WO2021080147A1

WO2021080147A1 - Device for providing personalized virtual assistant

Info

Publication number: WO2021080147A1
Application number: PCT/KR2020/011156
Authority: WO
Inventors: 방성철
Original assignee: 주식회사 유니온플레이스
Priority date: 2019-10-24
Filing date: 2020-08-21
Publication date: 2021-04-29
Also published as: KR102135859B1; JP7372693B2; JP2022534648A

Abstract

Provided is a device for providing a personalized virtual assistant comprising a communication unit, and a calculation processing unit for (a) managing one or more user profiles, (b) receiving, through the communication unit, a voice signal including an instruction command for a first service, (c) acquiring a personalized user profile, (d) selecting, on the basis of the voice signal and/or the personalized user profile, a first virtual assistant corresponding to the instruction command, (e) converting the voice signal into a first signal suitable for the first virtual assistant, (f) transmitting the first signal to a first virtual assistant platform through the communication unit and receiving a first response signal from the first virtual assistant platform through the communication unit, and (g) transmitting the first response signal to the electronic device.

Description

A device that provides a personalized virtual assistant

The present disclosure relates to an apparatus for providing a personalized virtual assistant.

In general, "virtual assistant" refers to a software agent that performs a task or service for a user based on a command or question. The virtual assistant is also referred to as an intelligent virtual assistant or intelligent personal assistant. The virtual assistant can operate based on text or voice. For example, the virtual assistant may perform a task or service for a user based on a user's voice input input through an electronic device such as a so-called "smart speaker" and a terminal. The smart speaker includes an audio input unit for inputting an audio signal such as a microphone, an audio output unit for outputting an audio signal such as a speaker, and a communication unit for wired or wireless communication connection with a virtual assistant. The smart speaker may further include a display that outputs a video signal. The terminal includes an audio input unit such as a microphone, an audio output unit such as a speaker, a communication unit providing a wired or wireless communication connection, and an operation processing unit such as a CPU for executing a virtual assistant. The terminal preferably further includes a display unit for outputting a video signal. The terminal may include, for example, a mobile communication terminal such as a smart phone, a portable terminal such as a notebook or tablet computer, and a computing device such as a personal computer.

For example, Korean Patent Registration No. 10-1683083 (Patent Document 1) entitled "Use of context information to facilitate command processing in a virtual assistant" filed by Apple and registered on November 30, 2016 is a virtual The secretary is opening.

In order to implement a virtual assistant, a virtual assistant platform using artificial intelligence such as "Siri" from Apple, "Assistant" from Google, "Alexa" from Amazon, and "Bixby" from Samsung Electronics is provided. In particular, with the development of artificial intelligence technology, the number of users using virtual assistants is increasing.

However, the conventional virtual assistant has the following disadvantages.

First of all, the virtual assistant of the existing smart speaker or terminal is subordinate to a specific virtual assistant platform.

For example, a smart speaker for a "Siri" virtual assistant or a "Siri" virtual assistant running on a terminal is dependent on the "Siri" virtual assistant platform provided by Apple. Therefore, it is very difficult to use a virtual assistant platform provided by another company, such as Samsung Electronics' "Bixby" virtual assistant platform, using a smart speaker for, for example, a "Siri" virtual assistant. In addition, in order for the terminal user to use the "Siri" virtual assistant, the "Siri" virtual assistant must be executed, and in order for the terminal user to use the "Bixby" virtual assistant, the "Bixby" virtual assistant must be executed. For example, even if a user wants to receive weather information using Google's "Assistant", music information using Apple's "Siri", and book information using Amazon's "Alexa", "Siri" virtual The "Siri" virtual assistant running on the smart speaker or terminal for the assistant provides information acquired using the "Siri" virtual assistant platform provided by Apple.

In addition, the existing smart speaker or the virtual assistant of the terminal is activated based on a voice command such as a predetermined wakeup command. The start command is also referred to as a trigger command. For example, in order to check the weather information using the "Siri" virtual assistant, the user must input a start command such as "Hey, Siri" and an instruction command "Check out the weather today."

In order to improve these shortcomings, for example, Korean Patent Laid-Open Patent No. 10-2019-0019401 (Patent Document 2) entitled "Electronic Device and Its Control Method" filed by LG Electronics Co., Ltd. and published on February 27, 2019. Discloses a configuration for transmitting voice to a plurality of artificial intelligence servers (ie, a plurality of virtual assistant platforms).

According to Korean Patent Publication No. 10-2019-0019401, when a voice including a start command and an instruction command is input, the electronic device generates a plurality of start commands corresponding to the plurality of virtual assistant platforms and instructs the plurality of start commands. A voice signal that combines commands is generated and then transmitted to each of the plurality of virtual assistant platforms. Thereafter, the electronic device receives and outputs result information corresponding to an instruction command from each of the plurality of virtual assistant platforms.

In addition, for example, Korean Patent Publication No. 10-2018-0115628 entitled "Management Layer for Multiple Intelligent Personal Assistant Services" filed by Harman International Industries, Inc. and published on October 23, 2018 (Patent Document 3) discloses a configuration for transmitting a startup command and an instruction command to a plurality of virtual assistant platforms.

However, according to Korean Laid-Open Patent Publication No. 10-2019-0019401, there is a disadvantage in that the instruction command is transmitted to the virtual assistant platform that the user does not want, and the result from the virtual assistant platform that the user does not want is also received and output. In addition, according to Korean Patent Application Publication No. 10-2018-0115628, there is a disadvantage in that the user must explicitly specify the virtual assistant platform he/she wants. In other words, "Google. What is my schedule for today? And Siri. Play music from my playlist."

In addition, the existing smart speaker or virtual assistant of the terminal only assumes a specific user and does not consider a number of users. Therefore, it is very difficult to provide a virtual assistant suitable for each of a plurality of users by using the existing smart speaker or the virtual assistant of the terminal.

For example, in the case of a smart speaker placed in a home, even though all members of the family can use it, it is difficult to provide a virtual assistant suitable for each user. For a user whose voice is not clear, such as an elderly person, it is very difficult to provide a virtual assistant by recognizing the user's voice. In addition, when a virtual assistant is provided to a temporary visitor, not a family member, through a smart speaker, functions that cannot be provided to the visitor may be provided through the smart speaker. For example, when the virtual assistant controls home appliances in the home, a function of controlling the home appliances in the home may be provided to the visitor without limitation.

[Prior technical literature]

[Patent Literature]

Patent Document 1. Korean Patent Registration No. 10-1683083.

Patent Document 2. Korean Laid-Open Patent No. 10-2019-0019401.

Patent Document 3. Korean Laid-Open Patent No. 10-2018-0115628.

An object of the technology described herein is to provide a personalized virtual assistant using a user profile set for each user.

In order to achieve the above technical problem, according to an aspect of the technology described herein, a communication unit for providing a wired or wireless communication interface; And (a) managing one or more user profiles each including voice characteristic information, service identification information, and virtual assistant platform identification information corresponding to the service identification information. (b) a process of receiving a voice signal including an instruction command for a first service from an electronic device through the communication unit; (c) processing of obtaining a personalized user profile from among the one or more user profiles based on the voice signal; (d) processing for selecting a first virtual assistant corresponding to the instruction command for the first service from among one or more virtual assistants based on at least one of the voice signal and the personalized user profile; (e) processing for converting the voice signal into a first signal suitable for the first virtual assistant; (f) transmitting the first signal to a first virtual assistant platform providing the first virtual assistant through the communication unit, and transmitting a first response signal corresponding to the first signal from the first virtual assistant platform to the communication unit Processing received through; And (g) an operation processor configured to transmit the first response signal to the electronic device.

According to the technology described herein, a personalized virtual assistant may be provided based on a user profile set for each user. For example, a personalized user profile may be obtained based on a voice signal transmitted from a smart speaker or a terminal, and a personalized virtual assistant may be provided for each user based on at least one of the voice signal or the personalized user profile.

1 is a diagram showing an exemplary configuration of an apparatus for providing a personalized virtual assistant according to a first embodiment of the technology described herein.

Fig. 2 is a diagram showing an exemplary configuration of a system environment including an apparatus for providing a personalized virtual assistant according to the first embodiment.

3 is a diagram illustrating a process performed by an operation processing unit of the apparatus for providing personalized advertisements according to the first embodiment.

4 is a diagram illustrating a process performed by an operation processing unit of the apparatus for providing a personalized advertisement according to the first embodiment.

5 is a diagram illustrating a user profile of a device providing personalized advertisements according to the first embodiment.

6 is a diagram illustrating a process performed by an operation processing unit of the apparatus for providing personalized advertisements according to the second embodiment.

Hereinafter, an embodiment of an apparatus for providing a personalized virtual assistant according to the technology described herein will be described in more detail with reference to the accompanying drawings. Meanwhile, in the drawings for describing embodiments of the technology described herein, for convenience of description, only a part of the actual configuration may be shown, some of the actual components may be omitted, or modified, or a different scale may be shown.

Referring to FIG. 1, an apparatus 100 for providing a personalized virtual assistant includes a communication unit 110 and an operation processing unit 130. Also, referring to FIG. 1, the apparatus 100 for providing a personalized virtual assistant may further include a user profile storage unit 150. Hereinafter, the device 100 providing a personalized virtual assistant is also simply referred to as the device 100.

The device 100 according to the techniques described herein may be implemented using computing devices such as, for example, a personal computer and a dedicated computer.

The communication unit 110 provides a wired or wireless communication interface. For example, the communication unit 110 provides a wired or wireless communication interface with the electronic device 200 in FIG. 2. In addition, the communication unit 110 provides a wired or wireless communication interface with a virtual assistant platform (300 in FIG. 2). The communication unit 110 may be implemented using a semiconductor chip that performs wired or wireless communication.

The operation processing unit 130 may be implemented by, for example, a semiconductor device such as a central processing unit (CPU). Some functions of the operation processing unit 130 may be implemented by a first CPU, and some other functions of the operation processing unit 130 may be implemented by a second CPU. The detailed configuration of the operation processing unit 130 will be described later.

The user profile storage unit 150 stores one or more user profiles. The user profile storage unit 150 may be implemented by, for example, a semiconductor device such as a semiconductor memory.

Referring to FIG. 2, the system environment includes a device 100 for providing a personalized virtual assistant, an electronic device 200, and one or more virtual assistant platforms 300. The electronic device 200 includes a smart speaker 210 or a terminal 230.

As described above, the smart speaker 210 includes an audio input unit for inputting an audio signal such as a microphone, an audio output unit for outputting an audio signal such as a speaker, and a communication unit for wired or wireless communication connection with a virtual assistant. do. As described above, the terminal 230 includes an audio input unit such as a microphone, an audio output unit such as a speaker, a communication unit providing a wired or wireless communication connection, and an operation processing unit such as a CPU for executing a virtual assistant.

The one or more virtual assistant platforms 300 may include, for example, a virtual assistant platform 300-1, a virtual assistant platform 300-2, and a virtual assistant platform 300-n. n is a natural number of 3 or more, for example. For example, the virtual assistant platform 300-1 is a platform for the "Siri" virtual assistant of Apple, and the virtual assistant platform 300-2 is a platform for the "Assistant" virtual assistant of Google, and the virtual assistant platform 300- n) May be a platform for Samsung's "Bixby" virtual assistant

Hereinafter, the operation processing unit 130 according to the first embodiment will be described in more detail.

3 and 4 are views exemplarily showing processing performed by an operation processing unit of the apparatus for providing personalized advertisements according to the first embodiment.

Referring to FIG. 3, the operation processing unit 130 performs a process S110 of managing one or more user profiles each including voice characteristic information, service identification information, and virtual assistant platform identification information corresponding to the service identification information.

Process S110 may include, for example, a process of generating a regular user profile based on an interface for user registration. Regular user profiles are managed as one of one or more user profiles.

For example, a first user using the electronic device 200 inputs his/her voice using an interface for user registration of the electronic device 200, and the user profile of the first user, that is, the first user Profiles can be created. Since the first user profile is generated based on the interface for user registration, it is a regular user profile.

5 is a diagram illustrating a user profile of a device that provides a personalized advertisement according to the first embodiment.

Referring to FIG. 5, a first user profile to a third user profile are illustrated by way of example.

The first user profile of FIG. 5 is, that is, a regular user profile generated based on an interface for user registration.

Initially, the first user profile may include only an item corresponding to voice characteristic information. Thereafter, the first user may input service identification information and virtual assistant platform identification information corresponding to the service identification information through an interface for user registration. For example, the first user may input service identification information and virtual assistant platform identification information corresponding to the service identification information through a voice input such as "tell me about weather information using an assistant."

Meanwhile, the first user profile may include service identification information automatically generated based on service preference information at the initial stage of creation and virtual assistant platform identification information corresponding to the service identification information. For example, when a general user prefers "Assistant" for a "weather information" service and a general user prefers "Siri" for a "movie reservation" service, as illustrated in FIG. 5, the service of the first user profile As the identification information and the virtual assistant platform identification information, a correspondence relationship between the service identification information and the virtual assistant platform identification information may be automatically generated, such as "weather information, assistant", "movie reservation, Siri".

Meanwhile, after the first user profile is created, as the first user uses the electronic device 200, service identification information of the first user profile and virtual assistant platform identification information corresponding to the service identification information may be generated or updated. . For example, when the first user transmits a voice input such as "Siri. How is the weather today?" to the device 100 through the electronic device 200, the service identification information of the first user profile and the virtual assistant platform identification information are The correspondence relationship may be created or updated, such as "weather information, Siri". That is, based on the first virtual assistant selected through processing S140 to be described later, service identification information and virtual assistant platform identification information corresponding to the service identification information may be generated or updated.

Process S110 may include, for example, a process of generating a temporary user profile based on a voice signal. Temporary user profiles are managed as one of one or more user profiles.

For example, when the electronic device 200 is installed in a home and used by a family member, a regular user profile is created for a family member, and a temporary user profile is created for a temporary user such as a visitor, for example.

5, the third user profile is a temporary user profile. Initially, the third user profile may include only an item corresponding to voice characteristic information. Thereafter, as the temporary user uses the electronic device 200, service identification information of the third user profile and virtual assistant platform identification information corresponding to the service identification information may be generated or updated. For example, when a third user transmits a voice input such as "Siri. How is the weather today?" to the device 100 through the electronic device 200, the service identification information of the third user profile and the virtual assistant platform identification information are The correspondence relationship may be created or updated, such as "weather information, Siri".

Meanwhile, the user profile may further include information on whether a service can be provided corresponding to the service identification information. For example, referring to FIG. 5, the third user profile is defined as "not available" for "refrigerator temperature control". In other words, temporary users cannot use services such as "refrigerator temperature control". Service availability information may be set so that not only temporary users but also family members, that is, regular users, cannot use services such as "refrigerator temperature control". For example, even if a child is a family member, a service such as "refrigerator temperature control" may be set to be disabled. The service availability information may be set as unavailable for, for example, a service providing sensitive information and a service controlling a dangerous home appliance.

Process S110 may include a process of converting a temporary user profile to a regular user profile or deleting a temporary user profile based on a predetermined condition.

That is, even a temporary user can be converted to a regular user through an interface for user registration. Alternatively, if a temporary user's voice input corresponding to the temporary user profile is periodically received through the electronic device 200 after being registered as a temporary user profile, the temporary user profile may be converted to a regular user profile. In addition, if there is no voice input of the temporary user corresponding to the temporary user profile for more than a certain period after being registered as the temporary user profile, the temporary user profile may be deleted.

Referring back to FIG. 3, the operation processing unit 130 performs a process S120 of receiving a voice signal including an instruction command for a first service from the electronic device 200 through the communication unit 110. The processing of receiving the audio signal is similar to the conventional configuration, so a detailed description thereof is omitted.

Thereafter, the operation unit 130 performs a process S130 of obtaining a personalized user profile from among one or more user profiles based on the voice signal received through the process S120.

The operation unit 130 may select a personalized user profile from among one or more user profiles by comparing the voice signal received through the processing S120 with voice characteristic information of one or more user profiles, for example.

Referring back to FIG. 3, the operation processing unit 130 responds to an instruction command for a first service among one or more virtual assistants based on at least one of a voice signal received through processing S120 or a personalized user profile obtained through processing S130. A process S140 of selecting a corresponding first virtual assistant is performed.

When the voice signal received through the process S120 includes a start command, the process S140 may include a process of selecting a first virtual assistant corresponding to the start command.

For example, if the voice signal includes a start command specifying "Siri" as a first service, that is, a virtual assistant for weather information, such as "Siri. How is the weather today?", the first service corresponding to the first service through processing S140 "Siri" is selected as the virtual assistant.

If the voice signal received through the process S120 does not include a specific start command, the process S140 is a first virtual assistant based on the service identification information and the virtual assistant platform identification information included in the personalized user profile obtained through the process S130. It may include a process of selecting.

That is, when receiving a voice signal such as "how is the weather today?", the operation processing unit 130 selects the first virtual assistant based on the service identification information and the virtual assistant platform identification information included in the personalized user profile. For example, when the personalized user profile is the first user profile shown in FIG. 5, the operation processing unit 130 selects “Assistant” as a first service, that is, a first virtual assistant corresponding to “weather information” through processing S140. .

That is, in the process S140, if the voice signal received through the process S120 includes a start command corresponding to the first virtual assistant, the first virtual assistant corresponding to the start command is selected, and the voice signal received through the process S120 is removed. 1 If the startup command corresponding to the virtual assistant is not included, the first virtual assistant may be selected based on the service identification information and the virtual assistant platform identification information included in the personalized user profile.

Meanwhile, the voice signal may include an electronic device start command. That is, in order to drive the electronic device 200, not to select a specific virtual assistant, the voice signal may include an electronic device start command. For example, if the start command of the smart speaker 210 is defined as "speaker", the voice signal may include an electronic device start command such as "speaker. How is the weather today?" When the voice signal includes an electronic device startup command, the operation processing unit 130 may select the first virtual assistant based on the service identification information and the virtual assistant platform identification information included in the personalized user profile obtained through processing S130. have.

Referring to FIG. 4, the operation processing unit 130 performs a process S150 of converting the voice signal received through the process S120 into a first signal suitable for the first virtual assistant.

The process S150 may include, for example, a process of generating a first signal including a start command corresponding to the first virtual assistant and an instruction command for the first service.

For example, even if the voice signal is in a form in which a specific virtual assistant is not designated, such as "How is the weather today?", the operation processing unit 130 is a first signal including a start command corresponding to the first virtual assistant selected through processing S140. Can be created. For example, when "Assistant" is selected as the first virtual assistant, the operation processing unit 130 includes a startup command corresponding to the first virtual assistant and an instruction command for the first service, such as "Assistant. How's the weather today?" 1 can generate signals.

The process S150 includes, for example, a process of converting an instruction command into text information using a speech to text (STT) function; Processing of generating corrected text information in which the text information is clearly corrected; And processing of generating a first signal including a correction instruction command obtained by converting the corrected text information using a text to speech (TTS) function.

For example, even when the instruction instruction is not grammatically clear such as "weather today" or includes dialect, the operation processing unit 130 converts the instruction instruction into text information such as "weather today". Thereafter, the operation processing unit 130 analyzes the text information and generates corrected text information more naturally and clearly, such as "how is the weather today?" Thereafter, the operation processing unit 130 may generate a first signal including a correction instruction command that converts the corrected text information using the TTS function.

The correction instruction command may be generated, for example, using the TTS function based on standard voice characteristics. That is, the correction instruction command may be generated similarly to the original voice of the user by using voice characteristic information included in the personalized user profile, but may be generated using standard voice characteristics. When a correction instruction command is generated using standard voice characteristics, the first virtual assistant platform may more accurately recognize the first signal. The first signal including the modification instruction command may further include a start command corresponding to the first virtual assistant.

In addition, when a correction instruction command is generated using the TTS function based on the standard voice characteristic, the user's voice characteristic information is not included in the correction instruction command. Therefore, since the user's voice characteristic information is not transmitted to the first virtual assistant platform, the first virtual assistant may be provided through the first virtual assistant platform without exposing the user's voice characteristic information.

The operation processing unit 130 may perform processing S150 using an artificial intelligence platform. That is, the operation processing unit 130 may perform processing such as speech recognition processing, natural language understanding processing, natural language generation processing and STT processing, and TTS processing using an artificial intelligence platform. By using the artificial intelligence platform to perform processing by the arithmetic processing unit 130, the arithmetic processing unit 130 may more efficiently generate the first signal.

Referring back to FIG. 4, the operation processing unit 130 transmits the first signal generated through processing S150 to the first virtual assistant platform through the communication unit 110, and transmits the first signal corresponding to the first signal from the first virtual assistant platform. 1 A process S160 of receiving a response signal through the communication unit 110 is performed.

For example, when "Assistant" is selected as the first virtual assistant, when the first signal is "Assistant. How is the weather today?", the operation processing unit 130 uses the virtual assistant platform 300-2 as the first virtual assistant platform. And transmits a first signal to the virtual assistant platform 300-2. After that, the operation processing unit 130 receives a first response signal from the virtual assistant platform 300-2.

Preferably the user profile includes user identification information. The user identification information may be information that absolutely identifies a user, but may also be information used to relatively identify a user of the electronic device 200.

The process S160 may include a process of transmitting the user identification information of the personalized user profile together with the first signal to the first virtual assistant platform through the communication unit 110. When the first virtual assistant platform, for example, the virtual assistant platform 300-2 receives the user identification information of the personalized user profile, it can confirm that a user corresponding to the user identification information of the electronic device 200 has requested a service.

Thereafter, the operation processing unit 130 performs a process S170 of transmitting the first response signal received through the process S160 to the electronic device 200.

As described above, according to the first embodiment, a personalized virtual assistant may be provided based on a user profile set for each user even if the user does not input a startup command for the virtual assistant. In addition, a startup command for the virtual assistant transmitted to the virtual assistant platform may be automatically generated by the operation processing unit 130.

According to the second embodiment, the voice signal further includes an instruction command for the second service. For example, "How's the weather today? And play music.", the voice signal is a command command for the first service, ie, "How is the weather today?", and a command command for the second service, ie, "Play music." It may include.

The instruction command for the first service is processed similarly to the above-described first embodiment. That is, the operation processing unit 130 receives the first response signal from the first virtual assistant platform and transmits it to the electronic device 200.

The instruction command for the second service is also processed in the same manner as in the second embodiment described above.

More specifically, the operation processing unit 130 includes processing S180 of selecting a second virtual assistant corresponding to an instruction command for a second service from among one or more virtual assistants based on at least one of a voice signal or a personalized user profile, and a voice signal. A process S190 of converting the second signal into a second signal suitable for the second virtual assistant, and the second signal is transmitted through the communication unit 110 to a second virtual assistant platform providing a second virtual assistant, and the second virtual assistant platform transmits the second signal. Processing S200 of receiving a second response signal corresponding to the signal through the communication unit 110 and processing S210 of transmitting the second response signal to the electronic device 200 are further performed. The second virtual assistant may be the same as or different from the first virtual assistant. Since the processes S180 to S210 are substantially the same as the processes S140 to S170 described with reference to FIGS. 3 to 4, detailed descriptions are omitted.

As described above, according to the second embodiment, in addition to the effects according to the above-described first embodiment, even if the user does not input a startup command for the virtual assistant for each of the plurality of services, the user profile set for each user is As a basis, a personalized virtual assistant may be provided for each of a plurality of services. In addition, a startup command for a plurality of virtual assistants transmitted to the plurality of virtual assistant platforms may be automatically generated for each of the plurality of virtual assistants by the operation processing unit 130.

Although embodiments of the technology described herein have been specifically described, this is merely illustrative of the technology described herein, and those of ordinary skill in the art to which the technology described herein belongs Various modifications will be possible without departing from the essential characteristics of the technology.

For example, the service identification information illustrated with reference to FIG. 5 and the virtual assistant platform identification information corresponding to the service identification information are exemplary only, and the service identification information may be variously changed.

Therefore, the embodiments described in the present specification are not intended to limit the technology described herein, but to describe it, and the spirit and scope of the technology described herein are not limited by these embodiments. The scope of the rights of the technology described herein should be construed by the claims below, and all technologies within the scope equivalent thereto are to be construed as being included in the scope of the technology described herein.

[Explanation of code]

100: a device that provides a personalized virtual assistant

110: communication department

130: operation processing unit

150: user profile storage unit

200: electronic device

210: smart speaker

230: terminal

300: virtual service platform

Claims

A communication unit that provides a wired or wireless communication interface; And

(a) processing of managing one or more user profiles each including voice characteristic information, service identification information, and virtual assistant platform identification information corresponding to the service identification information; (b) a process of receiving a voice signal including an instruction command for a first service from an electronic device through the communication unit; (c) processing of obtaining a personalized user profile from among the one or more user profiles based on the voice signal; (d) processing for selecting a first virtual assistant corresponding to the instruction command for the first service from among one or more virtual assistants based on at least one of the voice signal and the personalized user profile; (e) processing for converting the voice signal into a first signal suitable for the first virtual assistant; (f) transmitting the first signal to a first virtual assistant platform providing the first virtual assistant through the communication unit, and transmitting a first response signal corresponding to the first signal from the first virtual assistant platform to the communication unit Processing received through; And (g) an operation processing unit that transmits the first response signal to the electronic device.

Device for providing a personalized virtual assistant comprising a.
The method of claim 1,

The apparatus for providing a personalized virtual assistant wherein the at least one user profile further includes information on whether or not a service is available corresponding to the service identification information.
The method of claim 1,

The processing (a) includes (a-1) generating a regular user profile based on an interface for user registration and adding it to the one or more user profiles.
The method of claim 1,

The processing (a) includes (a-2) managing the one or more user profiles based on service preference information.
The method of claim 1,

The processing (a) includes (a-3) generating a temporary user profile based on the voice signal and adding it to the one or more user profiles.
The method of claim 5,

The processing (a) includes (a-4) converting the temporary user profile to a regular user profile or deleting the temporary user profile based on a predetermined condition. .
The method of claim 1,

The processing (a) includes (a-5) updating the personalized user profile based on the first virtual assistant.
The method of claim 1,

The processing (c) includes (c-1) comparing the voice signal and the voice characteristic information of the one or more user profiles to select the personalized user profile from among the one or more user profiles. A device that provides a virtual assistant.
The method of claim 1,

The voice signal further includes a start command corresponding to the first virtual assistant,

The process (d) includes (d-1) a process of selecting the first virtual assistant corresponding to the start-up command.
The method of claim 1,

The processing (d) includes processing (d-2) selecting the first virtual assistant based on the service identification information and the virtual assistant platform identification information included in the personalized user profile. A device that provides a secretary.
The method of claim 10,

Wherein the voice signal further comprises an electronic device startup command.
The method of claim 1,

The process (e) includes (e-1) a process of generating the first signal including a start command corresponding to the first virtual assistant and the instruction command. .
The method of claim 12,

The processing (e) includes: (e-2) a processing of converting the instruction command into text information using a speech to text (STT) function; (e-3) processing of generating corrected text information based on the text information; And (e-4) processing for generating the first signal including a correction instruction command obtained by converting the corrected text information using a text to speech (TTS) function. .
The method of claim 13,

Wherein the first signal further comprises a start command corresponding to the first virtual assistant.
The method of claim 13,

The apparatus for providing a personalized virtual assistant wherein the modification instruction command is generated using the TTS based on standard voice characteristics.
The method of claim 1,

The one or more user profiles each further include user identification information,

The processing (f) includes (f-1) transmitting the user identification information of the personalized user profile together with the first signal to the first virtual assistant platform through the communication unit. A device that provides.
The method of claim 1,

The voice signal further includes an instruction command for a second service,

The calculation processing unit,

(h) processing for selecting a second virtual assistant corresponding to the instruction command for the second service from among one or more virtual assistants based on at least one of the voice signal and the personalized user profile; (i) processing of converting the voice signal into a second signal suitable for the second virtual assistant; (j) transmitting the second signal to a second virtual assistant platform providing the second virtual assistant through the communication unit, and transmitting a second response signal corresponding to the second signal from the second virtual assistant platform to the communication unit Processing received through; And (k) transmitting the second response signal to the electronic device.
The method of claim 1,

User profile storage unit for storing the one or more user profiles

An apparatus for providing a personalized virtual assistant further comprising.