KR-20260062772-A - METHOD OF OPERATION OF A SERVER THAT PROVIDES A GENERATIVE MODEL-BASED CUSTOMER CONSULTING SERVICE SPECIALIZED IN DOMAIN-SPECIFIC KNOWLEDGE

Abstract

The present disclosure relates to a method of operation of a server including a callbot module. The method of operation comprises the steps of: providing a User Interface (UI) for inputting document data related to the enterprise of an enterprise customer to a terminal of an enterprise customer; obtaining first voice data of an enterprise customer from a terminal of an enterprise customer; obtaining second voice data generated based on document data through a callbot module with respect to the first voice data of the enterprise customer; and transmitting the second voice data to the terminal of the enterprise customer.

Inventors

• 유승재

Assignees

• (주)페르소나에이아이

Dates

Publication Date

20260507

Application Date

20241231

Priority Date

20241029

Claims (7)

1. In a method of operation of a server including a callbot module, A step of providing a User Interface (UI) for inputting document data related to the corporate customer's company into the corporate customer's terminal; A step of obtaining the first voice data of the general customer from the general customer's terminal; Regarding the first voice data of the general customer, the step of obtaining second voice data generated based on the document data through the callbot module; and A method of operation of a server comprising the step of transmitting the second voice data to the terminal of the general customer.
2. In paragraph 1, The step of obtaining second voice data generated based on the document data through the above callbot module is: A step of obtaining a first text by converting the first voice data into a text format through a speech-to-text (STT) model included in the callbot module; A step of inputting the first text into a generative model included in the callbot module to obtain the output second text; and A method of operation of a server comprising the step of inputting the second text into a TTS (text to speech) model to obtain the second voice data converted into an audio signal format.
3. In paragraph 2, The step of obtaining the above second text is, A method of operation of a server, wherein the first text and the document data are input into the generative model, and the generative model extracts at least one keyword related to the first text from the document data to obtain a second text generated.
4. In paragraph 2, The step of obtaining second voice data generated based on the document data through the above callbot module is: A step of obtaining evaluation information for the generative model based on the first text and the second text through a conversation evaluation model included in the callbot module; Based on the above evaluation information, a step of selecting one of a plurality of large language models pre-trained with domain-specific knowledge and the above generative model; and A method of operation of a server comprising the step of obtaining the second text through the selected model.
5. In paragraph 4, The step of obtaining evaluation information for the above generative model is, Identifying the general customer's intention information of the general customer's terminal based on the first text through the above conversation evaluation model, and Identifying the similarity between the identified intention information and the second text, and A method of operation of a server that calculates an evaluation score for the generative model based on the similarity identified above.
6. In paragraph 5, Based on the above evaluation information, the step of selecting one of the large language model pre-trained for domain-specific knowledge and the above generative model is, If the evaluation score for the above generative model is above the threshold, the above generative model is selected, and A method of operation of a server that, if the evaluation score of the above generative model is below a threshold, selects a large language model among the plurality of large language models that is pre-trained with knowledge of the field corresponding to the enterprise of the enterprise customer.
7. In paragraph 2, The method of operation of the above server is, A step of generating summary data for the first text and the second text; and A method of operation of a server comprising the step of providing the summary data to the terminal of the corporate customer.

Description

Method of operation of a server that provides a generative model-based customer consulting service specialized in domain-specific knowledge The present disclosure relates to a method of operation of a server including a callbot module, and more specifically, to a method of operation of a server including a callbot module that performs consultation with a general customer based on document data related to an enterprise customer. Recently, an increasing number of companies are adopting AI to improve the quality of customer consultation services. As a result, time for both customers and agents can be saved by handling repetitive or simple inquiries quickly and efficiently through AI. However, while it is necessary to train AI based on large amounts of data to provide high-quality responses depending on the types of products and the complexity of services, difficulties in AI training are arising due to the lack of proper integration among internal customer service channels. Furthermore, since most companies utilize rule-based AI, responses are limited to fixed scenarios, and the recognition rate is low, resulting in significant customer dissatisfaction. Therefore, there is a need to ensure the scalability of conversation construction by moving away from the decision tree approach, which relies solely on fixed scenarios, and instead building a knowledge system that indexes the entire data to enable responses even outside the scope of the scenarios. FIG. 1 is a flowchart illustrating the operation of a server performing consultation with a general customer according to one embodiment of the present disclosure. FIG. 2 is a diagram illustrating the operation of a server communicating with a corporate customer's terminal and a general customer's terminal according to one embodiment of the present disclosure. FIG. 3 is a diagram illustrating the operation of a callbot module generating second voice data based on first voice data according to one embodiment of the present disclosure. FIG. 4 is a drawing for explaining the operation of a server according to one embodiment of the present disclosure. FIG. 5 is a diagram illustrating the operation of a server providing summary data to a corporate customer's terminal according to one embodiment of the present disclosure. FIG. 6 is a block diagram illustrating the configuration of a server according to one embodiment of the present disclosure. Before specifically describing the present disclosure, the method of description in the specification and drawings is described. First, the terms used in this specification and claims have been selected based on general terms considering their functions in the various embodiments of this disclosure. However, these terms may vary depending on the intent of those skilled in the art, legal or technical interpretations, and the emergence of new technologies. Additionally, some terms have been arbitrarily selected by the applicant. Such terms may be interpreted according to the meanings defined in this specification; in the absence of specific definitions, they may be interpreted based on the overall content of this specification and common technical knowledge in the relevant field. In addition, the same reference numbers or symbols described in each drawing attached to this specification represent parts or components that perform substantially the same function. For convenience of explanation and understanding, the same reference numbers or symbols are used to describe different embodiments. That is, even if components having the same reference number are all depicted in multiple drawings, the multiple drawings do not imply a single embodiment. Additionally, in this specification and claims, terms including ordinal numbers, such as "first," "second," etc., may be used to distinguish between components. These ordinal numbers are used to distinguish identical or similar components from one another, and the meaning of the terms should not be limited by the use of such ordinal numbers. For example, the order of use or arrangement of components combined with such ordinal numbers should not be restricted by the number. If necessary, each ordinal number may be used interchangeably. In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "consisting of" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. In the embodiments of the present disclosure, terms such as "module," "unit," "part," etc. are used to refer to a component that performs at least one function or operation, and such component may be implemented in hardware or software, or in a combination of hardware