WO2019177101A1 - Ai speaker system, method for controlling ai speaker system, and program - Google Patents

Abstract

A settlement account notifying unit 117 refers to an account database in a storage unit 118, identifies a settlement user account associated with a speaker user account identified by means of an identifying unit 113, and issues a notification of the speaker user account to a settlement system 3. Furthermore, the settlement account notifying unit 117 notifies the settlement system 3 of the price of service provision by a service providing device 2. A service account notifying unit 116 refers to the account database in the storage unit 118, identifies a service user account associated with the speaker user account identified by means of the identifying unit 113, and issues a notification of the service user account to the service providing device 2. A learning unit 119 learns a correspondence relationship between the speaker user account, the service user account, and the settlement user account by means of a learning algorithm such as machine learning.

Description

AI speaker system, control method of AI speaker system, and program

The present invention relates to a technique for controlling an AI speaker system.

AI (Artificial Intelligence) speakers have been developed that accept user voice instructions and perform actions desired by the user. In Patent Document 1, when performing a voice conversation with a user, it is more effective when the voice conversation is made into a natural content according to the user or when the conversation is used as a means for achieving a certain purpose. A technique for promoting dialogue is disclosed.

JP 2004-021121 A

With the spread of AI speakers, it is expected that all services desired by users will be realized via the AI speaker system. However, there are various services desired by users including those of third parties, and user accounts are individually set for these services. For this reason, when a user uses each service, an account must be individually input. The troublesome input of such an account is the same in the case of performing a settlement service that settles the price of the service used by the user. Therefore, it is practically difficult for a user to easily use a paid service via an AI speaker system.

An object of the present invention is to realize an AI speaker system that can easily provide a pay service desired by a user.

The present invention relates to a correspondence relationship between a speaker user account that identifies a user who speaks in an AI speaker, a service user account that identifies a user in a service providing apparatus that provides a service, and a settlement user account that identifies a user in a settlement system. Learning means for learning; specifying means for specifying the speaker user account based on a user's voice input in the AI speaker; and the service user account associated with the specified speaker user account, the service user account A service account notifying means for notifying the providing apparatus, and a payment account notifying means for notifying the payment system of the payment user account associated with the specified speaker user account A To provide a speaker system.

It may be configured to include a determination unit that determines the priority order of the plurality of service providing devices, and a presentation unit that presents the service providing device to the user who has spoken according to the determined priority order.

The determination unit may determine the priority based on an analysis result obtained by analyzing the meaning of the input user's voice.

The determination unit selects one or more service providing devices from a plurality of the service providing devices based on the analysis result, and the one or more selected service providing devices correspond to the specified speaker user account. The priority order may be determined according to user attributes.

The specifying unit may specify the speaker user account corresponding to the input voice print by referring to a database in which each user's voice print is associated with the speaker user account of the user. .

The learning unit may perform learning related to the priority order, and the determination unit may determine the priority order using the learning result.

The learning unit may perform the learning according to a date and time or a place where a user's voice is input through the AI speaker.

The learning means may perform the learning according to the user's emotion estimated from the input user's voice.

The payment account notification means may notify the payment system of the value of service provided by the service providing device.

The present invention relates to a correspondence relationship between a speaker user account that identifies a user who speaks in an AI speaker, a service user account that identifies a user in a service providing apparatus that provides a service, and a settlement user account that identifies a user in a settlement system. A learning step of learning, a specifying step of specifying the speaker user account based on a user's voice input in the AI speaker, and the service user account associated with the specified speaker user account, the service user account A service account notifying step for notifying the providing apparatus, and a settlement account notifying step for notifying the settlement system of the settlement user account associated with the identified speaker user account. To provide a control method for AI speaker system and a flop.

The present invention provides a computer with a speaker user account for identifying a user who has spoken at an AI speaker, a service user account for identifying a user at a service providing apparatus that provides a service, and a payment user account for identifying a user at a payment system. A learning step of learning correspondence, a specifying step of specifying the speaker user account based on a user's voice input in the AI speaker, and the service user account associated with the specified speaker user account A service account notifying step for notifying the service providing apparatus, and a settlement account for notifying the settlement system of the settlement user account associated with the identified speaker user account. It provides a program for executing a count notification step Toto.

According to the present invention, an AI speaker system that can easily provide various services desired by a user can be realized.

The figure which shows the structure of the AI speaker system 9 which concerns on this embodiment. The figure which shows an example of the hardware constitutions of AI speaker 1. The figure which shows the voiceprint database memorize | stored in the memory | storage part. The figure which shows the account database memorize | stored in the memory | storage part. The figure which shows the user attribute database memorize | stored in the memory | storage part. The figure which shows the functional structure of the AI speaker. The sequence chart which shows the flow of the basic operation | movement of AI speaker system 9. FIG. The sequence chart which shows the flow of learning operation | movement of AI speaker system 9. FIG.

DESCRIPTION OF SYMBOLS 1 ... AI speaker, 11 ... Control part, 12 ... Memory | storage part, 12 ... Communication part, 14 ... Input part, 15 ... Output part, 111 ... Input part, 112 ... Analysis part, 113 ... Identification part, 114 ... Determination part, DESCRIPTION OF SYMBOLS 115 ... Presentation part, 116 ... Service account notification part, 117 ... Payment account notification part, 118 ... Memory | storage part, 119 ... Learning part, 2 ... Service provision apparatus, 3 ... Payment system, 4 ... Network, 9 ... AI speaker system.

<Embodiment>
<Overall configuration of AI speaker system>
FIG. 1 is a diagram showing a configuration of an AI speaker system 9 according to the present embodiment. The AI speaker system 9 includes an AI speaker 1, a service providing device group 2, a settlement system 3, and a network 4 that connects these devices so that they can communicate with each other. The AI speaker system 9 may include a plurality of AI speakers 1, a service providing device 2, a payment system 3, and a network 4.

The AI speaker 1 is a device that inputs a user's voice through a microphone or the like and outputs a voice through a dynamic speaker or an electrostatic speaker. In addition to the name AI speaker, for example, an information processing device called a smart speaker or a home speaker It is. The service providing apparatus group 2 is one or more information processing apparatuses that provide a service requested from the AI speaker 1 via the network 4. For example, the service providing device group 2 provides a service for selling products to the user. The payment system 3 includes, for example, a credit company and bank system, and performs a payment process related to the consideration of selling a product to a user. In the service providing apparatus 2 and the payment system 3, a user account is individually assigned to one user. That is, a user account assigned in a certain service can identify the user only in that service. The network 4 is a communication line that connects the AI speaker 1, the service providing device group 2, and the payment system 3 so as to be communicable, and is, for example, the Internet.

<Configuration of AI speaker>
FIG. 2 is a diagram illustrating an example of the configuration of the AI speaker 1. The AI speaker 1 includes a control unit 11, a storage unit 12, a communication unit 13, an input unit 14, and an output unit 15.

The control unit 11 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory), and a computer program (hereinafter simply referred to as a program) in which the CPU is stored in the ROM and the storage unit 12. Are read out and executed to control each part of the AI speaker 1.

The communication unit 13 is a communication circuit connected to the network 4 by wire or wireless. The AI speaker 1 exchanges information with the service providing apparatus 2 connected to the network 4 by the communication unit 13.

The input unit 14 is a microphone or the like for inputting sound, and sends a sound signal indicating the input sound to the control unit 11.

The output unit 15 is, for example, a dynamic speaker or an electrostatic speaker, and emits sound according to a signal instructed by the control unit 11.

The storage unit 12 is a large-capacity storage unit such as a solid state drive or a hard disk drive, and stores various programs and data read by the CPU of the control unit 11. The storage unit 12 stores, for example, a voiceprint database (hereinafter referred to as DB), a user account DB, and a user attribute DB.

FIG. 3 is a diagram showing the voiceprint DB stored in the storage unit 12. The voiceprint DB stores each user's voiceprint data in association with a speaker user account that is identification information for identifying the user in the AI speaker 1.

FIG. 4 is a diagram showing the account DB stored in the storage unit 12. The account DB identifies a speaker user account that is identification information for identifying the user in the AI speaker 1, a service (services A, B, and C) of each service providing device 2, and a service of the payment system 3. Service user accounts (service A user account, service B user account, service C user account, settlement user account), which are identification information to be stored, are stored in association with each other. That is, in the account DB, a speaker user account of a certain user, one or more service user accounts of the user, and a settlement user account of the user are associated with each other. The payment user account includes information for identifying a user and performing a payment process in the payment system 3, such as a credit card number or a bank account number. .

FIG. 5 is a diagram showing the user attribute DB stored in the storage unit 12. The user attribute DB stores a speaker user account, which is identification information for identifying the user in the AI speaker 1, and a user attribute of the user in association with each other. User attributes include, for example, age, sex, hobbies, search history in the network 4, browsing history in the network 4, purchase history of products or services in mail order sales via the network 4, and the like.

<Functional configuration of AI speaker>
FIG. 6 is a diagram showing a functional configuration of the AI speaker 1. When the control unit 11 of the AI speaker 1 reads and executes the program stored in the storage unit 12, the AI speaker 1 has the input unit 111, the analysis unit 112, the identification unit 113, the determination unit 114, the presentation unit 115, the service. Functions such as an account notification unit 116, a settlement account notification unit 117, a storage unit 118, and a learning unit 119 are realized.

The input unit 111 inputs a user's voice.

The analysis unit 112 analyzes the meaning of the user's voice input to the input unit 111 using, for example, a hidden Markov model.

The storage unit 118 stores a voice print DB in which each user's voice print data is associated with a speaker user account that is identification information for identifying the user in the AI speaker 1. In addition, the storage unit 118 has a learning result by the learning unit 119, which will be described later, specifically, a speaker user account for identifying a user who speaks in the AI speaker 1, and a service user for identifying a user in the service providing apparatus 2 that provides a service. An account DB including a correspondence relationship between an account and a settlement user account for identifying a user in the settlement system 3 is stored. In addition, the storage unit 118 stores a user attribute DB in which a speaker user account that is identification information for identifying the user in the AI speaker 1 and a user attribute of the user are associated with each other.

The identifying unit 113 identifies a speaker user account that identifies a user who has emitted the voice based on the voice input to the input unit 111. More specifically, the specifying unit 113 refers to the voice print DB in the storage unit 118 and specifies a speaker user account corresponding to the voice print of the input voice.

The determination unit 114 determines the priority order of the plurality of service providing apparatuses 2 based on the analysis result obtained by the analysis unit 112 analyzing the meaning of the input user's voice. More specifically, the determination unit 114 selects one or more service providing apparatuses 2 from the plurality of service providing apparatuses 2 based on the analysis result. Then, the determination unit 114 refers to the user attribute DB of the storage unit 118 and determines the priority order of the selected one or more service providing apparatuses 2 according to the user attribute corresponding to the speaker user account specified by the specifying unit 113. Determine.

The presentation unit 115 presents the service providing apparatus 2 to the uttered user according to the priority order determined by the determination unit 114.

The settlement account notifying unit 117 refers to the account DB of the storage unit 118, identifies the settlement user account associated with the speaker user account identified by the identifying unit 113, and notifies the settlement system 3 of the identified settlement user account. Further, the settlement account notifying unit 117 notifies the settlement system 3 of the price of service provision by the service providing apparatus 2.

The service account notifying unit 116 refers to the account DB in the storage unit 118, identifies the service user account associated with the speaker user account identified by the identifying unit 113, and notifies the service providing apparatus 2 of it.

The learning unit 119 uses a learning algorithm such as machine learning to identify a user who speaks at the AI speaker 1, a service user account that identifies the user at the service providing apparatus 2 that provides the service, and a payment system 3 Learn the correspondence with the payment user account that identifies the user.

<Operation of AI speaker>
The operation of the AI speaker is roughly classified into a learning operation for performing learning and a basic operation performed based on the learning. In the following, for easy understanding, the basic operation will be described first, and then the learning operation will be described.

<Basic operation of AI speaker>
FIG. 7 is a sequence chart showing a flow of basic operations of the AI speaker system 9. First, the input unit 111 receives an input of voice spoken by the user (step S11). This voice is, for example, “I want to buy a battery”, “I want to buy a T-shirt”, “I want to go to a resort”, or “I want to move”.

Next, the analysis unit 112 analyzes the meaning of the input user's voice (step S12). In this speech semantic analysis, various known semantic analysis algorithms may be used.

Next, the specifying unit 113 analyzes the input voice, generates voice print data, and collates it with the voice print data included in the voice print DB (step S13). In this voiceprint generation, various known voiceprint generation algorithms may be used. If there is a voice print DB that matches the voice print data input by the user, the identifying unit 113 refers to the speaker user account corresponding to the voice print data in the voice print DB, so that the user who emitted the voice Identify the speaker user account for. If there is no voice print DB that matches the voice print data input by the user, the specifying unit 113 performs a predetermined error process, and the process shown in FIG. 7 ends.

Next, the determination unit 114 selects the service providing apparatus 2 that provides a service corresponding to the meaning of the analyzed voice (step S14). For example, if the user's voice is “I want to buy a battery”, the determination unit 114 determines that the user wants to purchase a battery, and provides one or more services that provide a service for mail-ordering daily commodities. The providing device 2 is selected. For example, if the user's voice is “I want to buy a T-shirt”, the determination unit 114 determines that the user wants to purchase a T-shirt, and provides a service for mail-ordering clothes. One or more service providing apparatuses 2 are selected. For example, if the user's voice is “I want to go to a resort”, the determination unit 114 determines that the user wants to travel to a resort, and provides one or more services related to travel. The service providing device 2 is selected. Also, for example, if the user's voice is “I want to ask for moving”, the determination unit 114 determines that the user is looking for a moving company, and determines one or more service providing apparatuses 2 that provide the moving service. select.

Next, the determination unit 114 determines the priority order of the selected one or more service providing apparatuses 2 according to the user attribute corresponding to the specified speaker user account (step S15). For example, if the selected service is a daily commodities mail order and the user is a housewife, the determination unit 114 gives higher priority to the service providing apparatus 2 that provides the daily commodities mail order service in descending order of evaluation that the price is lower. Give. Further, for example, if the selected service is mail order of clothes and the user is a teenage male, the determination unit 114 performs the mail order sales service of clothes in descending order of evaluation from the male teenager. Is given a high priority. For example, if the selected service is a service related to travel and the user is a woman in her 20s, the determination unit 114 gives higher priority to the service providing apparatus 2 that provides services related to travel in descending order of evaluation from women in their 20s. Give a ranking. For example, if the selected service is a moving service and the user is a single living man, the determination unit 114 provides the moving service to the service providing apparatus 2 that performs the moving service in ascending order of the cost of the moving pack for living alone. Give high priority.

Next, the presenting unit 115 presents the service providing device to the user according to the determined priority (Step S16). Specifically, artificial speech that explains the service contents of the selected service providing apparatus 2 is generated and emitted in descending order of priority. For this reason, the presentation unit 115 obtains in advance the service contents of the service providing apparatus 2 from each service providing apparatus 2 or the like via the network 4.

When the user who has heard the presentation content specifies any one of the service providing apparatuses 2 by voice, the service account notifying unit 116 makes a request to the designated service providing apparatus 2 (step S17). This request includes a service user count for identifying the user in the service providing apparatus 2.

The service providing apparatus 2 performs processing according to this request (step S18). For example, the service providing apparatus 2 that performs mail order sales of daily commodities confirms that the service user account included in the request is included in the account database in the own apparatus, and then uses the voice regarding the order of the product by the voice from the user. Are exchanged via the AI speaker 1. At this time, the presentation unit 115 of the AI speaker 1 performs sound emission processing according to the audio data transmitted from the service providing apparatus 2.

When the user designates the product to be purchased by voice, and when the payment of the price is instructed by voice, the settlement account notification unit 117 makes a request to the settlement system 3 (step S19). This request includes a payment user account for identifying a user in the payment system 3 and a service provision price by the service providing apparatus 2.

The settlement system 3 performs a settlement process according to this request (step S20). That is, a process for withdrawing the consideration from the user's bank account at a predetermined time using the settlement user account is performed.

<AI speaker learning operation>
FIG. 8 is a sequence chart showing the flow of the learning operation of the AI speaker system 9. The process shown in the figure is executed over an initial period in which the user starts using the AI speaker 1, for example, and over a period in which the user uses the AI speaker 1, for example. The input unit 111 receives an input of voice spoken by the user (step S111). As in step S11 of FIG. 7, this voice includes, for example, “I want to buy a battery”, “I want to buy a T-shirt”, “I want to go to a resort”, and “I want to move”.

Next, the analysis unit 112 analyzes the meaning of the input user's voice using a well-known semantic analysis algorithm, similarly to step S12 (step S112).

Next, as in step S13, the specifying unit 113 analyzes the input voice using a known voiceprint generation algorithm to generate voiceprint data, and collates it with the voiceprint data included in the voiceprint DB (step S113). If there is a voice print DB that matches the voice print data input by the user, the identifying unit 113 refers to the speaker user account corresponding to the voice print data in the voice print DB, so that the user who emitted the voice Identify the speaker user account for. If there is no voice print DB that matches the voice print data input by the user, the specifying unit 113 performs a predetermined error process, and the process shown in FIG. 8 ends.

Next, as in step S14, the determination unit 114 selects the service providing apparatus 2 that provides a service corresponding to the meaning of the analyzed voice (step S114).

Next, as in step S15, the determination unit 114 determines the priority order of the selected one or more service providing apparatuses 2 according to the user attribute corresponding to the specified speaker user account (step S15).

Next, the presentation unit 115 presents the service providing apparatus to the user according to the determined priority order, similarly to step S16 (step S16).

When the user who listens to the presentation content designates any one of the service providing apparatuses 2 by voice, the service account notifying unit 116 prompts the user to input a service user account in the designated service providing apparatus 2. When the user inputs a service user account in the specified service providing apparatus 2, the service account notifying unit 116 receives this input (step S117).

Next, the service account notification unit 116 makes a request to the designated service providing apparatus 2 (step S118). This request includes a service user count for identifying the user in the service providing apparatus 2.

Here, the learning unit 119 learns the correspondence between the speaker user account specified in step S113 and the service user account input in step S117, and stores the correspondence in the storage unit 118 (account DB). (Step S119).

Meanwhile, the service providing apparatus 2 performs processing according to the request (Step S120). For example, the service providing apparatus 2 that performs mail order sales of daily commodities confirms that the service user account included in the request is included in the account database in the own apparatus, and then uses the voice regarding the order of the product by the voice from the user. Are exchanged via the AI speaker 1. At this time, the presentation unit 115 of the AI speaker 1 performs sound emission processing according to the audio data transmitted from the service providing apparatus 2.

When the product to be purchased is designated by voice and the payment for the price is instructed by voice, the settlement account notifying unit 117 prompts the user to input a settlement user account in the settlement system 3. When the user inputs a payment user account, the payment account notification unit 117 accepts this input (step S121).

Next, the settlement account notification unit 117 makes a request to the settlement system 3 (step S122). This request includes a payment user account for identifying a user in the payment system 3 and a service provision price by the service providing apparatus 2.

Here, the learning unit 119 learns the correspondence between the speaker user account specified in step S113 and the settlement user account input in step S121, and stores the correspondence in the storage unit 118 (account DB). (Step S123).

On the other hand, the payment system 3 performs a payment process in response to this request (step S124). That is, a process for withdrawing the consideration from the user's bank account at a predetermined time using the settlement user account is performed.

Through the above learning operation, a speaker user account for identifying a user who has spoken in the AI speaker 1, a service user account for identifying a user in the service providing apparatus 2 that provides the service, and a settlement user account for identifying the user in the settlement system 3 Is learned and stored in the storage unit 118 (account DB). In the basic operation described above, each process is executed based on the contents of the storage unit 118 (account DB) as the learning result.

According to the above embodiment, even if the service user account of the user is individually set in each service, the user desires via the AI speaker system without inputting the service user account each time. Service can be provided. Furthermore, it is possible to provide a pay service desired by the user via the AI speaker system without inputting the settlement user account each time. Furthermore, since these processes are automatically learned, high usability is realized.

<Modification>
The above is the description of the embodiment, but the contents of this embodiment can be modified as follows. Further, the following modifications may be combined.

In the embodiment, the learning unit 119 has learned the correspondence relationship between the speaker user account, the service user account, and the settlement user account, but may further learn about the priority order used by the determination unit. As described above, for example, if the user's voice is “I want to buy a battery”, the determination unit 114 determines that the user wants to purchase a battery, and provides a service for mail-ordering daily commodities. One or more service providing apparatuses 2 are selected. And the priority according to the attribute of the user was determined with respect to these one or more service providing apparatuses 2. The learning unit 119 learns the relationship between the user attribute and the priority of the service providing apparatus 2 using a learning algorithm such as machine learning. For example, when there is a service providing device 2 that is frequently selected by the user among the plurality of service providing devices 2 presented to the user, the learning unit 119 determines the user attribute of the user and the service providing device 2. It learns that there is a strong relevance between and, in that user attribute, the service providing apparatus 2 is associated with a high priority and stored. Thereby, for example, when the user attribute is a housewife and the user's voice is “I want to buy a battery”, a service that provides a service that sells daily commodities cheaper than others by learning of the learning unit 119 An example in which the priority order of the providing device 2 becomes higher can be considered. Further, when the user attribute is a single male and the user's voice is “I want to buy sneakers”, the learning of the learning unit 119 increases the priority of the service providing apparatus 2 with many types of shoe sales. Examples are possible.

Furthermore, the learning unit 119 may perform learning related to the above priority order according to the date and time or place where the user's voice is input in the AI speaker 1. That is, the above learning is performed for each certain time zone or certain area. Through such learning, for example, when a user utters “I want to buy a book” at home, a comprehensive book sales service in which books according to the user's hobbies are sold by learning of the learning unit 119. An example in which the priority order of the service providing apparatus 2 that provides the service is high can be considered. On the other hand, for example, when the user utters “I want to buy a book” at work or school, the learning of the learning unit 119 provides a book sales service in which a more specialized book is sold. An example where the priority is higher is conceivable. For example, when the user utters “I want to eat something” in the middle of the day, the learning unit 119 learns and the priority of the service providing apparatus 2 that provides a delivery service of, for example, pizza and noodles suitable for lunch is high. An example can be considered. On the other hand, for example, when the user speaks “I want to eat something” at night, the priority of the service providing apparatus 2 that provides a food delivery service with a food and drink menu including alcoholic beverages is high due to the learning of the learning unit 119. An example can be considered.

Further, the learning unit 119 may perform learning related to the priority order according to the emotion of the user estimated from the input user's voice. Specifically, the learning unit 119 stores an algorithm for estimating an emotion from the user's voice, and estimates the emotion from the user's voice using the algorithm. And the learning part 119 performs said learning for every user's emotion. With such learning, for example, if the emotion when the user utters “I want to go on a trip” is “pleasant”, the learning unit 119 learns mainly a trip with a resort or a downtown area. An example in which the priority order of the service providing apparatuses 2 sold in the market is higher is conceivable. On the other hand, for example, if the emotion when the user says “I want to go on a trip” is “tired”, the learning unit 119 learns mainly to travel to a region with a lot of nature or a small population. An example in which the priority order of the service providing apparatuses 2 sold in the market is higher is conceivable.

For example, a part of the functional configuration of the AI speaker 1 illustrated in FIG. 6 may be omitted, or another function may be added. The functions of the AI speaker 1 shown in FIG. 6 may be implemented by any device or terminal belonging to the AI speaker system 9. In addition, a computer device group consisting of a plurality of devices may cooperate to implement a function equivalent to the AI speaker 1 shown in FIG.

The processing executed by the control unit 11 of the AI speaker 1 can be considered as a method for controlling the AI speaker 1. That is, the present invention provides a specific step of identifying the speaker user account based on the input user's voice, a speaker user account for identifying a user who has spoken at an AI speaker, and a service providing apparatus that provides a service. The service user account that is associated with the identified speaker user account is identified by referring to a storage unit that stores the service user account to be identified and the settlement user account that identifies the user in the settlement system in association with each other. A service account notifying step for notifying the service providing device, and a payment account for referring to the storage unit and identifying the settlement user account associated with the identified speaker user account and notifying the settlement system It may be provided as a control method for AI speaker system comprising a cement notification step. Note that the steps of processing performed in the AI speaker system 9 are not limited to the example described in the above-described embodiment. The steps of this process may be interchanged as long as there is no contradiction.

The program executed by the control unit 11 of the AI speaker 1 is a recording medium readable by a computer device, such as a magnetic recording medium such as a magnetic tape and a magnetic disk, an optical recording medium such as an optical disk, a magneto-optical recording medium, or a semiconductor memory. Can be provided in a stored state. The program may be downloaded via a communication line such as the Internet. Note that various devices other than the CPU may be applied as the control means exemplified by the control unit 11 described above. For example, a dedicated processor or the like is used.

Claims (11)

Learning means for learning a correspondence relationship between a speaker user account for identifying a user who speaks in an AI speaker, a service user account for identifying a user in a service providing apparatus that provides a service, and a payment user account for identifying a user in a payment system When,
Identifying means for identifying the speaker user account based on a user's voice input at the AI speaker;
Service account notification means for notifying the service providing device of the service user account associated with the identified speaker user account;
An AI speaker system comprising: a payment account notification means for notifying the payment system of the payment user account associated with the specified speaker user account. Determining means for determining a priority order of a plurality of the service providing devices;
The AI speaker system according to claim 1, further comprising: a presentation unit that presents the service providing apparatus to the user who has spoken according to the determined priority order. The AI speaker system according to claim 2, wherein the determination unit determines the priority order based on an analysis result obtained by analyzing a meaning of an input user's voice. The determination means selects one or more service providing devices from the plurality of service providing devices based on the analysis result,
The AI speaker system according to claim 3, wherein the priority order of the selected one or more service providing devices is determined according to the user attribute corresponding to the specified speaker user account. The identifying means refers to a database in which each user's voiceprint and the user's speaker user account are associated with each other, and identifies the speaker user account corresponding to the input voiceprint. The AI speaker system according to any one of the above. The learning means performs learning related to the priority order,
The AI speaker system according to any one of claims 2 to 5, wherein the determination unit determines the priority using the learning result. The AI speaker system according to claim 6, wherein the learning unit performs the learning according to a date and time or a place where a user's voice is input to the AI speaker. The AI speaker system according to claim 6, wherein the learning unit performs the learning according to the user's emotion estimated from the input user's voice. The AI speaker system according to any one of claims 1 to 8, wherein the payment account notification means notifies the payment system of a value for service provision by the service providing device. A learning step of learning a correspondence relationship between a speaker user account for identifying a user who speaks in an AI speaker, a service user account for identifying a user in a service providing apparatus that provides a service, and a payment user account for identifying a user in a payment system When,
A specifying step of specifying the speaker user account based on a user voice input at the AI speaker;
A service account notification step of notifying the service providing apparatus of the service user account associated with the identified speaker user account;
A method for controlling an AI speaker system, comprising: a payment account notification step of notifying the payment system of the payment user account associated with the specified speaker user account. On the computer,
Learning and storing a correspondence relationship between a speaker user account for identifying a user who has spoken in an AI speaker, a service user account for identifying a user in a service providing apparatus that provides a service, and a payment user account for identifying a user in a payment system Learning steps to
A specifying step of specifying the speaker user account based on a user voice input at the AI speaker;
A service account notification step of notifying the service providing apparatus of the service user account associated with the identified speaker user account;
A payment account notification step of notifying the payment system of the payment user account associated with the specified speaker user account.

Images