query response device 2. It may include input on any medium, such as visual input of
- the tactile operation on the entire query response device 2 may include, for example, an operation of tilting and / or moving the query response device 2 in a predetermined direction. Both operations are determined by the query response device 2 detecting the angular velocity and acceleration of the tilt and movement of the device. With regard to any of the tilt and movement of the device, the combination of a predetermined direction, a predetermined distance, and / or a predetermined speed may be associated with the content of the query, and the query response device 2 may determine the query. For example, the operation in which the user tilts the query response device 2 45 degrees to the right corresponds to a query for inquiring the balance of the account.
- the visual input to the query response device 2 is, for example, an operation in which the user or a part of the user (such as a face) approaches a predetermined range of the query response device 2 and an operation in which the predetermined color and brightness are held over the query response device 2 Etc. may be included. Both actions are determined by detecting the proximity of the query response device user to the device, light, and the like. For any of these operations, the query response device 2 may determine the query, with predetermined proximity, color and / or intensity combinations associated with the content of the query. For example, an operation in which the user wears black clothes and approaches within 50 cm of the query response device 2 corresponds to a query for querying the balance of the account.
- the query response device 2 includes a touch screen and a camera as input devices, and includes a proximity sensor, an acceleration sensor, an angular velocity sensor, an optical sensor and the like as detection elements.
- the output of the response may include, for example, an output on any medium, such as a visual display on the display of the query response device 2.
- the query response device 2 includes a display or the like as an output device.
- the query input described above may be accepted by the client device 3 instead of the query response device 2.
- a value (angular velocity or the like) corresponding to the operation may be transmitted to the query response device 2, and the query response device 2 may determine the query content corresponding to the operation. Also, the client computer 3 may determine the contents of the query.
- the response output may also be output by the client device 3 instead of the query response device 2.
- the response may be transmitted to the client computer 3 and displayed on a display or the like.
- pairing information of a single query response device 2 and a single client computer 3 is registered in the server computer 1
- the pairing information associates a plurality of pairs. May be registered.
- N client computers 3 may be associated with M query response devices 2 and registered in the user data table 300 or 800.
- each of two client computers 3 is paired with the query response device 2
- client computer notification IDs of both of the two client computers 3 are registered in the user data table 300 or 800 in association with the user ID and the use device name.
- a connection confirmation request message is sent from the server computer 1 to both of the two client computers 3, and it is confirmed whether both are connected to the query response device 2.
- the present invention is not limited to such configurations.
- the response to the utterance is output by the server computer 1 and the query response device 2 cooperating with each other, the above process may be implemented by a single computer device on which a speaker is implemented. Also, some of the processing performed by the server computer 1 described above may be performed by the query response device 2 without departing from the basic concept of the present invention, and vice versa .
- the query response device described above (that is, the device on which the speaker and the microphone are mounted) has been described as being implemented as an AI speaker, the query response device does not necessarily have to have a learning function.
- a query response device is implemented without having a learning function, a correspondence relationship in which a predetermined keyword is associated with request processing and the like is stored, and processing is executed according to the correspondence relationship.
- the data structures such as user data table 300 described above, are exemplary only, and may include other necessary data items, and may be alternative without departing from the basic concept of the present invention. Data tables may be used.
- the order of the processes described in the above embodiment need not necessarily be performed in the order described, and may be performed in an arbitrary order.
- additional steps may be added newly or the described steps may be omitted without departing from the basic concept of the present invention.
- server computer query response device 3 client computer 4 network 300 user data table 400 account data table 800 user data table A user
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Transfer Between Computers (AREA)
Abstract
Provided is a query response system. This query response device receives an input, and generates a query telegram on the basis of the input to transmit the query telegram to a server computer. The server computer transmits a connection confirmation request telegram to a client computer, receives a connection confirmation response telegram including a first identifier from the client computer, identifies the query response device by using the first identifier, determines that a connection is established between the client computer and the query response device by determining whether the first identifier matches a stored second identifier, and generates a response telegram corresponding to the query telegram to transmit the response telegram to the query response device in response to the determination that the first identifier matches the second identifier. The query response device outputs the response telegram.
Description
本発明は、クエリ応答システムおよびクエリ応答方法に関し、特に、セキュリティが確保された方式で、ユーザからクエリに対して応答するクエリ応答システムおよびクエリ応答方法に関する。
The present invention relates to a query response system and a query response method, and more particularly to a query response system and a query response method for responding to queries from a user in a secure manner.
ユーザが発した音声(発話)を認識し、その発話に対応する音声応答を返却するクエリ応答システムが知られている。例えば、Amazon(登録商標)Echoは、AI(Artificial Intelligence)スピーカとしてユーザからの発話を分析し、適切な音声応答を返却する。このようなAIスピーカは、ユーザからの発話に対してどのように応答するかを学習し、使用を重ねるにつれて適切な応答を返却する。
There is known a query response system that recognizes a voice (utterance) uttered by a user and returns a voice response corresponding to the utterance. For example, Amazon (R) Echo analyzes speech from a user as an AI (Artificial Intelligence) speaker, and returns an appropriate voice response. Such an AI speaker learns how to respond to the user's speech and returns an appropriate response as the use is repeated.
AIスピーカは、特定のサービスを提供するアプリケーションと連携することができる。AIスピーカは、ユーザからの発話における特定のキーワードに基づいて(つまり、ユーザの発話に対して適切な応答を返却する処理を実装する)アプリケーションを選択する。選択されたアプリケーションが発話に応じた応答を返却する処理を実行し、AIスピーカがその応答を音声出力する。
An AI speaker can work in conjunction with an application that provides a particular service. The AI speaker selects an application based on a specific keyword in the utterance from the user (that is, implementing a process of returning an appropriate response to the user's utterance). The selected application executes a process of returning a response according to the utterance, and the AI speaker outputs the response in voice.
上述したアプリケーションと連携する場合、AIスピーカは、当該アプリケーションを実装したコンピュータデバイスと通信する必要がある。通常、このような通信を行う場合、WiFiなどの無線通信が利用される。従来技術では、無線通信を行う際には、ユーザは認証コードを発話することが要求される。例えば、ユーザが「テレビをつけて」と発話すると、AIスピーカが「PINコードを教えてください」などと応答し、ユーザがPINコードを発話することによって認証が行われる。
When cooperating with the application described above, the AI speaker needs to communicate with the computer device on which the application is implemented. Usually, when performing such communication, wireless communication such as WiFi is used. In the prior art, when performing wireless communication, the user is required to utter an authentication code. For example, when the user utters "Turn on TV", the AI speaker responds such as "Tell me PIN code" etc., and the user utters the PIN code, whereby the authentication is performed.
上述したAIスピーカで口座情報の残高などを含む秘密情報を問い合わせることも考えられる。このようなケースでは、秘密情報をやり取りすることになるので、セキュリティが確保された方式、つまり所定の認証を介して発話応答が行われる必要がある。AIスピーカにおける認証方式として、(1)発話による認証コードを介した認証、(2)発話音声の生体認証、および(3)携帯電話(スマートフォンなど)のコンピュータデバイスと連携した二経路認証などが考えられる。
It is also conceivable to inquire about secret information including the balance of account information and the like from the above-mentioned AI speaker. In such a case, since secret information is to be exchanged, it is necessary to make a speech response via a method in which security is ensured, that is, predetermined authentication. As authentication methods in AI speakers, (1) authentication via an authentication code by speech, (2) biometric authentication of speech, and (3) two-path authentication in cooperation with a computer device of a mobile phone (such as a smartphone) are considered. Be
認証コードを介した認証は、ユーザが毎回認証コードを発話する必要が生じ、利便性に欠ける。また、認証コードを音声で発することになるので、認証コードが第三者に漏洩するおそれもある。発話音声の生体認証は、音声に基づいてユーザが正当な者であるかを判定する音声による生体認証方式である。音声による生体認証は、指紋や虹彩に比べ認証精度が劣り、信頼性に欠ける。
Authentication via the authentication code requires the user to utter the authentication code every time, which is inconvenient. In addition, since the authentication code is issued by voice, the authentication code may be leaked to a third party. The biometric authentication of the uttered voice is a voice biometric authentication method that determines whether the user is a valid person based on the voice. Voice biometric authentication is less accurate and less reliable than fingerprints and irises.
二経路認証は、オンラインバンキングなどで採用されている認証方式である。この方式では、相互接続された2つのコンピュータデバイスのうちの第1のコンピュータデバイスで取引データを作成し、第2のコンピュータデバイスで承認することで取引が成立する。
Two-pass authentication is an authentication method employed in online banking and the like. In this method, a transaction is established by creating transaction data at a first computer device of two interconnected computer devices and approving at a second computer device.
このような方式は、セキュリティを確保することができるが、ユーザによるコンピュータデバイスを介した操作を必要とし、利便性に欠ける。特に、AIスピーカでは、音声により情報が交換される以上、人間による操作を簡易化することが要求される。このことは、例えば、車を運転しているとき、キッチンで作業しているときなどコンピュータデバイスを介した操作を簡易化することが要求されるときに有益となる。
Such a scheme can ensure security, but requires operation by a user via a computer device and lacks convenience. In particular, in the AI speaker, as long as information is exchanged by voice, it is required to simplify human operations. This is useful, for example, when driving a car, when working in a kitchen, or when it is required to simplify the operation through a computer device.
特許文献1は、顧客からの音声に応じて音声により銀行取引を実行する技術を開示している。特許文献1に係る技術では、顧客からの口座番号などの情報、およびサービスコードの音声入力によって本人確認を行っている。また、特許文献1に係る技術では、予め登録されている顧客の声紋と、顧客本人が発声した音声データの声紋とを比較することによって本人確認を行っている。
Patent Document 1 discloses a technology for executing a bank transaction by voice according to voice from a customer. In the technique according to Patent Document 1, the user is identified by voice input of information such as an account number from a customer and a service code. Further, in the technology according to Patent Document 1, the identification is performed by comparing the voiceprint of the customer registered in advance with the voiceprint of the voice data uttered by the customer himself.
特許文献1に係る技術は、上述した(1)発話による認証コードを介した認証、および(2)発話音声の生体認証を利用するものである。この技術は、上述した課題を何ら解決するものではない。
The technology according to Patent Document 1 uses (1) authentication through an authentication code by speech and (2) biometric authentication of speech. This technology does not solve the above-mentioned problems at all.
本発明は、このような問題に鑑みてなされたものであり、セキュリティが確保された方式でユーザからのクエリに対する応答を返却するクエリ応答システムを提供することにある。
The present invention has been made in view of such problems, and an object thereof is to provide a query response system that returns a response to a query from a user in a security secured manner.
上記の課題を解決するために、本発明に係るクエリ応答システムは、サーバコンピュータおよびクエリ応答デバイスを備えたコンピュータシステムであって、前記クエリ応答デバイスは、入力を受信し、前記入力に基づいてクエリ電文を生成してサーバコンピュータに送信し、前記サーバコンピュータは、クライアントコンピュータに接続確認要求電文を送信し、前記クライアントコンピュータから第1の識別子を含む接続確認応答電文を受信し、前記第1の識別子は、前記クエリ応答デバイスを識別し、前記第1の識別子と記憶された第2の識別子とが一致するかを判定することによって、前記クライアントコンピュータと前記クエリ応答デバイスとの間の接続が確立されていると判定し、前記第1の識別子と前記第2の識別子とが一致したと判定したことに応答して、前記クエリ電文に対応する応答電文を生成して前記クエリ応答デバイスに送信し、前記クエリ応答デバイスは、前記応答電文を出力することを特徴とする。
In order to solve the above problems, a query response system according to the present invention is a computer system comprising a server computer and a query response device, wherein the query response device receives an input and a query based on the input. A telegram is generated and transmitted to the server computer, the server computer transmits a connection confirmation request telegram to the client computer, receives a connection confirmation reply telegram including the first identifier from the client computer, and the first identifier A connection between the client computer and the query response device is established by identifying the query response device and determining whether the first identifier and the stored second identifier match. And the first identifier matches the second identifier. In response to determining that generates a response message corresponding to the query message sent to the query response device, wherein the query response device, and outputs the response message.
本発明に係るクエリ応答システムによれば、ユーザからのクエリに対する応答を返却する際のセキュリティを確保することができる。
According to the query response system according to the present invention, it is possible to secure security when returning a response to a query from a user.
以下、添付図面を参照して、本発明に係るクエリ応答システムを詳細に説明する。本実施形態に係るクエリ応答システムは、ユーザからのクエリに対する応答を介して特定のアプリケーションサービスを提供する。図1は、本発明の一実施形態に係るクエリ応答システムを構成するコンピュータシステム全体の構成の例を示す図である。以下で説明する実施形態では、本発明に係るクエリ応答システムがユーザからの発話(つまり、音声入力)に対して音声の形式で応答を返却する(つまり、音声出力)処理を実行する例として説明する。
Hereinafter, the query response system according to the present invention will be described in detail with reference to the attached drawings. The query response system according to the present embodiment provides a specific application service via a response to a query from a user. FIG. 1 is a diagram showing an example of the overall configuration of a computer system that constitutes a query response system according to an embodiment of the present invention. In the embodiment described below, the query response system according to the present invention is described as an example in which a response is returned in the form of voice (that is, voice output) in response to an utterance from the user Do.
クエリ応答システムは、サーバコンピュータ1、クエリ応答デバイス2、およびクライアントコンピュータ3を備える。サーバコンピュータ1と、クエリ応答デバイス2およびクライアントコンピュータ3とは、ネットワーク4(インターネットなどの公衆ネットワーク)を介して接続され、その接続方式は、有線または無線のいずれかであってもよい。クエリ応答デバイス2とクライアントコンピュータ3とは、Bluetooth(登録商標)などの近距離無線通信規格に従った接続方式で接続される。
The query response system comprises a server computer 1, a query response device 2, and a client computer 3. The server computer 1 and the query response device 2 and the client computer 3 are connected via the network 4 (a public network such as the Internet), and the connection method may be either wired or wireless. The query response device 2 and the client computer 3 are connected by a connection method in accordance with a short distance wireless communication standard such as Bluetooth (registered trademark).
サーバコンピュータ1は、ユーザにアプリケーションサービスを提供するサーバコンピュータデバイスである。サーバコンピュータ1は、ユーザからのクエリ応答デバイス2を通じたクエリ(発話)に応答して、実装するアプリケーションサービスに対応するアプリケーションプログラムを実行し、発話に応じた応答電文を生成する。応答電文は、クエリ応答デバイス2に送信される。
The server computer 1 is a server computer device that provides application services to users. The server computer 1 executes an application program corresponding to the application service to be implemented in response to a query (speech) from the user through the query response device 2, and generates a response message corresponding to the speech. The response message is sent to the query response device 2.
サーバコンピュータ1は、単一のデバイスで実装されてもよく、または複数のデバイスで実装されてもよい。また、サーバコンピュータ1は、それぞれが異なるサービスを実装する複数のデバイスで実装されてもよい。この場合、クエリ応答デバイス2がユーザからの発話を認識して(例えば、発話における特定のキーワードに基づいて)、発話に対応するアプリケーションサービスを実装するサーバコンピュータ1を選択することになる。
The server computer 1 may be implemented by a single device or may be implemented by a plurality of devices. Also, the server computer 1 may be implemented by a plurality of devices each implementing a different service. In this case, the query response device 2 recognizes an utterance from the user (for example, based on a specific keyword in the utterance), and selects a server computer 1 implementing an application service corresponding to the utterance.
クエリ応答デバイス2は、ユーザからの発話に応答して音声で応答する学習機能付きのスピーカデバイスである。クエリ応答デバイス2は、クライアントコンピュータ3が一定の範囲内にあることを検出することによって、クライアントコンピュータ3と接続する(Bluetooth接続を介して)。そして、ユーザの発話がされるとその音声を認識し、サーバコンピュータ1に、発話に対応するクエリ電文を送信し、クエリ電文に対応する応答電文を受信して音声に変換して音声出力する。つまり、クエリ応答デバイス2は、ユーザからの発話に対応する応答を返却するサーバコンピュータ1を選択し、サーバコンピュータ1からの応答を音声出力する仲介デバイスとして機能する。クエリ応答デバイス2も、単一のデバイスで実装されてもよく、または複数のデバイスで実装されてもよい。
The query response device 2 is a speaker device with a learning function that responds with speech in response to speech from the user. The query response device 2 connects with the client computer 3 (via Bluetooth connection) by detecting that the client computer 3 is within a certain range. Then, when the user utters, the voice is recognized, a query message corresponding to the utterance is transmitted to the server computer 1, a response message corresponding to the query message is received, converted to voice, and voice output is performed. That is, the query response device 2 selects the server computer 1 which returns a response corresponding to the speech from the user, and functions as an intermediary device which outputs the response from the server computer 1 as voice. The query response device 2 may also be implemented by a single device or by multiple devices.
クライアントコンピュータ3は、ユーザの認証に使用されるコンピュータデバイスである。クライアントコンピュータ3は、クエリ応答デバイス2との接続を確立すると、サーバコンピュータ1に接続情報を送信する。クライアントコンピュータ3とクエリ応答デバイス2との間での接続をサーバコンピュータ1が確認することによって、ユーザの認証が行われる。本実施形態では、クライアントコンピュータ3は、スマートフォンを含む携帯電話を想定しているが、そのような機器に限定されず、近距離無線通信機能(特に、Bluetooth接続機能)を有するコンピュータデバイスであれば、特に機器は限定されない。例えば、ユーザが衣服などに身に着けるウェアラブルデバイス(スマートウォッチおよびスマートグラス)などであってもよい。
The client computer 3 is a computer device used to authenticate a user. When establishing the connection with the query response device 2, the client computer 3 transmits connection information to the server computer 1. The server computer 1 confirms the connection between the client computer 3 and the query response device 2 to authenticate the user. In the present embodiment, the client computer 3 is assumed to be a mobile phone including a smartphone, but is not limited to such a device, and it is a computer device having a short distance wireless communication function (in particular, a Bluetooth connection function) Especially, the equipment is not limited. For example, it may be a wearable device (smart watch and smart glass) worn by a user on clothes or the like.
次に、図2を参照して、本発明の一実施形態に係るクエリ応答システムを構成するコンピュータシステムの詳細な構成の例を説明する。図2に示すように、サーバコンピュータ1は、制御装置11、メモリ12、記憶装置13、および通信装置14を含み、それらの構成要素がバスを介して相互に結合されている。
Next, with reference to FIG. 2, an example of a detailed configuration of a computer system constituting the query response system according to an embodiment of the present invention will be described. As shown in FIG. 2, the server computer 1 includes a control unit 11, a memory 12, a storage unit 13, and a communication unit 14. These components are mutually connected via a bus.
制御装置11は、プロセッサなどで実装され、中央処理装置(CPU)、グラフィックプロセシングユニット(GPU)、および/または1つもしくは複数のプロセッサコアを含んでもよい。サーバコンピュータ1は、DBMS(Database Management System)を実装し、制御装置11は、OS(Operating System)を実装する各プログラム、アプリケーションサービスを実装する各アプリケーションプログラム、およびDBMSが実装する各プログラムを実行する。メモリ12は、揮発性記憶領域として機能し、例えば、ランダムアクセスメモリ(RAM)、ダイナミックRAM、またはキャッシュメモリを含んでもよい。制御装置11は、上述したプログラムをメモリ12に読み出して実行する。
The controller 11 may be implemented by a processor or the like, and may include a central processing unit (CPU), a graphic processing unit (GPU), and / or one or more processor cores. The server computer 1 implements a DBMS (Database Management System), and the control device 11 executes programs implementing an OS (Operating System), application programs implementing an application service, and programs implementing a DBMS. . The memory 12 functions as a volatile storage area, and may include, for example, random access memory (RAM), dynamic RAM, or cache memory. The control device 11 reads the program described above into the memory 12 and executes the program.
記憶装置13は、サーバコンピュータ1に内蔵された記憶装置であり、例えば、ハードディスクドライブ、ソリッドステートドライブ、光ディスク、およびフラッシュドライブを含んでもよい。記憶装置13は、上述したプログラム、および後述する供給先データテーブルなどのデータベーステーブル13aを記憶する。通信装置14は、例えば、ネットワークインタフェースカード(例えば、LANカード)などで実装され、ネットワーク4を介してクエリ応答デバイス2およびクライアントコンピュータ3との間でデータを送受信する。
The storage device 13 is a storage device built in the server computer 1 and may include, for example, a hard disk drive, a solid state drive, an optical disk, and a flash drive. The storage device 13 stores the above-described program and a database table 13a such as a supply destination data table described later. The communication device 14 is implemented by, for example, a network interface card (for example, a LAN card) and transmits and receives data between the query response device 2 and the client computer 3 via the network 4.
クエリ応答デバイス2は、制御装置21、メモリ22、記憶装置23、通信装置24、入力ドライバ25、マイクロフォン26、出力ドライバ27、およびスピーカ28を含み、それらの構成要素がバスを介して相互に結合されている。
The query response device 2 includes a control unit 21, a memory 22, a storage unit 23, a communication unit 24, an input driver 25, a microphone 26, an output driver 27, and a speaker 28, whose components are mutually coupled via a bus. It is done.
制御装置21は、サーバコンピュータ1と同様にCPUなどで実装される。制御装置21は、OSを実装するプログラム、音声信号をデジタルデータに符号化し、デジタルデータを音声信号に復号するコーデックを実装したプログラム(コーデックプログラム)、および学習機能を実装するプログラム(学習プログラム)などをメモリ22に読み出して実行する。なお、学習機能は、ニューラルネットワークモデルなどの既知の技術を使用して実装されてもよい。
The control device 21 is implemented by a CPU or the like as in the server computer 1. The control device 21 implements a program for implementing the OS, a program for implementing a codec that encodes audio signals into digital data and decodes the digital data into audio signals (codec program), and a program for implementing a learning function (learning program) Are read into the memory 22 and executed. Note that the learning function may be implemented using known techniques, such as neural network models.
記憶装置23は、サーバコンピュータ1と同様に、ハードディスクドライブなどを含む内蔵された記憶装置であり、ニューラルネットワークモデルなどに従って蓄積した学習データを記憶する。通信装置24は、無線LANカードおよびBluetoothアダプタなどを含む。通信装置24は、サーバコンピュータ1との間でネットワーク4を介してデータを送受信し、クライアントコンピュータ3との間でBluetooth接続を介してデータを送受信する。
Similar to the server computer 1, the storage device 23 is a built-in storage device including a hard disk drive and the like, and stores learning data accumulated according to a neural network model or the like. The communication device 24 includes a wireless LAN card, a Bluetooth adapter, and the like. The communication device 24 transmits and receives data to and from the server computer 1 via the network 4, and transmits and receives data to and from the client computer 3 via a Bluetooth connection.
マイクロフォン(入力装置)26は、入力ドライバ25を介して、ユーザによる発話を受信する。受信した発話は、制御装置21(コーデック)によって符号化されてクエリ電文に変換される。スピーカ(出力装置)28は、サーバコンピュータ1から受信した応答電文から変換された音声信号を音声出力する。応答電文は、制御装置21(コーデック)によって復号されて、出力ドライバ27を介してスピーカ28に送信される。
The microphone (input device) 26 receives an utterance by the user via the input driver 25. The received speech is encoded by the control device 21 (codec) and converted into a query message. The speaker (output device) 28 outputs the voice signal converted from the response message received from the server computer 1 as voice. The response message is decoded by the control device 21 (codec) and transmitted to the speaker 28 via the output driver 27.
クライアントコンピュータ3は、図示しない通信装置および制御装置を有する。通信装置は、サーバコンピュータ1との間でネットワーク4を介してデータを送受信し、クライアントコンピュータ3との間でBluetooth接続を介してデータを送受信する。制御装置は、OSおよびサーバコンピュータ1が実行するアプリケーションプログラムに対応するクライアントアプリケーションプログラムを実行する。なお、クライアントアプリケーションプログラムは、サーバコンピュータ1から事前にダウンロードされている。
The client computer 3 has a communication device and a control device (not shown). The communication device transmits / receives data to / from the server computer 1 via the network 4 and transmits / receives data to / from the client computer 3 via the Bluetooth connection. The control device executes a client application program corresponding to the OS and an application program executed by the server computer 1. The client application program is downloaded from the server computer 1 in advance.
次に、図3を参照して、本発明の一実施形態に係るユーザデータテーブル300のデータ構造の例を説明する。本実施形態では、サーバコンピュータ1が提供するアプリケーションサービスがインターネットバンキングサービスであるものとして説明する。そのことを前提に、ユーザからの発話が自身の口座の残高に関する問い合わせに関するものであり、それに対する応答が残高に関するものである。ユーザデータテーブル300は、サーバコンピュータ1の記憶装置13に記憶されたデータベーステーブル13aのうち、残高に関する問い合わせに対するユーザの認証に使用されるデータテーブルである。
Next, an example of the data structure of the user data table 300 according to the embodiment of the present invention will be described with reference to FIG. In the present embodiment, the application service provided by the server computer 1 will be described as the Internet banking service. On that premise, the utterance from the user relates to the inquiry about the balance of his account, and the response to it relates to the balance. The user data table 300 is a data table used for authenticating the user with respect to the inquiry about the balance among the database tables 13 a stored in the storage device 13 of the server computer 1.
図3に示すように、ユーザデータテーブル300は、データ項目「ユーザID」、「クライアントコンピュータ通知ID」、「利用デバイス名」、および「有効フラグ」を含む。各データレコードは、ユーザごとに生成される。ユーザIDは、サーバコンピュータ1が実装するアプリケーションサービス(以下、「アプリケーションサービス」)においてユーザごとに一意な識別子が設定される。クライアントコンピュータ通知IDは、サーバコンピュータ1がクライアントコンピュータに、後述する接続確認要求電文を送信するための送信先を識別するIDである。利用デバイス名は、対象のユーザが使用するクエリ応答デバイス2の名称(例えば、型式番号など)が設定される。つまり、対象のユーザと使用するクエリ応答デバイス2とが関連付けて記憶される。有効フラグは、残高に関する問い合わせに対して応答するサービスが有効であるか否かを示す値が設定され、例えば、「1:有効」または「2:無効」である。
As shown in FIG. 3, the user data table 300 includes data items “user ID”, “client computer notification ID”, “used device name”, and “valid flag”. Each data record is generated for each user. As the user ID, a unique identifier is set for each user in an application service (hereinafter, "application service") implemented by the server computer 1. The client computer notification ID is an ID by which the server computer 1 identifies a transmission destination for transmitting a connection confirmation request message described later to the client computer. The name (for example, model number etc.) of the query response device 2 which a target user uses is set as a utilization device name. That is, the target user and the query response device 2 to be used are stored in association with each other. The validity flag is set to a value indicating whether the service responding to the inquiry about the balance is valid, and is, for example, "1: valid" or "2: invalid".
ユーザデータテーブル300における対象のユーザのデータレコードは、例えば、対象のユーザが、アプリケーションサービスに登録するときに生成(追加)される。ユーザがアプリケーションサービスに登録する際に、一意なユーザIDが割り当てられ、所定の認証(パスワードなど)でログインする。このときに割り当てられるユーザIDが、ユーザデータテーブル300の対象のデータレコードに設定される。
The data record of the target user in the user data table 300 is generated (added), for example, when the target user registers in the application service. When the user registers with the application service, a unique user ID is assigned and logs in with a predetermined authentication (such as a password). The user ID assigned at this time is set in the target data record of the user data table 300.
また、後述する接続確認を行うための接続確認要求電文は、例えば、サーバコンピュータ1が実装するOSのプッシュ通知サービスを利用して送信される。プッシュ通知サービスは、例えば、クライアントコンピュータ3がiOS(登録商標)を実装している場合「APN(Apple Push Notification Service)」を利用し、Android(登録商標)を実装している場合「GCM(Google Cloud Messaging)」を利用する。よって、ユーザが上記プッシュ通知サービスに登録する際に割り当てられたIDが、ユーザデータテーブル300の対象のデータレコードに設定される(クライアントコンピュータ通知ID)。
Further, a connection confirmation request message for performing connection confirmation, which will be described later, is transmitted, for example, using a push notification service of the OS mounted on the server computer 1. The push notification service uses, for example, “APN (Apple Push Notification Service)” when the client computer 3 implements iOS (registered trademark) and “GCM (Google Google) when Android (registered trademark) is installed. Use Cloud Messaging). Therefore, the ID assigned when the user registers in the push notification service is set in the target data record of the user data table 300 (client computer notification ID).
なお、上述したプッシュ通知サービスは例示的なものにすぎず、上記例に限定されない。サーバコンピュータ1がクライアントコンピュータ3に接続確認要求電文を送信することができる任意の方式が利用されてもよい。
The above-described push notification service is merely an example, and is not limited to the above example. Any method by which the server computer 1 can transmit a connection confirmation request message to the client computer 3 may be used.
Bluetooth接続では、2つのデバイス(Bluetooth接続対応デバイス)との間でペアリングを行う必要がある。2つのデバイス間が通信可能な範囲内にあると、相互に他方のデバイスの存在を検出し、対象のデバイス名を認識する。その後、いずれかのデバイスでPINコードを入力すると、ペアリングが完了し、双方の機器にペアリング情報が記憶される(他方の機器のデバイス名など)する。クライアントコンピュータ3とクエリ応答デバイス2との間では事前にペアリングを行い、そのクエリ応答デバイス2のデバイス名がユーザデータテーブル300の利用デバイス名に設定される(つまり、クライアントコンピュータ3とクエリ応答デバイス2とのペアリング情報がユーザデータテーブル300に登録される)。
In Bluetooth connection, pairing needs to be performed between two devices (devices compatible with Bluetooth connection). When the two devices are within the communicable range, the presence of the other device is detected mutually and the device name of the object is recognized. Thereafter, when a PIN code is input in any device, pairing is completed, and pairing information is stored in both devices (such as the device name of the other device). The client computer 3 and the query response device 2 are paired in advance, and the device name of the query response device 2 is set to the use device name of the user data table 300 (that is, the client computer 3 and the query response device The pairing information with 2 is registered in the user data table 300).
Bluetooth接続では、2つのデバイスがIEEE(Institute of Electrical and Electronics Engineers)802.15.1で規定された2.4GHz帯の電波を放出し、数メートルから数10メートルの範囲で通信を行う。つまり、クライアントコンピュータ3とクエリ応答デバイス2との間の通信は、双方が数メートルから数10メートルの範囲内にあるときにのみ有効となる。
In Bluetooth connection, two devices emit radio waves in the 2.4 GHz band defined by IEEE (Institute of Electrical and Electronics Engineers) 802.15.1 and communicate in a range of several meters to several tens of meters. That is, communication between the client computer 3 and the query response device 2 is effective only when both are in the range of several meters to several tens of meters.
次に、図4を参照して、本発明の一実施形態に係る口座データテーブル400のデータ構造の例を説明する。口座データテーブル400は、上述したサービスを実装するために必要な、サーバコンピュータ1の記憶装置13に記憶されたデータベーステーブル13aのうち、ユーザごとの口座情報を記憶したデータテーブルである。
Next, with reference to FIG. 4, an example of the data structure of the account data table 400 according to the embodiment of the present invention will be described. The account data table 400 is a data table storing account information for each user among the database tables 13a stored in the storage device 13 of the server computer 1 necessary for implementing the above-described service.
図4に示すように、口座データテーブル400は、データ項目「ユーザID」、「口座番号」、「金融機関コード」、「支店コード」、「口座種別」、「名義人」、「残高」、および「限度額」を含む。各データレコードは、上述したインターネットバンキングサービスを利用するユーザごとに生成される。つまり、本実施形態に係るアプリケーションサービスを利用するユーザごとに、図3で説明したユーザデータテーブル300のデータレコードと対で生成される。限度額は、後述する残高に関する問い合わせに対して応答を制限する場合に所定の値が設定される。
As shown in FIG. 4, the account data table 400 includes data items "user ID", "account number", "financial institution code", "branch code", "account type", "nominee", "balance", And "limit amount". Each data record is generated for each user who uses the Internet banking service described above. That is, each user who uses the application service according to the present embodiment is generated as a pair with the data record of the user data table 300 described in FIG. The limit amount is set to a predetermined value when limiting the response to the inquiry about the balance described later.
<第1の実施形態>
次に、図5を参照して、本発明の第1の実施形態に係るクエリ応答システムが実行する処理を説明する。この処理では、ユーザAがクエリ応答システムに自身の残高を問い合わせる発話をし、それに対して残高に関する応答が音声で返却される。ユーザAの口座情報は、口座データテーブル400に登録されているものとする。 First Embodiment
Next, processing performed by the query response system according to the first embodiment of the present invention will be described with reference to FIG. In this process, the user A utters the query response system to inquire about its balance, and a response relating to the balance is audibly returned. It is assumed that the account information of the user A is registered in the account data table 400.
次に、図5を参照して、本発明の第1の実施形態に係るクエリ応答システムが実行する処理を説明する。この処理では、ユーザAがクエリ応答システムに自身の残高を問い合わせる発話をし、それに対して残高に関する応答が音声で返却される。ユーザAの口座情報は、口座データテーブル400に登録されているものとする。 First Embodiment
Next, processing performed by the query response system according to the first embodiment of the present invention will be described with reference to FIG. In this process, the user A utters the query response system to inquire about its balance, and a response relating to the balance is audibly returned. It is assumed that the account information of the user A is registered in the account data table 400.
また、クライアントコンピュータ3とクエリ応答デバイス2との間の通信は、Bluetooth接続により通信が行われ、双方の間で既にペアリングが行われ、ユーザAのユーザID、利用デバイス名、およびクライアントコンピュータ通知IDがユーザデータテーブル300に登録されているものとする。このような状態で、クライアントコンピュータ3が音声対応デバイスとBluetooth接続が可能な範囲に入ると、自動的に接続状態となる。
Also, communication between the client computer 3 and the query response device 2 is performed by Bluetooth connection, pairing is already performed between the two, user ID of the user A, use device name, and client computer notification It is assumed that the ID is registered in the user data table 300. In such a state, when the client computer 3 enters a range in which a Bluetooth connection can be made with the voice compatible device, the connection state is automatically established.
図5で説明する処理におけるサーバコンピュータ1によって実行される処理は、サーバコンピュータ1の制御装置11がアプリケーションプログラムを実行することによって実装される。また、クエリ応答デバイス2によって実行される処理は、クエリ応答デバイス2の制御装置21がコーデックプログラムおよび学習プログラムなどを実行することによって実装される。さらに、クライアントコンピュータ3によって実行される処理は、クライアントコンピュータ3の制御装置がクライアントアプリケーションプログラムを実行することによって実装される。
The process executed by the server computer 1 in the process described in FIG. 5 is implemented by the control device 11 of the server computer 1 executing an application program. Also, the processing executed by the query response device 2 is implemented by the control device 21 of the query response device 2 executing a codec program, a learning program, and the like. Furthermore, the processing executed by the client computer 3 is implemented by the control device of the client computer 3 executing the client application program.
まず、ステップS501では、クエリ応答デバイス2の通信装置24は、クライアントコンピュータ3の存在を検出する(または、クライアントコンピュータ3がクエリ応答デバイス2の存在を検出する)。この検出の処理は、クライアントコンピュータ3が放出する電波をクエリ応答デバイス2が検出することによって実行される。ステップS501でクエリ応答デバイス2がクライアントコンピュータ3の存在を検出すると、両者間での接続が確立し、クライアントコンピュータ3は、クエリ応答デバイス2のデバイス名を認識することができる。
First, in step S501, the communication device 24 of the query response device 2 detects the presence of the client computer 3 (or the client computer 3 detects the presence of the query response device 2). The process of this detection is executed by the query response device 2 detecting the radio wave emitted by the client computer 3. When the query response device 2 detects the presence of the client computer 3 in step S501, a connection between the two is established, and the client computer 3 can recognize the device name of the query response device 2.
次に、ユーザAは、クエリ応答デバイス2に対し、例えば、「口座の残高を教えてください」と発話する(ステップS502)。次に、クエリ応答デバイス2のマイクロフォン26が、ステップS502における発話を受信し音声信号に変換すると、制御装置21に送信する。制御装置21は、音声信号をデジタルデータに変換し、デジタルデータに含まれるキーワードを分析する(つまり、ユーザAの発話を分析する)(ステップS503)。キーワードの分析は、デジタルデータに変換された発話におけるキーワードを抽出することによって実行される。キーワード抽出は、例えば、TF-IDF(Term Frequency-Inverse Document Frequency)およびSVM(Support Vector Machine)などの既知の特徴語抽出アルゴリズムを使用して実行される。
Next, the user A utters, for example, "Please tell me the balance of the account" to the query response device 2 (step S502). Next, when the microphone 26 of the query response device 2 receives the speech in step S502 and converts it into an audio signal, the microphone 26 transmits the speech signal to the control device 21. The control device 21 converts the voice signal into digital data, and analyzes the keyword included in the digital data (that is, analyzes the speech of the user A) (step S503). The analysis of keywords is performed by extracting keywords in the speech converted to digital data. The keyword extraction is performed using known feature word extraction algorithms such as, for example, TF-IDF (Term Frequency-Inverse Document Frequency) and SVM (Support Vector Machine).
この分析の結果、ニューラルネットワークなどの学習データに従って、分析したキーワードに対応するアプリケーションサービスを特定する(ステップS504)。本実施形態では、例えば、学習データには、「残高」などのキーワードとインターネットバンキングサービスとが関連付けられたデータが記憶されている。ユーザからの発話にキーワード「残高」が含まれる場合に、制御装置21は、インターネットバンキングサービスを提供するサーバコンピュータ1を特定する。
As a result of this analysis, an application service corresponding to the analyzed keyword is specified according to learning data such as a neural network (step S504). In the present embodiment, for example, data in which a keyword such as “balance” and the Internet banking service are associated is stored in the learning data. When the utterance from the user includes the keyword "balance", the control device 21 specifies the server computer 1 providing the internet banking service.
次に、制御装置21は、クエリ電文を生成し、通信装置24がステップS504で特定したサーバコンピュータ1に送信する(ステップS505)。クエリ電文は、少なくともステップS501で取得したクライアントコンピュータ3のユーザAによる要求の種別を示す情報を含む。要求の種別は、所定のキーワードと関連付けて学習データに記憶されており、例えば、キーワード「残高」と、残高に関する問い合わせ要求とが関連付けられている。
Next, the control device 21 generates a query message, and the communication device 24 transmits it to the server computer 1 identified in step S504 (step S505). The query message includes at least information indicating the type of request by the user A of the client computer 3 acquired in step S501. The type of request is stored in the learning data in association with a predetermined keyword. For example, the keyword “balance” is associated with an inquiry request regarding the balance.
次に、サーバコンピュータ1の通信装置14が、ステップS505で送信されたクエリ電文を受信すると、制御装置11は、ユーザデータテーブル300から対象のデータレコードを取得する(ステップS506)。そして、制御装置11は、クライアントコンピュータ3とクエリ応答デバイス2との間で接続が確立されているかを確認するための接続確認要求電文を生成する。そして、通信装置14は、ステップS506で取得したユーザデータレコードに含まれるクライアントコンピュータ通知IDを宛先(つまり、クライアントコンピュータ3)として接続確認要求電文を送信する(ステップS507)。この接続確認要求電文は、上述したプッシュ通知サービスなどを利用して送信される。
Next, when the communication device 14 of the server computer 1 receives the query message transmitted in step S505, the control device 11 acquires a target data record from the user data table 300 (step S506). Then, the control device 11 generates a connection confirmation request message for confirming whether a connection is established between the client computer 3 and the query response device 2. Then, the communication device 14 transmits a connection confirmation request message with the client computer notification ID included in the user data record acquired in step S506 as the destination (that is, the client computer 3) (step S507). The connection confirmation request message is transmitted using the above-described push notification service or the like.
次に、クライアントコンピュータ3がステップS507で送信された接続確認要求電文を受信すると、接続確認応答電文を生成し、サーバコンピュータ1に送信する(ステップS508)。接続確認応答電文は、少なくともステップS501で取得したクエリ応答デバイス2のデバイス名を含む。
Next, when the client computer 3 receives the connection confirmation request message transmitted in step S507, the client computer 3 generates a connection confirmation response message and transmits it to the server computer 1 (step S508). The connection confirmation response message includes at least the device name of the query response device 2 acquired in step S501.
ステップS508でクライアントコンピュータ3が接続確認要求電文を受信した時点で、クライアントコンピュータ3上では、クライアントアプリケーションプログラムが実行されている必要がある。クライアントアプリケーションプログラムが実行されていない場合、クライアントコンピュータ3のOSがクライアントアプリケーションプログラムを起動してもよい。クライアントアプリケーションプログラムを起動する際は、接続確認要求電文のポート番号を参照することによって、対象のアプリケーションを特定することができる(ポート番号は、サーバコンピュータ1のアプリケーションプログラムに割り当てられたポート番号と関連付けられる)。
When the client computer 3 receives the connection confirmation request message in step S508, the client application program needs to be executed on the client computer 3. When the client application program is not executed, the OS of the client computer 3 may start the client application program. When starting the client application program, the target application can be identified by referring to the port number of the connection confirmation request message (the port number is associated with the port number assigned to the application program of the server computer 1) ).
次に、サーバコンピュータ1の通信装置14が、ステップS508で送信された接続確認応答電文を受信すると、制御装置11は、接続確認応答電文に含まれるデバイス名(第1の識別子)が、ステップS506で取得したユーザデータテーブルに含まれる利用デバイス名(第2の識別子)と一致するかを判定する(ステップS509)。この処理の際に、ユーザデータレコードに含まれる利用デバイス名が接続確認応答電文に含まれるデバイス名と一致する場合、クライアントコンピュータ3と対象のデバイス名に対応するクエリ応答デバイス2との間で接続が確立されていることを確認することができる。つまり、ユーザAからの発話が、対象のクエリ応答デバイス2と接続されたクライアントコンピュータ3を有するユーザであることを確認することができ、ひいてはユーザAの本人確認を行うことができる。
Next, when the communication device 14 of the server computer 1 receives the connection confirmation response message transmitted in step S508, the control device 11 determines that the device name (first identifier) included in the connection confirmation response message is in step S506. It is determined whether or not the used device name (second identifier) included in the user data table acquired in step S2 matches (step S509). In this process, if the used device name included in the user data record matches the device name included in the connection confirmation response message, connection is established between the client computer 3 and the query response device 2 corresponding to the target device name. Can be confirmed. That is, it can be confirmed that the speech from the user A is a user having the client computer 3 connected to the target query response device 2, and consequently, the identity of the user A can be confirmed.
なお、ステップS509の処理において、クエリ応答デバイス2とクライアントコンピュータ3との間の接続が確認できなかった場合、サーバコンピュータ1が対象のユーザに接続を促す旨の通知を行ってもよい。具体的には、サーバコンピュータ1の通信装置14が、予め定められたメッセージをクエリ応答デバイス2および/またはクライアントコンピュータ3に送信する。クエリ応答デバイス2および/またはクライアントコンピュータ3は、送信されたメッセージを音声または視覚的表示などの所定の形式で出力する。
If the connection between the query response device 2 and the client computer 3 can not be confirmed in the process of step S 509, the server computer 1 may notify the target user to urge the connection. Specifically, the communication device 14 of the server computer 1 sends a predetermined message to the query response device 2 and / or the client computer 3. The query response device 2 and / or the client computer 3 outputs the transmitted message in a predetermined format such as audio or visual display.
次に、制御装置11は、ステップS505で送信されたクエリ電文に含まれる要求の種別に従って処理を実行する。本実施形態では、要求の種別が残高に関する問い合わせ要求であるので、制御装置11は、ステップS506で取得したデータレコードに含まれるユーザIDに基づいて、口座データテーブル400から対象のデータレコードを取得する(ステップS510)。データレコードを取得すると、制御装置11は、取得したデータレコードに基づいて、クエリ応答電文を生成し、通信装置14がクエリ応答デバイス2に送信する(ステップS511)。クエリ応答電文は、少なくともステップS510で取得したデータレコードに含まれる残高を含む。
Next, the control device 11 executes processing in accordance with the type of request included in the query message transmitted in step S505. In the present embodiment, since the type of request is an inquiry request regarding balance, the control device 11 acquires the target data record from the account data table 400 based on the user ID included in the data record acquired in step S506. (Step S510). When acquiring the data record, the control device 11 generates a query response message based on the acquired data record, and the communication device 14 transmits it to the query response device 2 (step S511). The query response message includes at least the balance included in the data record acquired in step S510.
次に、クエリ応答デバイス2の通信装置24が、ステップS511で送信されたクエリ応答電文を受信すると、制御装置21は、クエリ応答電文を音声信号に変換する(ステップS512)。変換された音声信号は、出力ドライバ27を介してスピーカ28を通じて音声出力される(ステップS513)。例えば、音声出力は、「残高は1,029,389円です。」などである。
Next, when the communication device 24 of the query response device 2 receives the query response message transmitted in step S511, the control device 21 converts the query response message into an audio signal (step S512). The converted audio signal is output as audio through the speaker 28 via the output driver 27 (step S513). For example, the voice output is "The balance is 1,029,389 yen."
ステップS509でデバイス名が一致しなかった場合、つまりユーザAの本人確認ができず、その旨を通知するメッセージを通知する場合は通知してから所定の時間が経過した後、ステップS510の処理は実行されずにエラー応答が返却される。この場合、ステップS511では、本人確認ができなかったことを示すエラー応答電文が生成され、ステップS512およびステップS513の処理を通じてその旨が音声出力される。例えば、音声出力は、「本人確認ができなかったのでお答えできません。」などである。
If the device names do not match in step S509, that is, the identity of the user A can not be confirmed, and a message notifying that effect is notified, the process of step S510 is performed after a predetermined time has passed since the notification. An error response is returned without being executed. In this case, in step S511, an error response message indicating that the identity verification could not be made is generated, and that effect is voice-outputted through the processing in steps S512 and S513. For example, the voice output is "I can not answer because I could not confirm my identity."
上述したエラー応答は、ステップS510で取得した口座データテーブル400のデータレコードに含まれる残高が限度額を上回る場合にも適用されてもよい。これによって、一定の額を上回る残高をユーザが有している場合に、問い合わせに対する回答を制限することができる。
The error response described above may also be applied when the balance included in the data record of the account data table 400 acquired in step S510 exceeds the limit amount. This makes it possible to limit the response to the inquiry when the user has a balance exceeding a certain amount.
サーバコンピュータ1は、ステップS505の処理以降のいずれか(つまり、ユーザAから残高に関する問い合わせがあった後のいずれか)で、ユーザAから残高に関する問い合わせがあったことを示す通知をクライアントコンピュータ3に送信してもよい(例えば、電子メールの形式で)。この処理によって、第三者による残高に関する問い合わせに関する発話の試みが行われたことをユーザAが確認することができる。
The server computer 1 notifies the client computer 3 that there is an inquiry about the balance from the user A at any time after the process of step S 505 (that is, any after the user A has made an inquiry about the balance). It may be sent (eg, in the form of an email). This process allows the user A to confirm that an attempt has been made by a third party to make an inquiry regarding a balance inquiry.
また、ステップS509の処理以降のいずれか(つまり、ユーザAの本人確認に成功した後のいずれか)で、ユーザAから残高に関する問い合わせに対し応答したことを示す通知をクライアントコンピュータ3に送信してもよい。さらに、ステップS509の処理において本人確認に失敗した場合、その処理以降のいずれかで、ユーザAから残高に関する問い合わせに対し本人確認ができなかったことを示す通知をクライアントコンピュータ3に送信してもよい。いずれの処理によっても、第三者による残高に関する問い合わせに対する本人確認が成功/失敗したことをユーザAが確認することができる。
In addition, a notification indicating that the user A has responded to the inquiry about the balance is transmitted to the client computer 3 at any time after the process of step S 509 (that is, any one after the user A's identity verification is successful). It is also good. Furthermore, if identity verification fails in the process of step S509, a notification may be sent to the client computer 3 indicating that identity verification could not be performed in response to the inquiry regarding the balance from the user A, in any of subsequent processes. . With any of the processes, the user A can confirm that the identification of the third party's inquiry about the balance is successful or unsuccessful.
なお、ユーザAは、クライアントコンピュータ3の操作を介して上述した残高に関する問い合わせの機能を無効化することができる。この無効化のために、クライアントコンピュータ3の操作を介してユーザデータテーブル300の対象のデータレコードに含まれる有効フラグが「2:無効」に更新される。この処理によって、制御装置11は、上述したステップS508の処理以降において有効フラグに設定された値を設定することによって、クエリ応答デバイス2にエラー応答を返却することができる。
In addition, the user A can invalidate the function of the inquiry about the balance described above through the operation of the client computer 3. For this invalidation, the valid flag included in the target data record of the user data table 300 through the operation of the client computer 3 is updated to “2: invalid”. By this processing, the control device 11 can return an error response to the query response device 2 by setting the value set in the valid flag in the processing after step S508 described above.
以上のように、本発明に第1の実施形態に係るクエリ応答システムを説明した。本実施形態に係るクエリ応答システムによると、PINコードなどの発話も入力も必要とせずにユーザ認証を行うことができ、残高などの秘密情報の交換においてセキュリティを確保することができる。また、クライアントコンピュータ3とクエリ応答デバイス2との間でのBluetooth接続を介して応答が返却されるので、少なくともクライアントコンピュータ3がクエリ応答デバイス2から一定の範囲内にない応答が返却されない。よって、例えば、第三者のコンピュータデバイスがネットワークを介した遠隔操作などによって残高を盗聴するような行為を防止することができる。
As described above, the query response system according to the first embodiment of the present invention has been described. According to the query response system according to the present embodiment, user authentication can be performed without requiring an utterance or input of a PIN code or the like, and security can be secured in exchange of secret information such as a balance. Further, since the response is returned via the Bluetooth connection between the client computer 3 and the query response device 2, at least the client computer 3 is not returned from the query response device 2 as a response. Thus, for example, it is possible to prevent a third party computer device from tapping the balance by remote control or the like via the network.
また、クライアントコンピュータ3とクエリ応答デバイス2との間でのBluetooth接続が条件となることから、一方のデバイスが盗難などにあっても、双方のデバイスが揃わない限り、第三者による盗聴を防止することができる。
In addition, since the Bluetooth connection between the client computer 3 and the query response device 2 is a condition, even if one of the devices is stolen etc., it is possible to prevent eavesdropping by a third party unless both devices are aligned. can do.
さらに、クライアントコンピュータ3をロック機能付きのコンピュータデバイスで実装することによってさらにセキュリティを高めることができる。一方で、クライアントコンピュータ3をウェアラブルデバイスで実装する場合、ユーザによる操作を必要とせずに本発明に係るクエリ応答システムを利用することができる。ウェアラブルデバイスでは、ユーザがデバイスを装着している場合はロックが解除されている状態であるので(内部センサがユーザとデバイスとの密着状態を検出することによって)、ユーザがロックを解除する操作を必要としないからである。
Furthermore, security can be further enhanced by implementing the client computer 3 with a computer device with a lock function. On the other hand, when the client computer 3 is implemented by a wearable device, the query response system according to the present invention can be used without requiring an operation by the user. In the wearable device, when the user wears the device, the lock is released (by the internal sensor detecting the contact state between the user and the device), the user performs an operation to release the lock. It is because it is not necessary.
なお、第1の実施形態では、クライアントコンピュータ3とクエリ応答デバイス2との間の接続がBluetooth通信規格に従って確立されるものとして説明したが、そのような形式に限定されない。Bluetoothと同様のペアリングに相当する手順を行ういずれかの通信方式を採用してもよい。この場合、最初のデバイス間接続時に双方で他方のデバイスを識別する情報を保持し、その後の通信は、双方のデバイスが所定の範囲内にあるときに秘密鍵によって通信が暗号化されることになる。以下の第2の実施形態でも同様である。
In the first embodiment, the connection between the client computer 3 and the query response device 2 is described as being established according to the Bluetooth communication standard, but the present invention is not limited to such a format. Any communication method may be adopted which performs a procedure equivalent to pairing similar to Bluetooth. In this case, at the time of the first inter-device connection, both sides hold information identifying the other device, and the subsequent communication is that the communication is encrypted by the secret key when both devices are within the predetermined range. Become. The same applies to the second embodiment described below.
また、Bluetooth以外の通信方式を利用する場合でも、本実施形態に係るクエリ応答システムを一定の範囲内での使用に制限することによってセキュリティを確保するために、近距離無線接続の通信方式を利用することが望ましい。例えば、ZigBeeや赤外線を利用した近距離無線通信などである。
In addition, even when using a communication method other than Bluetooth, in order to ensure security by restricting the use of the query response system according to the present embodiment within a certain range, a communication method of near field wireless connection is used. It is desirable to do. For example, near-field wireless communication using ZigBee or infrared light is used.
<第2の実施形態>
次に、図6および図7を参照して、本発明の第2の実施形態に係るクエリ応答システムが実行する処理を説明する。図6は、ユーザからの発話におけるキーワードに基づいて識別される識別対象が複数あり、それぞれの識別対象が対応するキーワードに基づいて識別されることを示す。つまり、発話に含まれる1つまたは複数のキーワードに基づいて、アプリケーションサービスおよびユーザなどの複数の異なる識別対象が判定される。 Second Embodiment
Next, processing performed by the query response system according to the second embodiment of the present invention will be described with reference to FIGS. 6 and 7. FIG. 6 shows that there are a plurality of identification targets identified based on the keyword in the utterance from the user, and each identification target is identified based on the corresponding keyword. That is, based on one or more keywords included in the utterance, a plurality of different identification targets such as the application service and the user are determined.
次に、図6および図7を参照して、本発明の第2の実施形態に係るクエリ応答システムが実行する処理を説明する。図6は、ユーザからの発話におけるキーワードに基づいて識別される識別対象が複数あり、それぞれの識別対象が対応するキーワードに基づいて識別されることを示す。つまり、発話に含まれる1つまたは複数のキーワードに基づいて、アプリケーションサービスおよびユーザなどの複数の異なる識別対象が判定される。 Second Embodiment
Next, processing performed by the query response system according to the second embodiment of the present invention will be described with reference to FIGS. 6 and 7. FIG. 6 shows that there are a plurality of identification targets identified based on the keyword in the utterance from the user, and each identification target is identified based on the corresponding keyword. That is, based on one or more keywords included in the utterance, a plurality of different identification targets such as the application service and the user are determined.
本実施形態では、識別対象は、ユーザに提供するアプリケーションサービスを含み、つまり、ユーザからの発話における第1のキーワードに基づいてアプリケーションサービスが識別される(本実施形態では、インターネットバンキングサービス)。また、識別対象は、ユーザによる要求の種別を含み、ユーザからの発話における第2のキーワードに基づいて要求の種別が識別される(本実施形態では、残高に関する問い合わせ)。さらに、識別対象は、問い合わせたユーザを含み、つまり、ユーザからの発話における第3のキーワードに基づいて問い合わせたユーザが識別(認証)される。
In the present embodiment, the identification target includes the application service provided to the user, that is, the application service is identified based on the first keyword in the speech from the user (in the present embodiment, the Internet banking service). Further, the identification target includes the type of the request by the user, and the type of the request is identified based on the second keyword in the utterance from the user (in the present embodiment, the inquiry about the balance). Furthermore, the identification target includes the inquired user, that is, the inquired user is identified (authenticated) based on the third keyword in the speech from the user.
図6に示す例では、ユーザからの発話が「スミス、○○銀行の口座残高を教えて。」であるものとする。そして、この発話におけるキーワード「○○銀行」は第1のキーワードであり、本実施形態におけるインターネットバンキングサービスと関連付けられる。つまり、クエリ応答デバイス2は、ユーザからの発話に「○○銀行」が含まれると判定したことに応じて、対象のインターネットバンキングサービスを提供するサーバコンピュータ1を選択する。上記第1のキーワードとアプリケーションサービスとの関連付け(第1の対応関係)は、学習データに記憶される。
In the example shown in FIG. 6, it is assumed that the utterance from the user is "Smith, tell me the account balance of the ○○ bank." Then, the keyword “○○ bank” in this utterance is the first keyword, and is associated with the Internet banking service in the present embodiment. That is, the query response device 2 selects the server computer 1 that provides the target internet banking service, in response to the determination that the “soo bank” is included in the speech from the user. The association (first correspondence relationship) between the first keyword and the application service is stored in learning data.
発話におけるキーワード「残高」は第2のキーワードであり、本実施形態における残高に関する問い合わせの処理要求と関連付けられる。つまり、クエリ応答デバイス2は、ユーザからの発話に「残高」が含まれると判定したことに応じて、残高に関する問い合わせ要求であることを示す情報を含むクエリ電文を生成する。上記第2のキーワードと処理要求との関連付け(第2の対応関係)は、学習データに記憶される。
The keyword "balance" in the utterance is the second keyword, and is associated with the processing request for the balance related inquiry in the present embodiment. That is, the query response device 2 generates a query message including information indicating that it is an inquiry request regarding the balance, in response to the determination that the “speech” is included in the utterance from the user. The association (second correspondence relationship) between the second keyword and the processing request is stored in learning data.
発話におけるキーワード「スミス」は第3のキーワードであり、本実施形態における問い合わせたユーザと関連付けられる。このキーワードは、クエリ応答デバイス2から送信されるクエリ電文に含まれ、サーバコンピュータ1においてこのキーワードに基づいてユーザの認証が行われる。
The keyword "Smith" in the utterance is the third keyword, and is associated with the inquired user in the present embodiment. This keyword is included in the query message sent from the query response device 2 and the server computer 1 authenticates the user based on this keyword.
次に、図7を参照して、本発明の第2の実施形態に係るクエリ応答システムが実行する処理を説明する。第2の実施形態は、第1の実施形態と処理が重複する部分を有するので、異なる部分のみを説明する。図7におけるステップの処理のうち、以下の表1に示すステップの処理は、図5で説明したステップの処理に対応するので、説明を省略する。
Next, processing performed by the query response system according to the second embodiment of the present invention will be described with reference to FIG. The second embodiment has a portion in which the processing overlaps with the first embodiment, so only the different portion will be described. Among the processes of the steps in FIG. 7, the processes of the steps shown in Table 1 below correspond to the processes of the steps described in FIG.
ステップS702では、ユーザAは、クエリ応答デバイス2に対し、例えば、「スミス、○○銀行の口座残高を教えて。」と発話する。ステップS704では、クエリ応答デバイス2の制御装置21は、ステップS703で分析されたキーワードから、学習データに記憶された第1の対応関係に従って、対応するアプリケーションサービスを特定する。つまり、第1のキーワードである「○○銀行」がユーザAからの発話に含まれていると判定したことに応じて、インターネットバンキングサービスを提供するサーバコンピュータ1を特定する。
In step S702, the user A utters, for example, "Smith, tell me the account balance of the bank OO." To the query response device 2. In step S704, the control device 21 of the query response device 2 specifies the corresponding application service from the keyword analyzed in step S703 according to the first correspondence stored in the learning data. That is, in response to having determined that the first keyword "○○ bank" is included in the speech from the user A, the server computer 1 providing the internet banking service is specified.
ステップS705では、制御装置21は、ステップS703で分析されたキーワードから、学習データに記憶された第2の対応関係に従って、ユーザAからの要求の種別を決定する。そして、その要求の種別を示す情報、ステップS703で分析されたキーワードを含むクエリ要求を生成する。クエリ電文は、通信装置24によってサーバコンピュータ1に送信される。
In step S705, the control device 21 determines the type of request from the user A according to the second correspondence relationship stored in the learning data, from the keyword analyzed in step S703. Then, a query request including information indicating the type of the request and the keyword analyzed in step S703 is generated. The query message is transmitted to the server computer 1 by the communication device 24.
ステップS706では、サーバコンピュータ1の制御装置11がユーザデータテーブル300から対象のデータレコードを取得することになるが、図8を参照して、第2の実施形態におけるユーザデータテーブル800のデータ構造の例を説明する。
In step S706, the control device 11 of the server computer 1 acquires the target data record from the user data table 300. Referring to FIG. 8, the data structure of the user data table 800 in the second embodiment will be described. An example will be described.
図8に示すように、第2の実施形態に係るユーザデータテーブル800は、図3で説明したユーザデータテーブルに含まれるデータ項目に加え、認証キーワードを含む。認証キーワードは、ユーザによって任意のキーワードの値が設定される。
As shown in FIG. 8, the user data table 800 according to the second embodiment includes an authentication keyword in addition to the data items included in the user data table described in FIG. 3. The authentication keyword is set by the user to an arbitrary keyword value.
図7の説明に戻り、ステップS709では、ステップS706で取得したデータレコードに含まれるデバイス名に加え、認証キーワードに設定されたキーワードが、ステップS705で送信されたクエリ電文に含まれるかを判定する。つまり、ユーザAからの発話が予め設定したキーワードを含んでいるかが判定される。この処理によって、予め所定のキーワードを設定し、そのキーワードを発話に含ませることによって、ユーザAの認証を行うことができる。
Referring back to FIG. 7, in step S709, in addition to the device name included in the data record acquired in step S706, it is determined whether the keyword set in the authentication keyword is included in the query message transmitted in step S705. . That is, it is determined whether the speech from the user A includes a preset keyword. By this process, the user A can be authenticated by setting a predetermined keyword in advance and including the keyword in the utterance.
なお、本実施形態では、第3のキーワードの一致判定をサーバコンピュータ1において実行しているが、そのような形式に限定されない。例えば、クエリ応答デバイス2において、ユーザからの対話に設定した認証キーワードが含まれるかを判定してもよい。この場合、クエリ応答デバイス2の学習データに、クライアントコンピュータ3から取得したBDアドレスなどのデバイス識別子とキーワードとの対応関係が記憶され、ユーザからの対話に含まれるキーワードが学習データに従って判定される。
In the present embodiment, the determination of the third keyword match is executed in the server computer 1, but the present invention is not limited to such a format. For example, the query response device 2 may determine whether the authentication keyword set in the dialog from the user is included. In this case, the correspondence between the device identifier such as the BD address obtained from the client computer 3 and the keyword is stored in the learning data of the query response device 2, and the keyword included in the dialog from the user is determined according to the learning data.
以上のように、本発明に第2の実施形態に係るクエリ応答システムを説明した。本実施形態に係るクエリ応答システムによると、さらにセキュリティが確保された方式で秘密情報を交換することができる。また、第1の実施形態で説明した方式と同様に、ユーザによる操作を必要としないでユーザ認証を行うことができる。
As described above, the query response system according to the second embodiment of the present invention has been described. According to the query response system according to the present embodiment, secret information can be exchanged in a system in which security is further ensured. Also, as in the method described in the first embodiment, user authentication can be performed without requiring an operation by the user.
<その他の実施形態>
上述した構成に加え、クライアントコンピュータ3が予め定められた位置の範囲内にあるかを判定することによって、ユーザからの発話に対する応答を制限してもよい。この場合の処理の具体例を以下に示す。 <Other Embodiments>
In addition to the configuration described above, the response to the speech from the user may be limited by determining whether theclient computer 3 is within a predetermined position range. The specific example of the process in this case is shown below.
上述した構成に加え、クライアントコンピュータ3が予め定められた位置の範囲内にあるかを判定することによって、ユーザからの発話に対する応答を制限してもよい。この場合の処理の具体例を以下に示す。 <Other Embodiments>
In addition to the configuration described above, the response to the speech from the user may be limited by determining whether the
前提として、図5で説明したステップS507(第2の実施形態では、ステップS707)でクライアントコンピュータ3から送信される接続確認応答電文に、クライアントコンピュータ3の位置情報を含める。この場合、クライアントコンピュータ3は、GPS受信機を備える必要があり、GPS受信機が衛星から受信したGPS信号に基づいて位置情報が生成される。
As a premise, the connection confirmation response message transmitted from the client computer 3 in step S 507 (in the second embodiment, step S 707) described in FIG. 5 includes the position information of the client computer 3. In this case, the client computer 3 needs to have a GPS receiver, and the position information is generated based on the GPS signal received from the satellite by the GPS receiver.
また、上述したユーザデータテーブル300(第2の実施形態では、ユーザデータテーブル800)に、予め定められた範囲を示す位置情報を記憶する。そして、サーバコンピュータ1において、クライアントコンピュータ3から送信される接続確認応答電文に含まれる位置情報が、ユーザデータテーブル300の対象のデータレコードに含まれる位置情報の範囲内にあるかを判定する。
Further, position information indicating a predetermined range is stored in the above-described user data table 300 (in the second embodiment, the user data table 800). Then, the server computer 1 determines whether the position information included in the connection confirmation response message transmitted from the client computer 3 is within the range of the position information included in the target data record of the user data table 300.
本実施形態に係るクエリ応答システムによると、ユーザが指定した範囲内においてのみ、発話に対する応答が音声出力されない。このことは、例えば、クエリ応答デバイス2およびクライアントコンピュータ3のいずれもが盗難などにあった場合でも、指定した位置でのみの利用に制限されることから、第三者が残高を盗聴するような行為を防止することができる。
According to the query response system according to the present embodiment, the response to the utterance is not output in voice only within the range designated by the user. This is because, for example, even if both the query response device 2 and the client computer 3 are stolen etc., it is restricted to use only at the designated position, so that a third party eavesdrops on the balance. It can prevent the act.
また、上記実施形態では、クエリ応答デバイス2が、ユーザから発話(つまり、音声)によるクエリに応答して音声の形式で応答を返却する例を説明したが、そのような実施形態に限定されない。例えば、ユーザからのクエリは、音声以外の任意の媒体でのユーザからの入力が含まれてもよい。また、返却される応答も、音声以外の任意の媒体での応答の出力が含まれてもよい。いずれの場合も、クエリ応答デバイス2は、音声以外の媒体での入力を受け付け、応答を出力するマルチメディア応答デバイスとして実装されることになる。
In the above embodiment, an example is described in which the query response device 2 returns a response in the form of voice in response to a query from a user uttered (that is, voice). However, the present invention is not limited to such embodiment. For example, the query from the user may include input from the user on any medium other than voice. The returned response may also include the output of the response on any medium other than voice. In any case, the query response device 2 will be implemented as a multimedia response device that accepts input on media other than voice and outputs a response.
ユーザからの入力は、例えば、クエリ応答デバイス2に備えられた物理ボタンの押下、タッチスクリーンへの接触操作(タッチおよびスライドなど)、クエリ応答デバイス2全体に対する触覚的操作、ならびにクエリ応答デバイス2への視覚的入力などの任意の媒体での入力を含んでもよい。
The input from the user may be, for example, pressing of a physical button provided on the query response device 2, touch operation on a touch screen (touch and slide, etc.), tactile operation on the entire query response device 2, and the query response device 2. It may include input on any medium, such as visual input of
クエリ応答デバイス2全体に対する触覚的操作は、例えば、ユーザがクエリ応答デバイス2を所定の方向に傾ける操作、および/または移動させる操作などを含んでもよい。いずれの操作も、クエリ応答デバイス2がデバイスの傾きおよび移動の角速度および加速度などを検出することによって判定される。デバイスの傾きおよび移動のいずれについても、所定の方向、所定の距離および/または所定の速度の組み合せとクエリの内容とが関連付けられて、クエリ応答デバイス2がクエリを判定してもよい。例えば、ユーザがクエリ応答デバイス2を右45度に傾ける操作は、口座の残高を問い合わせるクエリに対応する。
The tactile operation on the entire query response device 2 may include, for example, an operation of tilting and / or moving the query response device 2 in a predetermined direction. Both operations are determined by the query response device 2 detecting the angular velocity and acceleration of the tilt and movement of the device. With regard to any of the tilt and movement of the device, the combination of a predetermined direction, a predetermined distance, and / or a predetermined speed may be associated with the content of the query, and the query response device 2 may determine the query. For example, the operation in which the user tilts the query response device 2 45 degrees to the right corresponds to a query for inquiring the balance of the account.
クエリ応答デバイス2への視覚的入力は、例えば、ユーザまたはユーザの一部(顔など)がクエリ応答デバイス2の所定の範囲に近づく動作、ならびに所定の色および輝度をクエリ応答デバイス2にかざす動作などを含んでもよい。いずれの動作もクエリ応答デバイスユーザのデバイスに対する近接度および光などを検出することによって判定される。それらの動作のいずれについても、所定の近接度、色および輝度またはそれらのいずれかの組み合わせとクエリの内容とが関連付けられて、クエリ応答デバイス2がクエリを判定してもよい。例えば、ユーザが黒い服を身に着けてクエリ応答デバイス2の50センチ以内に近づく動作は、口座の残高を問い合わせるクエリに対応する。
The visual input to the query response device 2 is, for example, an operation in which the user or a part of the user (such as a face) approaches a predetermined range of the query response device 2 and an operation in which the predetermined color and brightness are held over the query response device 2 Etc. may be included. Both actions are determined by detecting the proximity of the query response device user to the device, light, and the like. For any of these operations, the query response device 2 may determine the query, with predetermined proximity, color and / or intensity combinations associated with the content of the query. For example, an operation in which the user wears black clothes and approaches within 50 cm of the query response device 2 corresponds to a query for querying the balance of the account.
上記いずれの場合も、クエリ応答デバイス2は、入力装置としてタッチスクリーンおよびカメラなどを備え、検出素子として近接センサ、加速度センサ、角速度センサ、および光学センサなどを備える。
In any of the above cases, the query response device 2 includes a touch screen and a camera as input devices, and includes a proximity sensor, an acceleration sensor, an angular velocity sensor, an optical sensor and the like as detection elements.
応答の出力は、例えば、クエリ応答デバイス2のディスプレイ上の視覚的表示などの任意の媒体での出力を含んでもよい。この場合、クエリ応答デバイス2は、出力装置としてディスプレイなどを備える。
The output of the response may include, for example, an output on any medium, such as a visual display on the display of the query response device 2. In this case, the query response device 2 includes a display or the like as an output device.
上述したクエリ入力は、クエリ応答デバイス2の代わりにクライアントデバイス3によって受け付けられてもよい。例えば、クライアントコンピュータ3が傾き操作などを検出すると、その操作に対応する値(角速度など)をクエリ応答デバイス2に送信し、クエリ応答デバイス2が操作に対応するクエリ内容を判定してもよい。また、クライアントコンピュータ3がクエリ内容を判定してもよい。
The query input described above may be accepted by the client device 3 instead of the query response device 2. For example, when the client computer 3 detects a tilt operation or the like, a value (angular velocity or the like) corresponding to the operation may be transmitted to the query response device 2, and the query response device 2 may determine the query content corresponding to the operation. Also, the client computer 3 may determine the contents of the query.
応答出力も同様に、クエリ応答デバイス2の代わりにクライアントデバイス3によって出力されてもよい。例えば、クエリ応答デバイス2が応答を生成すると、その応答をクライアントコンピュータ3に送信し、ディスプレイなどに表示してもよい。
The response output may also be output by the client device 3 instead of the query response device 2. For example, when the query response device 2 generates a response, the response may be transmitted to the client computer 3 and displayed on a display or the like.
さらに、上記実施形態では、単一のクエリ応答デバイス2と単一のクライアントコンピュータ3とのペアリング情報をサーバコンピュータ1に登録する例を説明したが、そのペアリング情報は、複数のペアを関連付けて登録されてもよい。例えば、N個のクライアントコンピュータ3がM個のクエリ応答デバイス2と関連付けられてユーザデータテーブル300または800に登録されてもよい。
Furthermore, in the above embodiment, an example in which pairing information of a single query response device 2 and a single client computer 3 is registered in the server computer 1 has been described, but the pairing information associates a plurality of pairs. May be registered. For example, N client computers 3 may be associated with M query response devices 2 and registered in the user data table 300 or 800.
2つのクライアントコンピュータ3のそれぞれがクエリ応答デバイス2とペアリングされる例を説明する。この例では、2つのクライアントコンピュータ3双方のクライアントコンピュータ通知IDが、ユーザIDおよび利用デバイス名と関連付けてユーザデータテーブル300または800に登録される。認証処理では、サーバコンピュータ1から2つのクライアントコンピュータ3の双方に接続確認要求電文が送信され、双方がクエリ応答デバイス2と接続しているかを確認する。
An example in which each of two client computers 3 is paired with the query response device 2 will be described. In this example, client computer notification IDs of both of the two client computers 3 are registered in the user data table 300 or 800 in association with the user ID and the use device name. In the authentication process, a connection confirmation request message is sent from the server computer 1 to both of the two client computers 3, and it is confirmed whether both are connected to the query response device 2.
このような構成とすることによって、例えば、2つのクライアントコンピュータ3およびクエリ応答デバイス2のうちのいずれかのデバイスが盗難などにあっても、3つのデバイスが揃わない限り、第三者による盗聴を防止することができる。ひいては、セキュリティをさらに強固にすることができる。
With such a configuration, for example, even if one of the two client computers 3 and the query response device 2 is stolen or the like, eavesdropping by a third party can be performed unless the three devices are aligned. It can be prevented. As a result, security can be further strengthened.
上記実施形態で説明した構成はいずれも例示的なものにすぎず、そのような構成に限定されるものではない。例えば、本実施形態では、サーバコンピュータ1とクエリ応答デバイス2とが連携することによって発話に対する応答を出力しているが、スピーカを実装した単独のコンピュータデバイスによって上記処理を実装してもよい。また、上記説明したサーバコンピュータ1によって実行される処理のうちの一部が、本発明の基本的な概念を逸脱することなく、クエリ応答デバイス2で実行されてもよく、逆もまたそうである。
All the configurations described in the above embodiments are merely illustrative, and the present invention is not limited to such configurations. For example, in the present embodiment, although the response to the utterance is output by the server computer 1 and the query response device 2 cooperating with each other, the above process may be implemented by a single computer device on which a speaker is implemented. Also, some of the processing performed by the server computer 1 described above may be performed by the query response device 2 without departing from the basic concept of the present invention, and vice versa .
また、上記説明したクエリ応答デバイス(つまり、スピーカおよびマイクを実装したデバイス)は、AIスピーカとして実装されるものとして説明したが、クエリ応答デバイスは、学習機能を必ずしも有する必要はない。学習機能を有さないでクエリ応答デバイスを実装する場合、予め定められたキーワードと要求処理などを関連付けた対応関係を記憶し、その対応関係に従って処理が実行されることになる。
In addition, although the query response device described above (that is, the device on which the speaker and the microphone are mounted) has been described as being implemented as an AI speaker, the query response device does not necessarily have to have a learning function. When a query response device is implemented without having a learning function, a correspondence relationship in which a predetermined keyword is associated with request processing and the like is stored, and processing is executed according to the correspondence relationship.
上記説明したユーザデータテーブル300などのデータ構造は、例示的なものにすぎず、その他の必要なデータ項目が含まれてもよく、また本発明の基本的な概念から逸脱することなく代替的なデータテーブルが使用されてもよい。また、上記実施形態で説明した処理の順序は、必ずしも説明した順序で実行される必要がなく、任意の順序で実行されてもよい。さらに、本発明の基本的な概念から逸脱することなく、追加のステップが新たに加えられてもよく、説明したステップが省略されてもよい。
The data structures, such as user data table 300 described above, are exemplary only, and may include other necessary data items, and may be alternative without departing from the basic concept of the present invention. Data tables may be used. In addition, the order of the processes described in the above embodiment need not necessarily be performed in the order described, and may be performed in an arbitrary order. Furthermore, additional steps may be added newly or the described steps may be omitted without departing from the basic concept of the present invention.
1 サーバコンピュータ
2 クエリ応答デバイス
3 クライアントコンピュータ
4 ネットワーク
300 ユーザデータテーブル
400 口座データテーブル
800 ユーザデータテーブル
A ユーザ 1server computer 2 query response device 3 client computer 4 network 300 user data table 400 account data table 800 user data table A user
2 クエリ応答デバイス
3 クライアントコンピュータ
4 ネットワーク
300 ユーザデータテーブル
400 口座データテーブル
800 ユーザデータテーブル
A ユーザ 1
Claims (5)
- サーバコンピュータおよびクエリ応答デバイスを備えたコンピュータシステムであって、
前記クエリ応答デバイスは、入力を受信し、前記入力に基づいてクエリ電文を生成してサーバコンピュータに送信し、
前記サーバコンピュータは、
クライアントコンピュータに接続確認要求電文を送信し、
前記クライアントコンピュータから第1の識別子を含む接続確認応答電文を受信し、前記第1の識別子は、前記クエリ応答デバイスを識別し、
前記第1の識別子と記憶された第2の識別子とが一致するかを判定することによって、前記クライアントコンピュータと前記クエリ応答デバイスとの間の接続が確立されていると判定し、
前記第1の識別子と前記第2の識別子とが一致したと判定したことに応答して、前記クエリ電文に対応する応答電文を生成して前記クエリ応答デバイスに送信し、
前記クエリ応答デバイスは、前記応答電文を出力する
ことを特徴とするコンピュータシステム。 A computer system comprising a server computer and a query response device, comprising:
The query response device receives an input, generates a query message based on the input, and sends it to a server computer.
The server computer is
Send a connection confirmation request message to the client computer,
Receiving a connection acknowledgment message including a first identifier from the client computer, the first identifier identifying the query response device;
Determining that a connection between the client computer and the query response device has been established by determining whether the first identifier and the stored second identifier match;
In response to determining that the first identifier and the second identifier match, a response message corresponding to the query message is generated and transmitted to the query response device.
The computer system, wherein the query response device outputs the response message. - 前記クライアントコンピュータが所定の範囲内にあることを前記クエリ応答デバイスが検出することによって、前記第1の識別子が前記クライアントコンピュータによって取得されることを特徴とする請求項1に記載のコンピュータシステム。 The computer system according to claim 1, wherein the first identifier is obtained by the client computer by the query response device detecting that the client computer is within a predetermined range.
- 前記クライアントコンピュータと前記クエリ応答デバイスとの間の接続は、Bluetooth通信規格に従って確立されることを特徴とする請求項1または2に記載のコンピュータシステム。 The computer system according to claim 1 or 2, wherein a connection between the client computer and the query response device is established according to a Bluetooth communication standard.
- サーバコンピュータおよびクエリ応答デバイスを備えたコンピュータシステムによって実行される方法であって、
前記クエリ応答デバイスが、入力を受信し、前記入力に基づいてクエリ電文を生成してサーバコンピュータに送信するステップと、
前記サーバコンピュータが、
クライアントコンピュータに接続確認要求電文を送信するステップと、
前記クライアントコンピュータから第1の識別子を含む接続確認応答電文を受信するステップであって、前記第1の識別子は、前記クエリ応答デバイスを識別する、ステップと、
前記第1の識別子と記憶された第2の識別子とが一致するかを判定することによって、前記クライアントコンピュータと前記クエリ応答デバイスとの間の接続が確立されていると判定するステップと、
前記第1の識別子と前記第2の識別子とが一致したと判定したことに応答して、前記クエリ電文に対応する応答電文を生成し、前記クエリ応答デバイスに送信するステップと、
前記クエリ応答デバイスが、前記応答電文を出力するステップと
を備えたことを特徴とする方法。 A method performed by a computer system comprising a server computer and a query response device, the method comprising:
The query response device receiving an input, generating a query message based on the input and sending it to a server computer;
The server computer is
Sending a connection confirmation request message to the client computer;
Receiving from the client computer a connection acknowledgment message including a first identifier, wherein the first identifier identifies the query response device;
Determining that a connection between the client computer and the query response device has been established by determining whether the first identifier and the stored second identifier match.
Generating a response telegram corresponding to the query telegram in response to determining that the first identifier and the second identifier match, and transmitting the response telegram to the query response device;
And D. the query response device outputting the response message. - コンピュータ実行可能命令を含むコンピュータプログラムであって、前記コンピュータ実行可能命令は、コンピュータによって実行されると、前記コンピュータに請求項4に記載の方法を実行させることを特徴とするコンピュータプログラム。 A computer program comprising computer executable instructions, wherein the computer executable instructions, when executed by a computer, cause the computer to perform the method of claim 4.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2018534751A JP6694069B2 (en) | 2017-11-30 | 2017-11-30 | Query response system and query response method |
PCT/JP2017/043192 WO2019106811A1 (en) | 2017-11-30 | 2017-11-30 | Query response system and query response method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2017/043192 WO2019106811A1 (en) | 2017-11-30 | 2017-11-30 | Query response system and query response method |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019106811A1 true WO2019106811A1 (en) | 2019-06-06 |
Family
ID=66665508
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2017/043192 WO2019106811A1 (en) | 2017-11-30 | 2017-11-30 | Query response system and query response method |
Country Status (2)
Country | Link |
---|---|
JP (1) | JP6694069B2 (en) |
WO (1) | WO2019106811A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2022509837A (en) * | 2018-11-28 | 2022-01-24 | ビザ インターナショナル サービス アソシエーション | Audience certification |
JP7629439B2 (en) | 2019-07-08 | 2025-02-13 | キャピタル・ワン・サービシーズ・リミテッド・ライアビリティ・カンパニー | Authenticating voice transactions using payment cards |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005266946A (en) * | 2004-03-16 | 2005-09-29 | Ricoh Co Ltd | Information processing apparatus, information processing method, information processing program, and recording medium |
JP2010226336A (en) * | 2009-03-23 | 2010-10-07 | Denso It Laboratory Inc | Authentication method and authentication apparatus |
JP2015099515A (en) * | 2013-11-20 | 2015-05-28 | ヤフー株式会社 | Information providing apparatus, information providing method, terminal device, and information providing program |
JP2017167884A (en) * | 2016-03-17 | 2017-09-21 | 株式会社エヌ・ティ・ティ・データ | Personal authentication system, personal authentication device, personal authentication method, and program |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5531351B2 (en) * | 2008-06-27 | 2014-06-25 | Kii株式会社 | Social mobile search |
JP2016126670A (en) * | 2015-01-07 | 2016-07-11 | Kddi株式会社 | Risk-based authentication method and system using portable terminal |
-
2017
- 2017-11-30 JP JP2018534751A patent/JP6694069B2/en active Active
- 2017-11-30 WO PCT/JP2017/043192 patent/WO2019106811A1/en active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005266946A (en) * | 2004-03-16 | 2005-09-29 | Ricoh Co Ltd | Information processing apparatus, information processing method, information processing program, and recording medium |
JP2010226336A (en) * | 2009-03-23 | 2010-10-07 | Denso It Laboratory Inc | Authentication method and authentication apparatus |
JP2015099515A (en) * | 2013-11-20 | 2015-05-28 | ヤフー株式会社 | Information providing apparatus, information providing method, terminal device, and information providing program |
JP2017167884A (en) * | 2016-03-17 | 2017-09-21 | 株式会社エヌ・ティ・ティ・データ | Personal authentication system, personal authentication device, personal authentication method, and program |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2022509837A (en) * | 2018-11-28 | 2022-01-24 | ビザ インターナショナル サービス アソシエーション | Audience certification |
JP7361118B2 (en) | 2018-11-28 | 2023-10-13 | ビザ インターナショナル サービス アソシエーション | audible authentication |
JP7629439B2 (en) | 2019-07-08 | 2025-02-13 | キャピタル・ワン・サービシーズ・リミテッド・ライアビリティ・カンパニー | Authenticating voice transactions using payment cards |
Also Published As
Publication number | Publication date |
---|---|
JPWO2019106811A1 (en) | 2019-12-12 |
JP6694069B2 (en) | 2020-05-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11902254B2 (en) | Blockchain joining for a limited processing capability device and device access security | |
CN109076076B (en) | Two-factor authentication | |
US9730065B1 (en) | Credential management | |
CN106030598B (en) | Trust agent authentication method for mobile device | |
US9413758B2 (en) | Communication session transfer between devices | |
US20140230019A1 (en) | Authentication to a first device using a second device | |
WO2016061769A1 (en) | Verification information transmission method and terminal | |
US12076649B2 (en) | Apparatus and methods for simulating authentication within the metaverse | |
US20200034551A1 (en) | Systems and methods for providing interactions based on a distributed conversation database | |
KR20160129839A (en) | An authentication apparatus with a bluetooth interface | |
WO2016102336A1 (en) | Communication system comprising head wearable devices | |
JP2022511545A (en) | Proximity field information authentication method, proximity field information authentication device, electronic device and computer storage medium | |
JP7553665B2 (en) | Wireless connection system, information terminal, display device, and wireless connection method | |
WO2020031429A1 (en) | Terminal device, authentication server, control method for terminal device, authentication method, and program | |
US20250294360A1 (en) | Communication, Authentication, and Validation Using LoRaWAN Protocol | |
JP6694069B2 (en) | Query response system and query response method | |
US20250150798A1 (en) | System and method for determining device status using lorawan | |
JP2017102842A (en) | Personal identification system, personal identification information output system, authentication server, personal identification method, personal identification information output method, and program | |
KR20210127383A (en) | Electronic device for sending cryptocurrency to blockchain account and method of operating the same | |
KR20170111942A (en) | Electronic commercial transaction authentication method and system by specific infomation related otp | |
CN113506108A (en) | Account management method, device, terminal and storage medium | |
US12014348B2 (en) | Validating transactions between entities using LoRaWAN protocol | |
US12141780B2 (en) | Mobile device transaction processing system and method using LoRaWAN communications | |
US20230041559A1 (en) | Apparatus and methods for multifactor authentication | |
TWM641570U (en) | Contactless signing system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2018534751 Country of ref document: JP Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17933424 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17933424 Country of ref document: EP Kind code of ref document: A1 |