KR102380557B1 - Apparatus and method for interfacing artificial intelligence service - Google Patents

Abstract

In accordance with one embodiment, an artificial intelligence service interfacing apparatus includes: a telephone network interfacing part connected with a telephone network; a data network interfacing part connected with a data network; an artificial intelligence network interfacing part connected with an artificial intelligence service providing server; and a processor controlling a first voice received from a first terminal connected with the telephone network such that the first voice is delivered to the artificial intelligence service providing server, controlling a first text text-converted from the first voice when receiving the first text from the artificial intelligence service providing server such that the first text is delivered to a second terminal connected with the data network, controlling a second text received from the second terminal such that the second text is delivered to the artificial intelligence service providing server, and controlling a second voice voice-converted from the second text when receiving the second voice from the artificial intelligence service providing server such that the second voice is delivered to the first terminal. Therefore, the present invention is capable of enabling two terminal devices to communicate with each other through different modes in which one of the terminal devices maintains voice or video communication and the other one uses chatting messages.

Description

APPARATUS AND METHOD FOR INTERFACING ARTIFICIAL INTELLIGENCE SERVICE

The present invention relates to an apparatus and method for interfacing artificial intelligence services.

A smart device such as a smartphone or a smart pad provides a voice/video call function and a data communication function. A user may chat with a counterpart using a message through a data communication function provided by a smart device, and may also share content such as maps, music, news, or photos in real time.

Meanwhile, servers that provide artificial intelligence services have recently appeared. Since this artificial intelligence service providing server has a learning function, it can provide a higher level of service the more it is used.

The smart device can receive various services by accessing the artificial intelligence service providing server using a data communication function. For example, the user can ask not only simple questions about weather or personal schedule but also more complex and sophisticated information to the AI service providing server through a smart device, and receive a response.

Korean Patent Laid-Open Publication No. 2011-0041322 (published on April 21, 2011)

When a user attempts or proceeds with an audio or video call with a counterpart through a smart device, a situation in which it is difficult to make a call may occur depending on circumstances. For example, in a theater where a movie is playing or in a room where a baby is sleeping, it may be difficult for a user to make an audio or video call with the other party.

At this time, the user and the other party can end the call and chat through a message. However, it is the user who is difficult to make a call with, not the other party. Although a call is more convenient than chatting through a message as a counterpart, there may be a situation in which chatting through a message is required due to difficulties experienced by the user.

Accordingly, the problem to be solved by the present invention is that when a user tries to make a call with the other party through a smart device or when a call is difficult depending on the situation, the user can send a message without terminating the call. It is to provide a technology that allows not only to perform a conversation with the other party through chatting, but also to allow the other party not to end the call as well as to perform a call rather than chatting through a message.

In addition, in providing such technology, it may be included in the problem to be solved by the present invention to provide such technology through a system implemented in a public network, not a system implemented only in smart devices.

However, the problems to be solved of the present invention are not limited to those mentioned above, and other problems to be solved that are not mentioned can be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be.

An artificial intelligence service interfacing device according to an example includes a call network interfacing unit connected to a call network, a data network interfacing unit connected to a data network, and an artificial intelligence network interfacing unit connected to an artificial intelligence service providing server, the call network Controls so that the first voice received from the first terminal connected to is delivered to the artificial intelligence service providing server, and when the first text converted from the first voice is received from the artificial intelligence service providing server, the first text is control to be transmitted to a second terminal connected to the data network, control such that a second text received from the second terminal is transmitted to the artificial intelligence service providing server, and a second voice converted from the second text into the voice and a processor controlling the second voice to be transmitted to the first terminal when received from the artificial intelligence service providing server.

An artificial intelligence service interfacing method according to an embodiment is performed by an artificial intelligence service interfacing device, and when a first voice is received from a first terminal connected to a communication network, transmitting the first voice to an artificial intelligence service providing server and, when the first text obtained by converting the first voice into text is received from the artificial intelligence service providing server, delivering the first text to a second terminal connected to a data network; transmitting the second text to the artificial intelligence service providing server when receiving 1 is performed including the step of transmitting to the terminal.

According to an embodiment, while the terminal device and the terminal device are attempting or conducting an audio or video call with each other, one terminal device maintains an audio or video call, but the other terminal device uses a chatting method through a message. It can communicate with the device.

1 is a diagram conceptually illustrating a mobile communication network to which an artificial intelligence interfacing device according to an embodiment is applied.
FIG. 2 is a diagram illustrating the configuration of the artificial intelligence service providing server shown in FIG. 1 .
FIG. 3 is a diagram illustrating the configuration of the call processing network shown in FIG. 1 .
FIG. 4 is a diagram illustrating the configuration of the artificial intelligence interfacing device shown in FIG. 1 .
5 is a diagram illustrating a procedure of an artificial intelligence interfacing method according to an embodiment.
6 is a diagram illustrating a flow of an artificial intelligence interfacing method according to an embodiment.
7 and 8 are each an example of a screen of a terminal device according to an embodiment.
9 is a diagram for explaining an algorithm employed to compensate for delay in text input in a text-type call service.
10 is a diagram illustrating a flow of an artificial intelligence interfacing method according to an embodiment.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the art to which the present invention pertains It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims.

In describing the embodiments of the present invention, if it is determined that a detailed description of a well-known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, the terms to be described later are terms defined in consideration of functions in an embodiment of the present invention, which may vary according to intentions or customs of users and operators. Therefore, the definition should be made based on the content throughout this specification.

1 is a diagram conceptually illustrating a mobile communication network 10 to which an artificial intelligence interfacing apparatus 100 according to an embodiment is applied. However, since FIG. 1 is merely exemplary, the artificial intelligence interfacing device 100 (hereinafter referred to as an interfacing device) is not construed as being limited to being applied only to the mobile communication network 10 shown in FIG. 1 .

Referring to FIG. 1 , the mobile communication network 10 may include an interfacing device 100 , an artificial intelligence service providing server 200 , a communication network 410 , and a data network 420 . The first terminal device 300/1 and the second terminal device 300/2 each exemplarily show a terminal accessing the mobile communication network 10 (the call network 410 or the data network 420 of the mobile communication network 10). In this case, each terminal device 300/1,300/2 may be provided with a data service such as a voice/video call or an artificial intelligence service by accessing the mobile communication network 10.

First, the artificial intelligence service providing server 200 refers to a server providing an artificial intelligence service. The artificial intelligence service providing server 200 may be connected to various terminals or servers in addition to the terminal devices 300/1 and 300/2, and may provide various artificial intelligence services to each of them.

The configuration of the artificial intelligence service providing server 200 is shown in FIG. 2 . 2, the artificial intelligence service providing server 200 is an artificial intelligence processing unit (intelligence workflow, IWF) 210 for performing artificial intelligence processing, a natural language processing unit for processing natural language (natural language understand) 220, voice It may include a text to speech (TTS) module 230 for performing synthesis and a speech to text (STT) 240 for performing text synthesis. In addition, the artificial intelligence service providing server 200 may be formed as a server group consisting of a plurality of servers.

Among them, the artificial intelligence processing unit 210 is configured to provide artificial intelligence services by imitation of a human thinking processing method. For example, the artificial intelligence processing unit 210 may perform functions such as image recognition, image caption, language recognition or conversation, and for this purpose, it may be learned in advance by a method such as machine learning or deep learning.

Next, the natural language processing unit may employ an algorithm for recognizing and analyzing a question when it receives an input, and then providing a recommended response thereto. Here, the recommended response may be, for example, a response recommended based on the frequency of cases in which the same user answered the same or similar question in the past, or based on various cases in which other users answered the same or similar question.

Next, the TTS module 230 is a module that synthesizes or converts the text input by the user into voice and provides the same. In this case, various well-known algorithms for converting text into speech may be employed.

Meanwhile, the TTS module 230 may reflect an emotion or mood acquired for the user when synthesizing the voice. For example, even if the same 'hello' voice is a happy or welcome emotion, the volume or tone of the sound is relatively increased, or if it is sad or depressed, the volume or tone of the sound is relatively lowered. can synthesize speech.

To this end, the TTS module 230 must be able to grasp the user's emotion or mood, which can be implemented by employing any one of the following embodiments.

First, the TTS module 230 may additionally include a module for recognizing or interpreting emoticons that reflect a person's emotion or mood. For example, the TTS module 230 may include a table in which emotions are mapped for each emoticon. When a smiley emoticon, a crying emoticon, an angry emoticon, or a bewildered emoticon is received along with the text, the TTS module 230 may obtain an emotion corresponding to the emoticon from the above-described table.

Second, the TTS module 230 may obtain separate information representing a person's emotion or mood, etc. from the terminal device. To this end, for example, the terminal device 300/2 may be provided with a separate button for inputting the user's own emotion or mood. If the user presses a button indicating his or her emotion or mood while inputting a predetermined text, information on the user's emotion or mood corresponding to the pressed button may be transmitted to the TTS module 230 .

Third, the TTS module 230 may include a separate module for recognizing the context or atmosphere shown in the text. To this end, an algorithm for understanding and recognizing natural language may be employed in the TTS module 230 .

On the other hand, if the TTS module 230 starts inputting text from the chatting party, but the text input is not completed for a predetermined time, it emits a sound indicating that the text input completion is delayed between voices. It can be inserted into or in the middle of a voice. These types of sounds may include words that typically indicate a delay in a conversation, such as 'umm...', 'so...', 'wait a minute...'.

Next, the STT module 240 is a module that synthesizes or converts the user's voice into text and provides it. In this case, various well-known algorithms for converting speech into text may be employed.

On the other hand, the STT module 240 according to an embodiment may include a module learned to determine the user's emotion or mood from the voice input by the user. For example, the STT module 240 determines the state of the user who is emitting a voice, such as joy, sadness, depression, or anger, based on the frequency or pitch of the user's voice, the spacing between words, or the context recognized from the voice. , pleasure, embarrassment, or excitement.

In addition, when synthesizing text, the STT module 240 may reflect the above-described emotion or mood determined for the user. For example, with respect to the synthesized text 'Hello', if the user has a happy or welcome emotion, a corresponding emoticon may be added to the front, middle, or end of the text, and an emoticon corresponding to the sad or depressed emotion may be added to the text in the same/similar manner.

Referring back to FIG. 1 , the communication network 410 and the data network 420 are communication networks through which the terminal devices 300/1 and 300/2 connect. 3 is a diagram conceptually illustrating the configuration of the communication network 410 and the data network 420 .

Referring to FIG. 3 , the communication network 410 may be referred to as an intelligent network, for example, an IP multimedia subsystem (IMS).

The call network 410 includes a call session control function (CSCF) 411 , an application service node (telephony application server, TAS) 412 or a media resource function (MRF) 413 . In addition, HLR (Home Location Register) (414), MGCF (Media Gateway Control Function) (415), MGW (Media Gateway) (416), SCC AS (Service Centralization and Continuity Application Server) (417), etc. can do. In addition, the functions performed by each configuration to be described below are merely exemplary. Accordingly, each of the components may additionally perform functions other than the functions described below.

Among them, the switching network 411 obtains information on the location of each terminal device 300/1 and 300/2, that is, to which base station the corresponding terminal device is connected.

The application service node 412 processes basic functions related to a phone call and additional services (call hold, swap, forward) related to a phone call. For example, the application service node 412 may process a 'text-type call service request' received from the terminal device 300/2. The text-type call service means that while the terminal device 300/1 and the terminal device 300/2 are attempting or conducting an audio or video call with each other, any one of the terminal devices 300/1 is While maintaining a call, the other terminal device 300/2 refers to a service that communicates with the terminal device 300/1 in a chatting method through a message. We will look at this in more detail later.

Also, when a call is connected or disconnected between the terminal devices, the application service node 412 may notify the interfacing device 100 to be described below.

The media resource server 413 performs codec conversion. Through codec conversion, packet exchange between terminal devices of different specifications is possible. To this end, the media resource server 413 may include a codec conversion module.

The media resource server 413 forking media data. For example, the media resource server 413 may transmit audio/video between the first terminal device 300/1 and the second terminal device 300/2. Also, the media resource server 413 may transmit the media data received from each terminal device 300/1 and 300/2 to the interfacing device 100, and may also transmit it in the opposite direction. To this end, the media resource server 413 may include a communication module for communication with each of the terminal devices 300/1 and 300/2 or the interfacing device 100 .

The media resource server 413 mixes (mixing) (or muxing) media data. The media data may include, but is not limited to, voice/video calls, data packets, or dual tone multiple frequency (DTMF) signals provided to each terminal device 300/1,3002.

In mixing, the media resource server 413 may mix sound sources received from various objects with each other. For example, the media resource server 413 may mix the audio signal received from the artificial intelligence service providing server 200 with the audio/video to be delivered to each terminal device 300/1,300/2, and also each terminal device ( 300/1,300/2), a predefined sound source or image (hereinafter referred to as a standby sound source or a standby image) may be mixed with the audio/video to be delivered. In addition, the media resource server 413 may adjust the size of the sound sources to be mixed, and such adjustment may be performed by a command received from the application service node 412 . To this end, the media resource server 413 may include a mixing module.

On the other hand, in the case of the shown Home Location Register (HLR) 414, MGCF (Media Gateway Control Function) 415, MGW (Media Gateway) 416, SCC AS (Service Centralization and Continuity Application Server) (417) already Since it is the same as the known configuration, a description thereof will be omitted.

Next, the data network 430 is also referred to as a legacy call processing network, and may mean, for example, a mobile communication network such as WCDMA. The data network 430 may provide an audio/video service to each terminal device 300/1 and 300/2. Also, the data network 430 may provide an app-based predetermined service to each terminal device 300/1 and 300/2.

The data network 430 includes a mobile switching center (MSC) 421 or a home location register (HLR) 422, in addition to a Gateway General packet radio service Support Node (GGSN) 423, NodeB (424), RNC (Radio Network Controller) (425), SGSN (Serving General packet radio service Support Node) (426), CGS (Cellular Gateway Switch) (427), MME (mobility management entity) (431), PGW ( packet data network gateway) 432, and in addition to eNodeB 433, SGW (Serving Gateway) 434, PCRF (Policy & Charging Rule Function) 435, HSS (Home Subscriber Server) (436), etc. may include or be linked to. Since the data network 430 may have the same configuration as a known network, a detailed description of the data network 430 will be omitted.

Referring back to FIG. 1 , each terminal device 300/1,300/2 is located in any one (or two or more) cells of a plurality of cells constituting the communication network 410 or the data network 420, so that the voice / Data services such as video calls or artificial intelligence services may be provided. These terminal devices 300/1 and 300/2 can be implemented in various forms, such as a smart phone, a smart pad, or a tablet pad. If any one of the terminal devices 300/1 and 300/2 shown in FIG. 1 is a calling terminal, the other terminal device may be a receiving terminal.

FIG. 4 is a diagram illustrating the configuration of the interfacing device 100 shown in FIG. 1 . First, the interfacing apparatus 100 can be implemented in a server group that performs a function to be described below. In addition, the interfacing apparatus 100 may be referred to as an Augmented Communication System (ACS) or an Augmented Communicatin Platform (ACP).

The interfacing apparatus 100 may support a chatting-type call service (hereinafter, may also be referred to as a 'text-type call service') through the aforementioned message. That is, in the interfacing device 100, while the terminal device 300/1 and the terminal device 300/2 attempt or proceed with an audio or video call with each other, any one of the terminal devices 300/1 Alternatively, while maintaining the video call, the other terminal device 300/2 may support communication with the terminal device 300/1 through a chatting method through a message. To this end, the interfacing device 100 may provide a voice from the terminal device 300/1 to the artificial intelligence service providing server 200 to receive a text corresponding thereto, and transmit the received text to the terminal device 300 . /2) can be provided. In addition, the interfacing device 100 may provide a text from the terminal device 300/2 to the artificial intelligence service providing server 200 to receive a corresponding voice, and transmit the received voice to the terminal device 300 . /1) can be provided.

The interfacing apparatus 100 includes an artificial intelligence network interfacing unit 110 , a call network interfacing unit 120 , a data network interfacing unit 130 , and a processor 140 , with reference to FIG. 4 . However, FIG. 4 is only an example, and the interfacing apparatus 100 is not interpreted as being limited to that illustrated in FIG. 4 . For example, the interfacing apparatus 100 may further include a memory or may not include at least one of the components illustrated in FIG. 4 .

Each of the artificial intelligence network interfacing unit 110 , the call network interfacing unit 120 , and the data network interfacing unit 130 transmits packet data such as voice or video data or a chatting message to the artificial intelligence service providing server 200 , the communication network 410 . ) and the data network 420, respectively. To this end, each of the interfacing units 110 to 130 may include a wired or wireless communication module.

The processor 140 may be implemented by a memory for storing instructions programmed to perform a function to be described below, and a microprocessor for executing these instructions.

More specifically, the processor 140 may perform setting for each of the communication network 410 and the data network 420 . In addition, the processor 140 may control a packet such as a voice received from the communication network 410 or a text received from the data network 420 to be delivered to the artificial intelligence service providing server 200 . In addition, the processor 140 may control packets such as voice or text received from the artificial intelligence service providing server 200 to be delivered to the communication network 410 or the data network 240 .

also. The processor 140 estimates whether the output of the voice transmitted to any one of the terminal devices 300/1 and 300/2, for example, the terminal device 300/1, has been completed in the corresponding terminal device 300/1. and when it is estimated that the output is completed, it is possible to control the notification to be delivered to the other terminal device 300/2. To this end, the processor 140 may employ an algorithm for calculating the time it takes for the voice transmitted to the terminal device 300/1 to be output from the terminal device 300/1. Through this, the user of a certain terminal device may recognize whether the voice is being output to another terminal device or whether the output has been completed.

In addition, when the algorithm employed in the above-described TTS module 230, that is, input of text is not completed for a predetermined time, the processor 140 sounds a sound indicating that the completion of the text input is delayed. Algorithms that insert between or in the middle of speech may be employed. Here, according to the embodiment, if the above-described algorithm is not employed in the TTS module 230 , the processor 140 may employ such an algorithm, otherwise, if the above-described algorithm is employed in the TTS module 230 , The processor 140 may not employ such an algorithm.

As described above, according to an embodiment, while the terminal device 300/1 and the terminal device 300/2 are attempting or conducting an audio or video call with each other, any one of the terminal devices 300/1 maintains an audio or video call, but the other terminal device 300/2 may communicate with the terminal device 300/1 through a chatting method through a message.

5 is a diagram illustrating a procedure of an artificial intelligence interfacing method according to an embodiment. This interfacing method can be performed by the interfacing apparatus 100 shown in FIG. 4 . In addition, since the interfacing method shown in FIG. 5 is merely exemplary, the spirit of the present invention is not limited to the one shown in FIG. 5 .

Referring to FIG. 5 , when a first voice is received from a first terminal connected to a communication network, a step (S100) of transmitting the first voice to an artificial intelligence service providing server is performed.

In addition, when the first text obtained by converting the first voice into text is received from the artificial intelligence service providing server, the step of delivering the first text to the second terminal connected to the data network ( S110 ) is performed.

In addition, when the second text is received from the second terminal, the step of delivering the second text to the artificial intelligence service providing server (S120) is performed.

In addition, when the second voice obtained by converting the second text into voice is received from the artificial intelligence service providing server, the step of transmitting the second voice to the first terminal (S13) is performed.

Hereinafter, referring to FIG. 6 , an operation flow for a case in which a request for the aforementioned text type call service is received from the terminal device 300/2 while the terminal devices 300/1 and 300/2 are in an audio or video call with each other. Let's take a look at

Referring to FIG. 6 , a voice or video call is performed between the terminal devices 300/1 and 300/2 through the communication network 410 (①,②).

During execution, a request for a text-type call service may be received from the terminal device 300/2 to the call network 410 (③). As shown in FIG. 7 , when the user touches a predetermined icon 311 on the screen 310 of the terminal device 300/2 as shown in FIG. 7 , the request is transmitted from the terminal device 300/2 to the communication network 410 . may have been transmitted, but such a request may be received in other circumstances, which will be described later.

Referring back to FIG. 6 , the application service node of the communication network 410 , that is, the TAS 412 transmits this request to the interfacing device 100 (④).

Then, the interfacing device 100 controls, through the processor 140 , so that the voice from the terminal device 300/1 is converted into text and transmitted to the terminal device 300/2, as well as the terminal device 300/2. ), the text is converted into voice and transmitted to the terminal device 300/1. Let's take a look at this with reference to ⑤ to ⑭ shown in FIG.

As described above, it is assumed that a request corresponding to ④ is received from the TAS 412 to the interfacing device 100 .

After that, when the user's voice of the terminal device 300/1 is transmitted to the interfacing device 100 through the communication network 410 (⑤, ⑥), the interfacing device 100 transmits the voice to the artificial intelligence service providing server Deliver to (200) (⑦).

The artificial intelligence service providing server 200 converts the voice received in ⑦ into text using the STT module 240 and then transmits the text to the interfacing device 100 (⑧). In this case, the text may include an emoticon indicating the above-described mood or emotion, etc. identified by the STT module 240 for the user of the terminal device 300/1.

Then, the interfacing device 100 transmits the text received in ⑧ to the terminal device 300/2 through the data network 420 (⑨, ⑩). Accordingly, the user of the terminal device 300/2 may receive a text corresponding to the voice uttered by the user of the terminal device 300/1 as a chatting message.

Meanwhile, based on the text received in ⑩, the user of the terminal device 300/2 may input a response message to the text into the terminal device 300/2. The text for the input response message is transmitted to the interfacing device 100 through the data network 420 (⑪, ⑫).

Then, the interfacing device 100 transmits the text received in ⑫ to the artificial intelligence service providing server 200 (⑬).

Then, the artificial intelligence service providing server 200 uses the TTS module 230 to convert the text received in step ⑬ into a voice, and then transmits the voice to the interfacing device 100 (⑭). At this time, the converted voice may reflect the mood or emotion of the user of the terminal device 300/2 obtained by the TTS module 230, and accordingly, the tone or size of the voice may be changed. may have been converted.

Then, the interfacing device 100 transmits the voice received in ⑭ to the terminal device 300/1 through the communication network 410 (⑮, 16). Accordingly, the user of the terminal device 300/21 may receive a voice corresponding to the text input by the user of the terminal device 300/2.

That is, according to an embodiment, while the terminal device 300/1 and the terminal device 300/2 are attempting or conducting a voice or video call with each other, any one of the terminal devices 300/1 is Although the video call is maintained, the other terminal device 300/2 may communicate with the terminal device 300/1 through a chatting method through a message.

Meanwhile, as shown in FIG. 7 , when the user of the terminal device 300/2 directly touches a predetermined icon 311 on the screen 310 of the terminal device 300/2, the above-described text-type call service is possible, but the text-type call service may be started by other methods as well. For example, when the user of the terminal device 300/2 sets the terminal device 300/2 as a 'rest mode (silent mode, non-vibration mode, etc.)', the terminal device 300/2 performs a predetermined voice or video call. After performing for a period of time, even if the user does not touch the icon 311 , the above-described text type call service may be automatically started.

On the other hand, the user of the terminal device 300/2 not only receives the text in ⑪, but also transmits the voice uttered by the user of the terminal device 300/1, that is, to the interfacing device 100 through the processes ⑤ and ⑥. Even the spoken voice can be delivered along with these texts.

In this case, the user of the terminal device 300/2 may wish to receive only text and not voice. To this end, a toggle button for receiving only text and not receiving voice, ie, muting, may be disposed on the screen of the terminal device 300/2, which is illustrated by identification number 312 in FIG. 8 . Whenever the user presses such a toggle button, the user's voice of the terminal device 300/1 may or may not be output.

On the other hand, the natural language processing unit (or natural language processing unit) 220 of the artificial intelligence service providing server 200 may recognize and analyze the question received in ⑦, and then provide a recommended response. The recommendation response provided in this way is provided to the terminal device 300/2 through the data network 420 via the interfacing device 100 . Then, the user of the terminal device 300/2 can transmit the desired text to the user of the terminal device 300/1, which is the counterpart, simply by selecting such a recommendation response. That is, it takes time for the user of the terminal device 300/2 to input text. In order to shorten this time, the recommended response text may be provided to the user of the terminal device 300/2. Through this, the user of the terminal device 300/2 may be provided with a text-type call service more smoothly.

In addition, according to an embodiment, on the screen of the terminal device 300/2, a method toggle button capable of switching between a text-type call service and an audio/video-type call service may be provided, which is identified in FIG. 8 . It is shown as number 313. Whenever the user presses such a toggle button, the terminal device 300/2 may provide a text-type call service or an audio/video-type call service.

On the other hand, if it is assumed that a text-type call service is provided to the terminal device 300/2, the speed at which the user of the terminal device 300/2 inputs text depends on the user of the terminal device 300/1. In many cases, it is relatively slower than the speed at which the voice produces a voice. Accordingly, the TTS module 230 of the artificial intelligence service providing server 200, when the input of text is started and is not completed for a predetermined time, a sound indicating that the input of the text is delayed between voice or voice Can be inserted in the middle. Referring to FIG. 9 , if a predetermined time has elapsed from the start of text input and the end point has not arrived, a predetermined sound may be output from the counterpart's terminal device from the lapse of the predetermined time until just before the end point. Here, the type of sound may include, for example, a word indicating a delay in a conversation, such as 'Ummm...', 'So...', and 'Wait a minute...'.

10 is a diagram illustrating a flow of an artificial intelligence interfacing method according to an embodiment. However, FIG. 10 is only an exemplary bar, and the spirit of the present invention is not construed as being limited to that illustrated in FIG. 10 .

Referring to FIG. 10 , a voice or video call is performed between terminal devices 300/1 and 300/2 through a communication network 410 (S10).

During execution, a request for a text-type call service from the terminal device 300/2 may be received by the call network interfacing unit 120 or the data network interfacing unit 130 . This request may be transmitted from the terminal device 300/2 when the user touches a predetermined icon 311 on the screen 310 of the terminal device 300/2 as shown in FIG. 7 . .

Thereafter, the request in S11 received by the data network interfacing unit 130 is transferred to the processor 140 (S12). Then, the processor 140 requests re-invite to each of the terminal device 300/1 and the terminal device 300/2 through the communication network interfacing unit 120 (S13 to S15). When the re-invite is completed, each terminal device 300/1, 300/2 gives a response indicating that the re-invite is completed (S16, S17)

From then on, the interfacing device 100 controls, through the processor 140 , so that the voice from the terminal device 300/1 is converted into text and transmitted to the terminal device 300/2, as well as the terminal device 300/ Control is performed so that the text from 2) is converted into voice and delivered to the terminal device 300/1. This will be described with reference to S20 to S43 shown in FIG. 10 .

The user's voice of the terminal device 300/1, for example, 'Hello', is transmitted from the terminal device 300/1 through the call network interfacing unit 120 through the artificial intelligence network interfacing unit 110 through the artificial intelligence service providing server ( 200) is transmitted to the STT module 240 (S20 to S22). Then, the STT module 240 outputs a text corresponding to the voice 'hello'. The output text is transmitted to the artificial intelligence network interfacing unit 110 (S24).

In addition, the natural language processing unit 220 derives the recommended response text for the voice received in S22 and delivers it to the artificial intelligence network interfacing unit 110 (S23, S25)

Then, the text received in S24 and S25 is transmitted to the terminal device 300/2 through the data network interfacing unit 130 (S27). Accordingly, the text corresponding to the voice uttered by the user of the terminal device 300/1 is displayed on the terminal device 300/2 (S28). At this time, the recommended response text is also displayed on the terminal device 300/2. The user of the terminal device 300/2 can transmit a desired response to the other party simply by selecting the recommended response text.

Meanwhile, based on the text displayed in S28 , the user of the terminal device 300/2 may input a response message to the text to the terminal device 300/2. The text for the input response message is transmitted to the artificial intelligence network interfacing unit 110 through the data network interfacing unit 130 (S30, S31). In addition, this text is transmitted to the TTS module 230 of the artificial intelligence service providing server 200 through the artificial intelligence network interfacing unit 110 (S32).

Then, the TTS module 230 outputs a voice corresponding to the text. The output voice is transmitted to the artificial intelligence network interfacing unit 110 (S33).

The voice received in S33 is transmitted from the artificial intelligence network interfacing unit 110 to the terminal device 300/1 through the communication network interfacing unit 120 (S34 to S35). Accordingly, a voice corresponding to the text input by the user of the terminal device 300/2 is output to the terminal device 300/1 (S36).

Thereafter, it is recognized by the interfacing device 100 that the output of the voice in S36 is completed. This recognition is based on information received from the terminal device 300/1, or the processor 140 may calculate it. . Thereafter, the completion of voice output in S36 is transmitted to the terminal device 300/2 through the data network interfacing unit 130 (S42). Then, in the terminal device 300/2, the text input by the user of the terminal device 300/2 is displayed to the other party in the form of a voice in a predetermined manner (S43).

As described above, according to an embodiment, while the terminal device 300/1 and the terminal device 300/2 are attempting or conducting an audio or video call with each other, any one of the terminal devices 300/1 maintains an audio or video call, but the other terminal device 300/2 may communicate with the terminal device 300/1 through a chatting method through a message. Also, in this case, it takes time for the user of the terminal device 300/2 to input chatting, and in order to shorten this time, a recommended response text may be provided to the user of the terminal device 300/2. Through this, the user of the terminal device 300/2 may be provided with a text-type call service more smoothly.

On the other hand, the aforementioned method according to the spirit of the present invention is implemented by a computer readable recording medium storing a computer program programmed to perform each step included in the method, or a computer program stored in such a computer readable recording medium possible.

The above description is merely illustrative of the technical idea of the present invention, and various modifications and variations will be possible without departing from the essential quality of the present invention by those skilled in the art to which the present invention pertains. Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

According to an embodiment, the user may be provided with a text-type call service when he/she wants.

100: artificial intelligence interfacing device
200: artificial intelligence service providing server
300/1,2: terminal device

Claims (9)

A call network interfacing unit connected to the call network,
a data network interfacing unit connected to the data network;
An artificial intelligence network interfacing unit connected to an artificial intelligence service providing server;
After the first terminal and the second terminal are connected to the communication network, the first voice received from the first terminal is controlled to be delivered to the artificial intelligence service providing server, and the first text obtained by converting the first voice into text is When received from the artificial intelligence service providing server, the first text is controlled to be delivered to a second terminal connected to the data network, and the second text received from the second terminal is controlled to be delivered to the artificial intelligence service providing server, , When a second voice obtained by converting the second text into voice is received from the artificial intelligence service providing server, a processor for controlling the second voice to be transmitted to the first terminal,
The artificial intelligence network interfacing unit,
Receiving a recommendation response text from the position of the second terminal to the first text from the artificial intelligence service providing server,
The processor is
control so that the recommendation response text is delivered to the second terminal,
The second text is generated based on the selection of the second terminal for the recommendation response text
An interfacing device for artificial intelligence services. The method of claim 1,
The control of the processor for each of the first voice, the first text, the second voice, and the second text,
While the first terminal and the second terminal are performing a voice call through the communication network, if there is a request for a text-type call service from the position of the second terminal,
An interfacing device for artificial intelligence services. The method of claim 1,
The first text is
transmitted to the second terminal together with the first voice
An interfacing device for artificial intelligence services. delete The method of claim 1,
The recommendation response text is,
It is based on the frequency of each case in which the second terminal responded in response to the first text, or the artificial intelligence service providing server recommends based on a plurality of cases based on the first text
An interfacing device for artificial intelligence services. The method of claim 1,
The processor is
Estimating whether the output of the second voice is completed in the first terminal, and controlling to transmit a notification to the second terminal when it is estimated that the output is completed
An interfacing device for artificial intelligence services. The method of claim 1,
The processor is
Obtaining information on a time point at which input of the second text is started and an end point at which the input of the second text is terminated in the second terminal is obtained, and after a predetermined time has elapsed from the time point to the end point, a predetermined value indicating that the second text is being input Controlling the sound to be output from the first terminal
An interfacing device for artificial intelligence services. An artificial intelligence service interfacing method performed by an artificial intelligence service interfacing device, comprising:
After the first terminal and the second terminal are connected to the communication network, when a first voice is received from the first terminal, transmitting the first voice to an artificial intelligence service providing server;
When the first text obtained by converting the first voice into text is received from the artificial intelligence service providing server, transmitting the first text to the second terminal connected to the data network;
Controlling the recommendation response text to be delivered to the second terminal when receiving the recommendation response text from the artificial intelligence service providing server from the position of the second terminal with respect to the first text;
When a second text generated based on the selection of the second terminal for the recommendation response text is received from the second terminal, transmitting the second text to the artificial intelligence service providing server;
When a second voice obtained by converting the second text into voice is received from the artificial intelligence service providing server, transmitting the second voice to the first terminal
A method of interfacing artificial intelligence services. As a computer program stored in a computer-readable recording medium,
When the computer program is executed by the processor included in the interfacing device of the artificial intelligence service,
After the first terminal and the second terminal are connected to a communication network, when a first voice is received from the first terminal, controlling the first voice to be transmitted to an artificial intelligence service providing server;
when the first text obtained by converting the first voice into text is received from the artificial intelligence service providing server, controlling the first text to be transmitted to the second terminal also connected to the data network;
Controlling the recommendation response text to be delivered to the second terminal when receiving the recommendation response text from the artificial intelligence service providing server from the position of the second terminal with respect to the first text;
When a second text generated based on the selection of the second terminal for the recommendation response text is received from the second terminal, controlling the second text to be delivered to the artificial intelligence service providing server;
When a second voice obtained by converting the second text into a voice is received from the artificial intelligence service providing server, controlling the second voice to be transmitted to the first terminal.
A computer program stored in a computer-readable recording medium.

Images