Method and system for generating review-database broadcast cue sheets

Abstract

The present disclosure provides a review database broadcast cue sheet generation method, performed by at least one processor, that includes receiving user input including information identifying a product, extracting one or more keywords from review data associated with the product, and generating a broadcast cue sheet based on the one or more keywords using a language generation model.

Description

[Technical Field]

[0001] The present disclosure relates to a method and system for generating a broadcast cue sheet, and more particularly to a method and system for extracting keywords from review data related to a product selected by a user and generating a broadcast cue sheet based on the extracted keywords using a language generation model. [Background technology]

[0002] In recent years, as online commerce has grown rapidly with the development of the Internet, the size of online commerce markets such as TV shopping and live commerce has also expanded rapidly. Here, live commerce refers to a form of e-commerce in which products are introduced to and sold to viewers through real-time video streaming. In other words, online commerce can refer to the combination of online shopping and broadcasting that takes place in real time.

[0003] Meanwhile, in order to conduct a broadcast in the live commerce market, a broadcast script (or cue sheet) needs to be created. Traditionally, script creation has been outsourced to professional scriptwriters, broadcast production companies have researched and created related materials, or existing scripts have been recycled. However, these methods can require a lot of time and cost to create a script. In addition, recycling an existing script created for a product other than the new product can result in a decline in broadcast quality. Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure provides a review database-based broadcast cue sheet generation method and system (apparatus) for solving the above-mentioned problems. [Means for solving the problem]

[0005] The present disclosure may be embodied in numerous ways, including as a method, an apparatus (system), or a computer program stored on a computer readable storage medium.

[0006] According to one embodiment of the present disclosure, a method for generating a broadcast cue sheet from a reviews database may include receiving user input including information identifying a product, extracting one or more keywords from review data associated with the product, and generating a broadcast cue sheet based on the one or more keywords using a language generation model.

[0007] A computer-readable non-transitory recording medium may be provided that records instructions for executing a method for generating a cue sheet for broadcasting according to an embodiment of the present disclosure on a computer.

[0008] A system according to one embodiment of the present disclosure includes a communication module, a memory, and at least one processor coupled to the memory and configured to execute at least one computer-readable program contained in the memory, wherein the at least one program includes instructions for receiving user input including information for identifying a product, extracting one or more keywords from review data related to the product, and generating a broadcast cue sheet based on the one or more keywords using a language generation model. [Effects of the Invention]

[0009] According to various embodiments of the present disclosure, a user can conveniently receive a broadcast cue sheet simply by inputting information for identifying a product. In addition, the user can easily reflect the user's intentions in the broadcast cue sheet by modifying the automatically generated keywords or cue sheet. This significantly reduces the time and cost required to create a broadcast cue sheet.

[0010] According to various embodiments of the present disclosure, a user can easily check review keywords for a product by simply inputting information to identify the product. In addition, pre-processing and post-processing processes for answer data generated in response to product-related questions can reduce uncertainty in the subsequent keyword extraction stage and improve the reliability of review keywords.

[0011] According to various embodiments of the present disclosure, rather than directly generating long sentences for a broadcast cue sheet from keywords, by generating short sentences in the middle, it is possible to more appropriately process the sentences. Furthermore, by filtering out prohibited words and the like from the short sentences before generating the long sentences, it is possible to prevent the generation of long sentences that are inappropriate for broadcast.

[0012] According to various embodiments of the present disclosure, it is possible to generate longer sentences capable of conveying meaning from short bullet-type sentences generated from keywords related to a product. Furthermore, by dividing and modularizing a cue sheet into blocks using prompts, it is possible to generate a broadcast cue sheet with natural context and structure. Furthermore, by generating a cue sheet using keywords labeled based on predetermined questions, the structure of the cue sheet is structured, allowing the content of the cue sheet to be effectively conveyed.

[0013] According to various embodiments of the present disclosure, a user can easily understand the contents of a broadcast cue sheet by checking the overall structure of the broadcast cue sheet. Furthermore, a user can use a more complete cue sheet by modifying the contents of the broadcast cue sheet generated by a language generation model. Furthermore, a user can effectively rehearse a live commerce broadcast by playing back the voice synthesis results of the cue sheet that reflect the vocal style of the show host.

[0014] According to various embodiments of the present disclosure, the structure of a broadcast cue sheet can be subdivided / modularized by block. This allows the content of each block that makes up the cue sheet to be created more smoothly, and the overall structure of the cue sheet can be seamlessly connected. Furthermore, the structure of the cue sheet can be flexibly adjusted in response to prompts entered by the user.

[0015] The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by a person having ordinary knowledge in the field of technology to which the present disclosure belongs (referred to as an "ordinary engineer") from the description in the claims. [Brief explanation of the drawings]

[0016] Non-limiting examples of the present disclosure will be described with reference to the accompanying drawings, as described below, in which like reference numerals indicate like elements and in which:

[0017] [Figure 1] FIG. 1 is a diagram illustrating an example of a method for generating a broadcast cue sheet according to an embodiment of the present disclosure.

[0018] [Figure 2] 1 is a schematic diagram showing a configuration in which an information processing system is communicatively connected to multiple user terminals to generate a broadcast cue sheet according to one embodiment of the present disclosure.

[0019] [Figure 3] 1 is a block diagram showing an internal configuration of a user terminal and an information processing system according to an embodiment of the present disclosure.

[0020] [Figure 4] FIG. 1 is a diagram illustrating a procedure of a cue sheet generation method according to an embodiment of the present disclosure.

[0021] [Figure 5] FIG. 10 is a diagram illustrating an example of extracting keywords from review data according to an embodiment of the present disclosure.

[0022] [Figure 6] FIG. 10 is a diagram showing an example of generating a bullet-type short sentence from a keyword according to an embodiment of the present disclosure.

[0023] [Figure 7] FIG. 10 is a diagram illustrating an example in which a broadcast cue sheet is generated according to one embodiment of the present disclosure.

[0024] [Figure 8] FIG. 10 is a diagram illustrating an example of a broadcast cue sheet according to one embodiment of the present disclosure.

[0025] [Figure 9] FIG. 10 is a diagram illustrating an example of a broadcast cue sheet configuration according to one embodiment of the present disclosure.

[0026] [Figure 10] FIG. 10 is a diagram illustrating an example of an opening block of a broadcast cue sheet according to one embodiment of the present disclosure.

[0027] [Figure 11] FIG. 10 is a diagram illustrating an example of a body block of a broadcast cue sheet according to one embodiment of the present disclosure.

[0028] [Figure 12] FIG. 10 is a diagram illustrating an example of a closing block of a broadcast cue sheet according to one embodiment of the present disclosure.

[0029] [Figure 13] 1 is a flowchart illustrating an example of a method according to one embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0030] Hereinafter, specific contents for implementing the present disclosure will be described in detail with reference to the accompanying drawings. However, in the following description, specific descriptions of well-known functions and configurations will be omitted if they may obscure the gist of the present disclosure.

[0031] In the accompanying drawings, identical or corresponding components are denoted by the same reference numerals. In addition, in the following description of the embodiments, duplicated descriptions of identical or corresponding components will be omitted. However, omission of a description of a component does not mean that the component is not included in any of the embodiments.

[0032] The advantages and features of the disclosed embodiments, as well as methods for achieving them, will become apparent from the following examples, taken in conjunction with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below, and may be embodied in various other forms. The present embodiments are provided solely to complete the disclosure and fully convey the scope of the invention to those skilled in the art.

[0033] The terms used in this specification will be briefly explained, and the disclosed embodiments will be specifically described. The terms used in this specification are currently commonly used and general terms that have been selected as much as possible while taking into consideration the functions of the present disclosure. However, these terms may change depending on the intentions of engineers in the relevant field, legal precedents, the emergence of new technologies, etc. In addition, in certain cases, the applicant may arbitrarily select terms, and in such cases, the meanings of these terms will be described in detail in the relevant description of the invention. Therefore, the terms used in this disclosure should be defined based on the meanings of the terms and the overall content of the present disclosure, rather than simply by the names of the terms.

[0034] In this specification, the singular expression includes the plural expression unless the context clearly dictates otherwise. Furthermore, the plural expression includes the singular expression unless the context clearly dictates otherwise. Throughout this specification, when a part includes one element, this does not mean that it may further include other elements, but does not exclude other elements, unless otherwise specified.

[0035] Furthermore, the terms "module" and "module" used in this specification refer to software or hardware components, and the "module" or "module" may perform either function. However, the term "module" or "module" is not limited to software or hardware. A "module" or "module" may be configured to be provided on an addressable storage medium or to execute one or more processors. Thus, as an example, a "module" or "module" may include components such as software components, object-oriented software components, class components, and task components, as well as at least one of processes, functions, attributes, procedures, subroutines, program code segments, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. Components and "modules" or "modules" may be combined into fewer components and "modules" or "modules" so that the functionality provided by them can be further separated into additional components, "modules," or "modules."

[0036] According to one embodiment of the present disclosure, a "module" or "unit" may be embodied as a processor and memory. "Processor" should be broadly interpreted to include a general-purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, etc. In some environments, "processor" may refer to an application specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), etc. "Processor" may also refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors in conjunction with a DSP core, or any other such configuration. Additionally, "memory" should be broadly interpreted to include any electronic component capable of storing electronic information. "Memory" may refer to various types of processor-readable media, such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable-programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, magnetic or optical data storage devices, registers, etc. Memory is in electronic communication with a processor if the processor can read information from and / or store information in the memory. Memory that is integrated into a processor is in electronic communication with the processor.

[0037] In the present disclosure, a "system" may include at least one of a server device and a cloud device, but is not limited thereto. For example, a system may be configured with one or more server devices. As another example, a system may be configured with one or more cloud devices. As yet another example, a system may be configured and operated by both a server device and a cloud device.

[0038] In this disclosure, "display" may refer to any display device associated with a computing device, for example, any display device capable of displaying any information / data controlled by or provided by the computing device.

[0039] In the present disclosure, "each of a plurality of A's" or "each of a plurality of A's" may refer to each of all the components included in the plurality of A's, or may refer to each of some of the components included in the plurality of A's.

[0040] A "machine learning model" may include any model used to infer an answer to a given input. According to one embodiment, the machine learning model may include an artificial neural network model including an input layer, multiple hidden layers, and an output layer. Here, each layer may include multiple nodes. In this disclosure, each of multiple machine learning models is described as a separate machine learning model, but this is not limited thereto. Some or all of the multiple machine learning models may be embodied as a single machine learning model. Furthermore, a single machine learning model may include multiple machine learning models. In this disclosure, the terms machine learning model and artificial neural network model may be used interchangeably to refer to the same or similar models. In addition, in this disclosure, a "language model" or a "language generation model" may refer to a machine learning model or an artificial neural network model configured to calculate a probability for at least a portion of one or more word sequences or sentences, or to generate a word sequence or a portion of a sentence.

[0041] 1 illustrates an example method for generating a broadcast cue sheet 124 according to one embodiment of the present disclosure. A first operation 110 and a second operation 120 illustrated in FIG. 1 illustrate actions performed by a user via a user interface displayed on a display of a user terminal to generate a broadcast cue sheet.

[0042] Specifically, the first operation 110 exemplifies an operation in which a user inputs information 112 for identifying a product. The user can input information 112 for identifying a product to be introduced in a broadcast intended for product sales or product advertising, such as live commerce. Here, the information 112 for identifying a product may include a product name, a product number, a catalog ID related to a predetermined product, etc. Furthermore, a cue sheet related to the product may be generated by the user clicking or selecting a "generate cue sheet" button 114.

[0043] A second operation 120 illustrates an operation in which a cue sheet 124 related to the product is generated and output. One or more keywords may be extracted from review data related to the product identified by the user. A broadcast cue sheet may be generated from the extracted keywords. An example of the process of extracting keywords will be described in detail below with reference to FIG. 5.

[0044] In one embodiment, keywords may be provided to a user as selling points 122. Here, the selling points 122 may include product features, main customers and usage methods of the product, main ingredients and characteristics of the product, etc. For example, product features such as "brush is hidden inside" and "give a clean finish without drying out" may be output as selling points 122. Furthermore, main customers and usage methods of the product such as "recommended for acne scars" and "recommended for combination skin" may be output as selling points 122. Furthermore, main ingredients and characteristics of the product such as "sebum control" and "UV protection function" may be output as selling points 122. If there are no appropriate keywords that can be provided as selling points 122, the user may directly input keywords, or selling points may be generated from review data of other products similar to the product using a language model.

[0045] In one embodiment, a language generation model may be used to generate a broadcast cue sheet 124 based on one or more keywords. Specifically, bullet-type sentences including one or more keywords may be generated. Here, the bullet-type sentences may include sentences combining one or more keywords with nouns related to the product, or sentences combining one or more keywords with adjectives related to the product. Further, to convey the meaning of the broadcast cue sheet, complete sentences may be generated by adding sentence components to the bullet-type sentences. Furthermore, the broadcast cue sheet 124 may be generated based on prompts related to the configuration of the broadcast cue sheet and the generated sentences.

[0046] In one embodiment, the user can regenerate the cue sheet by clicking or selecting the "Regenerate Comments" button 126. Specifically, the user can modify the keywords in the automatically generated selling points 122. In this case, the user can have a new cue sheet generated based on the modified keywords by using the "Regenerate Comments" button 126. Additionally or alternatively, the user can modify the automatically generated cue sheet 124 directly. Additionally or alternatively, the user can have a new cue sheet provided by using the "Regenerate Comments" button 126 without modifying the selling points 122 and cue sheet 124.

[0047] The method of the present disclosure may be applied to various fields other than broadcast cue sheets. For example, the method of generating a script such as a cue sheet using the language generation model of the present disclosure may be applied to fields for creating different types of long texts, such as speeches, announcements, curator scripts, narration scripts, and film scripts, which require structured sentences based on specific themes or keywords. These may be used for various services such as speeches, presentations, docents, curators, narrators, TTS-based virtual avatars, virtual humans, videos, short forms, and video subtitle generation. The method may also be used for collecting materials and data for creative activities such as writing, and for creating drafts.

[0048] With this configuration, users can conveniently receive a broadcast cue sheet simply by entering information to identify a product. Furthermore, users can easily reflect their intentions in the broadcast cue sheet by modifying the automatically generated keywords or cue sheet. This significantly reduces the time and cost required to create a broadcast cue sheet.

[0049] 2 is a schematic diagram illustrating a configuration in which an information processing system 230 is communicably connected to a plurality of user terminals 210_1, 210_2, and 210_3 to generate a broadcast cue sheet according to an embodiment of the present disclosure. As shown, the plurality of user terminals 210_1, 210_2, and 210_3 may be connected to the information processing system 230, which can provide a broadcast cue sheet generation service, via a network 220. Here, the plurality of user terminals 210_1, 210_2, and 210_3 may include terminals of users who wish to receive the broadcast cue sheet generation service.

[0050] In one embodiment, information processing system 230 may include one or more server devices and / or databases, or one or more cloud computing service-based distributed computing devices and / or distributed databases, that can store, provide, and execute computer-executable programs (e.g., downloadable applications) and data related to providing a broadcast cue sheet generation service, etc.

[0051] The broadcast cue sheet generation service provided by the information processing system 230 may be provided to a user through a broadcast cue sheet generation service application web browser or a web browser extension program installed in each of the user terminals 210_1, 210_2, and 210_3. For example, the information processing system 230 may provide information corresponding to a broadcast cue sheet generation request received from the user terminals 210_1, 210_2, and 210_3 through the broadcast cue sheet generation service application or perform corresponding processing.

[0052] A plurality of user terminals 210_1, 210_2, and 210_3 can communicate with the information processing system 230 through the network 220. The network 220 may be configured to enable communication between the plurality of user terminals 210_1, 210_2, and 210_3 and the information processing system 230. Depending on the installation environment, the network 220 may be configured as a wired network such as Ethernet, a wired home network (Power Line Communication), a telephone line communication device, and RS-serial communication, a mobile communication network, a wireless network such as WLAN (Wireless LAN), Wi-Fi (registered trademark), Bluetooth (registered trademark), and ZigBee (registered trademark), or a combination thereof. The communication method is not limited, and may include a communication method utilizing a communication network that the network 220 may include (e.g., a mobile communication network, a wired Internet, a wireless Internet, a broadcast network, a satellite network, etc.), as well as short-range wireless communication between the user terminals 210_1, 210_2, and 210_3.

[0053] 2, a mobile phone terminal 210_1, a tablet terminal 210_2, and a PC terminal 210_3 are shown as examples of user terminals, but are not limited thereto, and the user terminals 210_1, 210_2, and 210_3 may be any computing device capable of wired and / or wireless communication and on which a broadcast cue sheet generation service application or a web browser, etc., can be installed and executed. For example, the user terminal may include an AI speaker, a smartphone, a mobile phone, a navigation system, a computer, a laptop, a digital broadcast terminal, a PDA (Personal Digital Assistant), a PMP (Portable Multimedia Player), a tablet PC, a game console, a wearable device, an IoT (Internet of Things) device, a VR (Virtual Reality) device, an AR (Augmented Reality) device, a set-top box, etc. Also, although FIG. 2 shows three user terminals 210_1, 210_2, and 210_3 communicating with the information processing system 230 through the network 220, this is not limited thereto, and other numbers of user terminals may be configured to communicate with the information processing system 230 through the network 220.

[0054] FIG. 3 is a block diagram illustrating the internal configuration of a user terminal 210 and an information processing system 230 according to an embodiment of the present disclosure. The user terminal 210 may refer to any computing device capable of executing applications, a web browser, and the like and capable of wired / wireless communication, and may include, for example, the mobile phone terminal 210_1, the tablet terminal 210_2, and the PC terminal 210_3 of FIG. 2 . As illustrated, the user terminal 210 may include a memory 312, a processor 314, a communication module 316, and an input / output interface 318. Similarly, the information processing system 230 may include a memory 332, a processor 334, a communication module 336, and an input / output interface 338. As illustrated in FIG. 3 , the user terminal 210 and the information processing system 230 may be configured to communicate information and / or data over the network 220 using their respective communication modules 316 and 336. Additionally, the input / output device 320 may be configured to input information and / or data to the user terminal 210 and output information and / or data generated from the user terminal 210 via the input / output interface 318 .

[0055] The memory 312, 332 may include any non-transitory computer-readable recording medium. According to one embodiment, the memory 312, 332 may include a permanent mass storage device such as a read only memory (ROM), a disk drive, a solid state drive (SSD), a flash memory, etc. As another example, a non-transitory mass storage device such as a ROM, an SSD, a flash memory, a disk drive, etc. may be included in the user terminal 210 or the information processing system 230 as a permanent storage device separate from the memory. The memory 312, 332 may also store an operating system and at least one program code.

[0056] Such software components may be loaded from a computer-readable recording medium separate from the memory 312, 332. Such separate computer-readable recording medium may include a recording medium directly connectable to the user terminal 210 and the information processing system 230, but may also include a computer-readable recording medium such as a floppy drive, a disk, a tape, a DVD / CD-ROM drive, or a memory card. As another example, the software components may be loaded into the memory 312, 332 via the communication module 316, 336 rather than from a computer-readable recording medium. For example, at least one program may be loaded into the memory 312, 332 based on a computer program installed by a file provided over the network 220 by a developer or a file distribution system that distributes application installation files.

[0057] The processors 314, 334 may be configured to process computer program instructions by performing basic arithmetic, logic, and input / output operations. The instructions may be provided to the processors 314, 334 by the memory 312, 332 or the communication modules 316, 336. For example, the processors 314, 334 may be configured to execute instructions received by program code stored in a storage device, such as the memory 312, 332.

[0058] The communication modules 316 and 336 may provide a configuration or function for the user terminal 210 and the information processing system 230 to communicate with each other via the network 220, and may provide a configuration or function for the user terminal 210 and / or the information processing system 230 to communicate with other user terminals or other systems (e.g., a separate cloud system, etc.). As an example, a request or data (e.g., a request to generate a cue sheet for broadcasting) generated by the processor 314 of the user terminal 210 via program code stored in a storage device such as the memory 312 may be transmitted to the information processing system 230 via the network 220 under the control of the communication module 316. Conversely, a control signal or command provided by the processor 334 of the information processing system 230 may be received by the user terminal 210 via the communication module 316 of the user terminal 210 via the communication module 336 and the network 220.

[0059] The input / output interface 318 may be a means for interfacing with the input / output device 320. For example, the input device may include a camera including an audio sensor and / or an image sensor, a keyboard, a microphone, a mouse, etc., and the output device may include a display, a speaker, a haptic feedback device, etc. As another example, the input / output interface 318 may be a means for interfacing with a device that integrates input and output configurations or functions, such as a touchscreen. For example, when the processor 314 of the user terminal 210 processes instructions of a computer program loaded in the memory 312, a service screen configured using information and / or data provided by the information processing system 230 or another user terminal may be displayed on the display through the input / output interface 318. Although FIG. 3 illustrates the input / output device 320 as not being included in the user terminal 210, this is not limiting and the input / output device 320 may be configured as a single device together with the user terminal 210. Furthermore, the input / output interface 338 of the information processing system 230 may be a means for interfacing with an input or output device (not shown) that may be connected to or included in the information processing system 230. Although the input / output interfaces 318 and 338 are shown in FIG. 3 as elements configured separately from the processors 314 and 334, the present invention is not limited thereto, and the input / output interfaces 318 and 338 may be configured to be included in the processors 314 and 334.

[0060] The user terminal 210 and the information processing system 230 may include more components than those shown in FIG. 3 . However, it is not necessary to explicitly show most of the conventional components. In one embodiment, the user terminal 210 may be embodied to include at least some of the input / output devices 320 described above. The user terminal 210 may also include other components such as a transceiver, a global positioning system (GPS) module, a camera, various sensors, a database, etc. For example, if the user terminal 210 is a smartphone, it may include components typically included in a smartphone, such as an acceleration sensor, a gyro sensor, a microphone module, a camera module, various physical buttons, buttons using a touch panel, an input / output port, a vibrator for vibration, etc.

[0061] While a program for a broadcast cue sheet generation service application or the like is running, the processor 314 can receive text, images, pictures, sounds, and / or actions, etc., entered or selected through an input device such as a touch screen, keyboard, camera including an audio sensor and / or image sensor, microphone, etc., connected to the input / output interface 318, and can store the received text, images, pictures, sounds, and / or actions, etc. in the memory 312 or provide them to the information processing system 230 via the communication module 316 and the network 220.

[0062] The processor 314 of the user terminal 210 may be configured to manage, process, and / or store information and / or data received from the input / output device 320, other user terminals, the information processing system 230, and / or multiple external systems. The information and / or data processed by the processor 314 may be provided to the information processing system 230 via the communication module 316 and the network 220. The processor 314 of the user terminal 210 may transmit and output information and / or data to the input / output device 320 via the input / output interface 318. For example, the processor 314 may display the received information and / or data on the screen of the user terminal 210.

[0063] The processor 334 of the information processing system 230 may be configured to manage, process, and / or store information and / or data received from multiple user terminals 210 and / or multiple external systems. The information and / or data processed by the processor 334 may be provided to the user terminal 210 via the communication module 336 and the network 220.

[0064] 4 is a diagram illustrating a procedure of a cue sheet generation method according to an embodiment of the present disclosure. In one embodiment, review data of products sold by a seller with broadcasting rights may be collected from blogs, online shopping sites, etc. (410). Furthermore, one or more frequently occurring keywords may be extracted from the review data related to the products (412). In this case, the extracted keywords may be stored in a keyword database 420.

[0065] In one embodiment, user input including information for identifying a product (e.g., product name, etc.) may be received (430). If keywords related to the product are stored in keyword database 420, the keywords stored in database 420 may be extracted (450). Alternatively, if keywords related to the product are not present in keyword database 420, keywords may be extracted from review data of other products similar to the product using language model 440 (450). Alternatively, the user may directly input keywords related to the product.

[0066] In one embodiment, a bullet-type short sentence may be generated based on the extracted keywords (460). Here, the bullet-type short sentence may include a short sentence combining one or more keywords with a noun related to a product, or a short sentence combining one or more keywords with an adjective related to a product. In this case, the nouns and / or adjectives related to the combined products may be combined before or after the keyword. In addition, predetermined prohibited words that cannot be used in broadcasting may be removed from the bullet-type short sentence.

[0067] In one embodiment, a longer sentence may be generated based on the bullet-type short sentence (470). Here, the longer sentence may refer to a sentence in which sentence components are added to the bullet-type short sentence so that the meaning of the broadcast cue sheet can be conveyed. That is, the longer sentence may be a complete sentence generated from the bullet-type short sentence. Furthermore, the longer sentence may be generated to reflect the writing style of the show host.

[0068] In one embodiment, a broadcast cue sheet may be generated (480) based on a long sentence and a prompt associated with the configuration of the broadcast cue sheet. Here, the prompt may include information associated with at least one of an opening block, a body block, or a closing block. Furthermore, a script may be generated by adding broadcast comments to the generated long sentence according to the configuration associated with the prompt. A cue sheet may be completed by combining multiple such scripts. Furthermore, the generated cue sheet may be stored in a cue sheet database 490.

[0069] All or part of the steps of the cue sheet generation method described above may be performed by a machine learning model, a language model, or a user associated with broadcast cue sheet generation.

[0070] FIG. 5 is a diagram illustrating an example of extracting keywords from review data 510 according to an embodiment of the present disclosure. In one embodiment, review data 510 may be collected from online shopping sites, blogs, community sites, internet news, company websites related to the product, etc. in response to information (e.g., product name, product number, catalog ID, etc.) for a user to identify a product. Furthermore, review data 510 may include not only general text but also text converted from audio included in review videos posted on the internet. For example, if a user identifies "Cosmetics A," review data related to the product may be collected, such as, "I gave this in the Melon Pop color as a birthday gift to a friend with cool-toned skin in the summer, and she was very happy with it and said it looked beautiful."

[0071] In one embodiment, the review data 510 may be informative review data collected by filtering spam / promotional review data from the overall review data. Here, informative review data may refer to review data including review information from actual users regarding various characteristics of a product. Also, only review data generated within a predetermined period (e.g., the past year) may be collected.

[0072] In one embodiment, the collected review data 510 may be preprocessed by using a language model to determine whether the review data 510 contains answers to at least some of the predetermined questions 530. Specifically, a Question Semantic Matching Model (QSM) may be used to determine whether answer data corresponding to at least some of the predetermined questions 530 can be extracted from the review data 510. Here, the question semantic matching model may be a machine learning model that learns from document data (or review data), question data, and a correct answer dataset to determine whether answers corresponding to the question data exist in the document data. In addition, the predetermined questions 530 may be questions related to the intended use of the product, the purchase intention of the product, the product's features, the product's drawbacks, purchase history / plans, related products / brands mentioned together with the product, where the product was purchased, the route of product awareness, product ingredients / application technology, product appearance, product nickname, product usage method, product collaboration / project, etc.

[0073] In one embodiment, when the review data 510 includes answers to at least some of the predetermined plurality of questions 530, a language model may be used to extract answer data corresponding to at least some of the predetermined plurality of questions 530 from at least some of the review data 510. Here, the language model for generating the answer data may be trained using a predetermined training dataset.

[0074] Specifically, question data and a document dataset may be input to a first generative model. In this case, the first generative model may pseudo-label at least a portion of the document data as first answer data to a specific question in the question data. Here, the first answer data may include noise answers that are inappropriate for the specific question. To reduce such noise answers, a second generative model may be further utilized. Specifically, a portion of the first answer data may be inspected and noise removed therefrom. For example, inspection of the answer data may be performed passively by a system operator. As a result, the inspected answer data may be considered correct answer data. Here, for efficient inspection, it is not necessary to inspect all of the first answer data. Furthermore, a second generative model may be trained using a specific question corresponding to the first answer data and the inspected answer data. Thereafter, the remaining uninspected portion of the first answer data is input to the second generative model, and the second generative model may label the portion of the first answer data as second answer data. Response data may be generated from the review data 510 using the second generative model trained in this way.

[0075] In one embodiment, the response data may be post-processed, such as by deleting duplicate sentences in the response data, deleting sentences that are inclusive of each other in the response data, or deleting sentences that differ from the original review data. Furthermore, similar response data may be clustered into at least one group, and representative keywords 540 for each of the at least one group may be extracted. In this case, the representative keywords 540 may be determined based on the frequency of keywords included in at least one group in the review data. For example, from the review data 510 for "Cosmetics A," a representative keyword 540 such as "As a birthday gift for a friend with cool-toned skin in summer" may be extracted in response to a question 530 related to the purchase intention of the product, such as "What was the purpose of purchasing this product?"

[0076] This configuration allows users to easily check review keywords for a product by simply entering information to identify the product. Furthermore, pre-processing and post-processing of answer data generated in response to product-related questions reduces uncertainty in the subsequent keyword extraction stage, improving the reliability of review keywords.

[0077] FIG. 6 illustrates an example of generating short bullet-type sentences from keywords according to an embodiment of the present disclosure. In one embodiment, keywords 620 extracted from review data 610 may be clustered. Specifically, each of the keywords 620 may be converted into an embedding vector using an artificial neural network-based model. The embedding vectors may then be clustered into at least one group based on the distance between the embedding vectors. For example, multiple embedding vectors may be clustered using, but not limited to, a K-means algorithm.

[0078] In one embodiment, a representative keyword 630 may be extracted from each of at least one group. Here, the representative keyword 630 may be determined based on the frequency of the keywords 620 included in at least one group. For example, if the keywords 620 of the clustered groups are "moderately strong flavor," "thick and delicious soup," and "deep and rich flavor," the representative keyword 630 may be determined as "deep and strong flavor" based on the frequency of the keywords 620.

[0079] In one embodiment, a bullet-type sentence 640 may be generated based on the representative keyword 630. Here, the bullet-type sentence may include a sentence combining a keyword and a noun related to the product, or a sentence combining a keyword and an adjective related to the product. For example, if the product is "Gomtang" and the representative keyword is "deep and rich in flavor," a bullet-type sentence such as "The soup is deep, rich, and delicious for the price" may be generated. Furthermore, the bullet-type sentence may include a sentence combining a keyword and a timely sentence. For example, a bullet-type sentence such as "The soup is deep and rich in flavor, perfect for the cold winter" may be generated. Furthermore, predetermined taboo words may be removed from the bullet-type sentence.

[0080] With this configuration, rather than directly generating long sentences for broadcast cue sheets from keywords, short sentences are generated along the way, allowing for more appropriate processing of sentences. Furthermore, before generating long sentences, prohibited words and the like can be filtered from the short sentences, preventing the generation of long sentences that are inappropriate for broadcast.

[0081] FIG. 7 is a diagram illustrating an example of generating a broadcast cue sheet according to an embodiment of the present disclosure. In one embodiment, original review data related to a product (e.g., "Dumpling A") may be collected (710) from an online shopping site, blog, etc. One or more keywords may be extracted from the original review data (720). One or more keywords may be clustered based on the similarity between the one or more keywords (730). In this case, a representative keyword may be extracted from each clustered group by summarizing the keywords included in the clustered groups (740). For example, a representative keyword "always stocked in the freezer" may be extracted from a group containing the keywords "always stocked in the home," "should always be stocked in the freezer," and "always in the freezer."

[0082] In one embodiment, a bullet-type short sentence may be generated based on the representative keyword (750). Specifically, the bullet-type short sentence may be generated by combining the representative keyword with a noun or adjective related to the product. Alternatively, the bullet-type short sentence may be generated by processing the representative keyword to be suitable for live commerce broadcasting. For example, if the representative keyword is "always stocked in the freezer," a bullet-type short sentence such as "should always be kept in the freezer" may be generated. As another example, if the representative keyword is "children," a bullet-type short sentence such as "a product that children also like" may be generated.

[0083] In one embodiment, a longer sentence and a cue sheet may be generated based on the bullet-type sentence (760). Specifically, the longer sentence may be generated by adding sentence components to the bullet-type sentence so that the meaning of the sentence in the broadcast cue sheet can be conveyed. Here, the sentence components may include a subject, an object, an adverbial phrase, a predicate, etc. For example, if the bullet-type sentence is "This should always be kept in the freezer," a longer sentence may be generated such as "This is a product that should always be kept in our freezer." As another example, if the bullet-type sentence is "This product is also popular with children," a longer sentence may be generated such as "This product is particularly popular with children, so it is often purchased as a snack for children." Furthermore, a machine learning model may reflect the writing style of the show host in the generated longer sentence.

[0084] In one embodiment, a broadcast cue sheet may be generated based on the long text, and further prompts related to the construction of the broadcast cue sheet may be required, where the prompts may include information related to at least one of an opening block, a body block, or a closing block.

[0085] In one embodiment, the cue sheet may be structured by using keyword labels extracted from the review data. For example, the cue sheet may include a product introduction structure based on keywords related to product features and nicknames, a product introduction structure based on keywords related to product usage, purchasing intentions, and intended use, a product introduction structure related to the main ingredients and appearance of the product, etc. Under such a structure, the cue sheet structure may be effectively communicated according to category.

[0086] With this configuration, longer sentences capable of conveying meaning may be generated from short bullet-type sentences generated from keywords related to a product. Furthermore, by dividing and modularizing a cue sheet into blocks using prompts, a broadcast cue sheet with natural sentence context and structure may be generated. Furthermore, by generating a cue sheet using keywords labeled based on predetermined questions, the structure of the cue sheet may be structured, allowing the content of the cue sheet to be effectively conveyed.

[0087] 8 is a diagram illustrating an example of a broadcast cue sheet 814 according to one embodiment of the present disclosure. A first operation 810 and a second operation 820 illustrated in FIG. 8 illustrate an example of an operation in which the broadcast cue sheet is output via a user interface displayed on a display of a user terminal.

[0088] Specifically, first operation 810 is an example in which a broadcast cue sheet 814 generated according to the present disclosure is output. By specifying a product (e.g., "Airy Powder Primer 8.5g"), the user can check selling points 812 and cue sheet 814 generated based on review data for the product on the display of the user terminal.

[0089] In one embodiment, the selling points 812 may include product features, main customers and usage methods of the product, main ingredients and characteristics of the product, etc. For example, product features such as "brush is hidden inside" and "provides a clean finish without drying out" may be output as selling points 812, main customers and usage methods of the product such as "recommended for acne scars" and "recommended for combination skin" may be output as selling points 812, and main ingredients and characteristics of the product such as "sebum control" and "UV protection" may be output as selling points 812. If there are no appropriate keywords for the selling points 812, the user may directly input keywords, or selling points may be generated from review data of other products similar to the product using a language model.

[0090] In one embodiment, a language generation model may be used to generate a broadcast cue sheet 814 based on keywords extracted from the review data. In this case, the broadcast cue sheet 814 may include paragraphs corresponding to an opening block, a body block, and a closing block. The broadcast cue sheet 814 may also reflect the writing style of the show host. Additionally, the cue sheet 814 may include a dialogue format for two or more speakers.

[0091] In one embodiment, the user can regenerate the cue sheet by, for example, selecting the "Regenerate Comments" button 816. Specifically, the user can modify the keywords in the selling points 812 or enter required keywords that must be included in the cue sheet. In this case, the user can have an updated cue sheet provided based on the modified or required keywords via the "Regenerate Comments" button 816. Additionally or alternatively, the user can modify the cue sheet 814 directly.

[0092] In one embodiment, the cue sheet 814 may be updated to reflect the speaker's characteristics. When a user inputs personal information about a show host (e.g., age, gender, specialty, etc.), the cue sheet may be updated to include expressions appropriate for the show host by reflecting the personal information. For example, if the user inputs "a woman in her 40s with children" as the show host's personal information, expressions that empathize with child-rearing may be reflected in the cue sheet. Also, a show host suitable for a live commerce broadcast may be recommended from a show host database based on the type of product and the cue sheet 814. Furthermore, a voice synthesis model may be used to generate a voice for the show host to read the broadcast cue sheet 814.

[0093] A second operation 820 illustrates an example of a detailed configuration of a broadcast cue sheet in table form. The cue sheet configuration 822 can be divided into an opening block, at least one body block, and a closing block. Each block may include detailed configurations 824. For example, the opening block may include detailed configurations such as an introductory greeting, a product summary, a price summary, a volume guide, a response prompt, and an opening greeting. The body block may include detailed configurations such as the beginning of a product introduction, the product configuration, and the product price. The closing block may include detailed configurations such as a question and answer session, promotional comments, reminders, and a closing greeting. Furthermore, comments 826 corresponding to the detailed configurations may be created for each detailed configuration 824. The user can review and modify the comments 826 corresponding to the detailed configurations.

[0094] With this configuration, users can easily understand the contents of a broadcast cue sheet by checking the overall structure of the broadcast cue sheet. Users can also use a more complete cue sheet by modifying the contents of the broadcast cue sheet generated by the language generation model. Furthermore, users can effectively rehearse a live commerce broadcast by playing back the voice synthesis results of the cue sheet that reflect the vocal style of the show host.

[0095] FIG. 9 is a diagram illustrating an example of a broadcast cue sheet configuration according to an embodiment of the present disclosure. As illustrated, the broadcast cue sheet may include an opening block 910, a body block 920, and a closing block 930. Each block may include detailed configurations. For example, the opening block 910 may include detailed configurations such as an opening greeting, a concept (e.g., a season-based, trend-based, or theme-based), a broadcast product introduction, an event introduction, and a broadcast opening comment. The body block 920 may include detailed configurations such as at least one product introduction and price introduction, a communication guideline with viewers, an event, a product trial, and answers to questions. The closing block 930 may include detailed configurations such as a broadcast product and price reminder, a communication guideline with viewers, an announcement of event winners, a product promotion comment, and a closing remark.

[0096] In one embodiment, a broadcast cue sheet may be generated based on prompts related to the structure or detailed structure of the cue sheet. For example, when a prompt related to an opening block is input to a language generation model, a script corresponding to the opening block may be generated. As another example, when a prompt related to a product introduction structure of a body block is input to a language generation model, a script corresponding to the product introduction structure of the body block may be generated.

[0097] The cue sheet configuration shown in FIG. 9 is merely an example and is not limiting. Some configurations may be deleted or new configurations may be added depending on the broadcast time. The order of the configurations in the cue sheet may also be changed by changing the input order of the prompts. Furthermore, by inputting the expected broadcast time into the prompts, a script appropriate for the expected broadcast time may be generated.

[0098] This structure allows the structure of a broadcast cue sheet to be subdivided and modularized into blocks. This allows the content of each block that makes up the cue sheet to be created more smoothly, and the structure of the entire cue sheet can be naturally connected. In addition, the structure of the cue sheet can be flexibly adjusted according to the prompts entered by the user.

[0099] FIG. 10 is a diagram illustrating an example of an opening block of a broadcast cue sheet according to an embodiment of the present disclosure. In one embodiment, the opening block may include an "introduction" section that starts the broadcast. The opening block may also include a "product summary" and a "price summary" section that summarizes the products and their prices. In this case, the regular price and special price of the products may be values ​​entered by the user. The opening block may also include a "volume guide" section that prompts viewers to check the volume. The opening block may also include a "response guidance" section that uses the characteristics of live commerce broadcasting to guide viewers' reactions, and an "opening comment" section that starts the broadcast.

[0100] FIG. 11 is a diagram illustrating an example of a body block of a broadcast cue sheet according to an embodiment of the present disclosure. In one embodiment, the body block may include a "Product Introduction Start" section that mentions the name of the product being sold. The body block may also include at least one "Product Composition" section that introduces products by product type. In this case, the comments included in the "Product Composition" section may be text generated based on review data. The body block may also include a "Product Price" section that provides information on the product's existing price and special price.

[0101] FIG. 12 illustrates an example of a closing block of a broadcast cue sheet according to an embodiment of the present disclosure. In one embodiment, the closing block may include a "Communication / Question and Answer" section that guides a Q&A or communication with viewers. The closing block may also include a "Promotional Comments" section that promotes the live commerce broadcast to viewers. The closing block may also include a "Reminder" section that summarizes the broadcast content. Finally, the closing block may include a "Closing Greetings" section that greets viewers while concluding the broadcast.

[0102] The cue sheet configurations shown in Figures 10 to 12 are merely examples and are not limited to these. For example, new configurations may be added depending on the type of product, broadcast time, etc., some configurations may be deleted, or the order between configurations may be changed.

[0103] 13 is a flowchart illustrating an example of a method 1300 according to an embodiment of the present disclosure. In one embodiment, the method 1300 may be performed by at least one processor. The method 1300 may begin with the processor receiving user input including information identifying a product (S1310).

[0104] The processor may then extract one or more keywords from the review data related to the product (S1320). Specifically, the processor may use the language model to generate at least one answer data from at least a portion of the review data based on a plurality of predetermined questions. The processor may also extract review keywords related to the product based on the at least one answer data.

[0105] The processor may then use the language generation model to generate a broadcast cue sheet based on one or more keywords (S1330). Specifically, the processor may generate short bullet-type sentences including the one or more keywords. The processor may also generate longer sentences based on the short bullet-type sentences. The processor may also receive prompts related to the configuration of the broadcast cue sheet. Here, the prompts may include information related to at least one of an opening block, a body block, or a closing block. The processor may then generate a broadcast cue sheet based on the longer sentences and the prompts.

[0106] In one embodiment, the processor can convert answer data to a plurality of predetermined questions into embedding vectors. The processor can also generate at least one group based on the distance between the embedding vectors. The processor can also extract representative keywords from each of the at least one group. In this case, the processor can determine the representative keywords based on the frequency of keywords included in the at least one group in the review data.

[0107] In one embodiment, the processor may remove predetermined prohibited words from the bullet-type short sentences. Here, the bullet-type short sentences may include short sentences combining one or more keywords with nouns related to the product, or short sentences combining one or more keywords with adjectives related to the product. Additionally or alternatively, the bullet-type short sentences may include short sentences combining one or more keywords with timely short sentences.

[0108] In one embodiment, the processor can generate longer sentences by adding sentence components to the short bullet sentences to convey the meaning of the broadcast cue sheet, and the processor can use a machine learning model to generate longer sentences based on the short bullet sentences that reflect the writing style of the show host.

[0109] In one embodiment, the processor may output at least one of one or more keywords or a broadcast cue sheet. The processor may also modify the one or more keywords in response to user input modifying the one or more keywords. In this case, the processor may update the broadcast cue sheet based on the modified one or more keywords.

[0110] In one embodiment, the processor can generate the voice of the show host from a broadcast cue sheet using a speech synthesis model, and the processor can respond to user input related to the show host's personal information by updating the broadcast cue sheet based on the show host's personal information.

[0111] The above-described method may be provided as a computer program stored on a computer-readable recording medium for execution by a computer. The medium may be a medium that continuously stores a computer-executable program or a medium that temporarily stores the program for execution or download. The medium may be various recording or storage means in the form of a single piece of hardware or a combination of multiple pieces of hardware. The medium is not limited to a medium directly connected to a computer system but may be distributed over a network. Examples of media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and media configured to store program instructions, including ROM, RAM, and flash memory. Other examples of media include recording or storage media managed by app stores that distribute applications or by websites or servers that provide or distribute various software.

[0112] The methods, operations, or techniques of the present disclosure may be embodied in various ways. For example, such techniques may be embodied as hardware, firmware, software, or a combination thereof. Those of ordinary skill in the art will appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the present disclosure may be embodied as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is embodied as hardware or software depends on the particular application and design requirements imposed on the overall system. Those of ordinary skill in the art may also implement the described functionality in various ways for each particular application, but such implementations should not be interpreted as departing from the scope of the present disclosure.

[0113] In a hardware implementation, the processing units used in performing the techniques may be embodied within one or more ASICs, DSPs, digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to have the functionality described in this disclosure, computers, or combinations thereof.

[0114] Accordingly, the various illustrative logic blocks, modules, and circuits described in connection with this disclosure may be embodied or implemented as a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination designed to have the functionality described herein. A general-purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be embodied as a combination of computing devices, such as a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other configuration.

[0115] In a firmware and / or software implementation, the techniques may be embodied as instructions stored on a computer-readable medium such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, compact disc (CD), magnetic or optical data storage device, etc. The instructions may be executable by one or more processors and may cause the processors to perform certain aspects of the functions described in this disclosure.

[0116] If embodied in software, the techniques described above may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes any medium that facilitates transfer of a computer program from one place to another, including both computer storage media and communication media. Storage media may be any available medium that can be accessed by a computer. By way of non-limiting example, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage, or other magnetic storage devices, or any other medium that can be used to transport or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium.

[0117] For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted wire, digital subscriber line (DSL), or wireless technologies such as infrared, radio, or microwave, the coaxial cable, fiber optic cable, twisted wire, digital subscriber line, or wireless technologies such as infrared, radio, and microwave are included within the definition of medium. As used herein, "disk" and "disc" include CDs, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where "disks" typically reproduce data magnetically and "discs" typically reproduce data optically using a laser. Combinations of the above should also be included within the scope of computer-readable media.

[0118] A software module may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium may be coupled to the processor such that the processor can read information from, and write information to, the storage medium. Alternatively, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. Alternatively, the processor and the storage medium may reside as discrete components in a user terminal.

[0119] While the embodiments described above utilize aspects of the presently disclosed subject matter on one or more stand-alone computer systems, the present disclosure is not limited thereto and may be implemented in connection with any computing environment, such as a network or distributed computing environment. Furthermore, aspects of the subject matter in the present disclosure may be implemented on multiple processing chips or devices, and storage may be affected similarly across multiple devices. Such devices may include PCs, network servers, and handheld devices.

[0120] Although the present disclosure has been described in connection with some embodiments herein, various modifications and changes that can be understood by those of ordinary skill in the art to which the present disclosure pertains can be made without departing from the scope of the present disclosure, and such modifications and changes should be considered to fall within the scope of the claims appended hereto.

Claims

1. 1. A method for generating a reviews database broadcast cue sheet, performed by at least one processor, comprising: receiving user input including information identifying the product; extracting one or more keywords from review data related to the product; generating a broadcast cue sheet based on the one or more keywords using a language generation model; A method for generating a broadcast cue sheet from a review database, comprising:

2. The step of extracting one or more review keywords from the review data related to the product includes: generating at least one answer data from at least a portion of the review data based on a plurality of predetermined questions using a language model; extracting review keywords related to the product based on the at least one response data; 2. The review database broadcast cue sheet generation method of claim 1, comprising:

3. The step of extracting review keywords related to the product based on the at least one response data includes: converting answer data to the predetermined plurality of questions into embedding vectors; generating at least one group based on the distance between the embedding vectors; extracting a representative keyword from each of the at least one group; 3. The review database broadcast cue sheet generation method of claim 2, comprising:

4. The step of extracting a representative keyword from each of the at least one group includes: determining the representative keyword based on the frequency of keywords included in the at least one group in the review data; 4. The review database broadcast cue sheet generation method of claim 3, comprising:

5. generating a broadcast cue sheet based on the one or more keywords using a language generation model, generating a bullet-type short sentence including the one or more keywords; 2. The review database broadcast cue sheet generation method of claim 1, comprising:

6. 6. The method for generating a broadcast cue sheet for a review database according to claim 5, wherein the bullet-type short sentences include short sentences that combine the one or more keywords with nouns related to the product, or short sentences that combine the one or more keywords with adjectives related to the product.

7. The method for generating a cue sheet for broadcasting a review database according to claim 5 , wherein the bullet-type short sentences include short sentences that combine the one or more keywords with timely short sentences.

8. The step of generating a bullet-type short sentence including the one or more keywords includes:

6. The method for generating a cue sheet for broadcasting a review database according to claim 5, further comprising the step of removing predetermined prohibited words from the bullet-type short sentences.

9. generating a broadcast cue sheet based on the one or more keywords using a language generation model, generating a long sentence based on the bullet-type short sentence; 6. The review database broadcast cue sheet generation method of claim 5, further comprising:

10. The step of generating a long sentence based on the bullet-type short sentence includes:

10. The method of claim 9, further comprising: adding sentence components to the bullet-type short sentences to generate the long sentences so that the meaning of the broadcast cue sheet can be conveyed.

11. The step of generating a long sentence based on the bullet-type short sentence includes:

10. The method of claim 9, further comprising: generating the long sentences that reflect the writing style of a show host based on the bullet-type short sentences using a machine learning model.

12. generating a broadcast cue sheet based on the one or more keywords using a language generation model, receiving prompts related to the construction of the broadcast cue sheet; generating a broadcast cue sheet based on the sentences and the prompts; further comprising 10. The method of claim 9, wherein the prompt includes information related to at least one of an opening block, a body block, or a closing block.

13. outputting at least one of the one or more keywords or the broadcast cue sheet; 10. The review database broadcast cue sheet generation method of claim 1, further comprising:

14. modifying the one or more keywords in response to user input modifying the one or more keywords; updating the broadcast cue sheet based on the modified one or more keywords; 14. The review database broadcast cue sheet generation method of claim 13, further comprising:

15. generating a show host voice from said broadcast cue sheet using a speech synthesis model; 10. The review database broadcast cue sheet generation method of claim 1, further comprising:

16. updating the broadcast cue sheet based on the show host's personal information in response to user input related to the show host's personal information; 10. The review database broadcast cue sheet generation method of claim 1, further comprising:

17. A non-transitory computer-readable recording medium having recorded thereon instructions for executing the method of claim 1 on a computer.

18. 1. A system comprising: a communication module; Memory and at least one processor coupled to the memory and configured to execute at least one computer-readable program contained in the memory; The at least one program receiving user input including information for identifying the product; extracting one or more keywords from review data related to the product; and instructions for generating a broadcast cue sheet based on the one or more keywords using a language generation model.

Images