System

Abstract

To provide a system.SOLUTION: A system includes extraction means for analyzing a user's remark and extracting information regarding a video work that the user has watched in the past, selection means for selecting the content related to the video work on the basis of the extracted information, and providing means for providing the user with the selected content.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including a description and related instruction sentence regarding the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] Traditional video streaming apps have the challenge of making it difficult for users to reminisce about past movies and TV shows or find related content. [Means for solving the problem]

[0005] A system according to one aspect of the present disclosure includes an extraction means for analyzing a user's comments and extracting information about movies or television programs that the user has previously watched, a selection means for selecting content related to the movie or television program based on the information, and a provision means for providing the content to the user.

[0006] In a system according to one aspect of the present disclosure, the selection means selects the content in which at least one of the title, genre, cast, and theme song of the movie or TV program is the same or similar.

[0007] In a system according to one aspect of the present disclosure, the system further includes an identification means for identifying the era in which the movie or television program was screened or broadcast based on the information, and the selection means selects the content that indicates at least one of the background and events of the era.

[0008] In a system according to one aspect of the present disclosure, the system further includes an emotion engine that recognizes the user's emotion, and the selection means customizes the content to be selected based on the user's emotion.

[0009] As a result, a system according to one aspect of the present disclosure allows users to reminisce about past movies and TV shows and easily find related content. [Brief description of the drawings]

[0010] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] 1 shows an example of main functions of a data processing device and smart glasses according to a second embodiment.

[0011] This is a conceptual diagram. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] 4 is a sequence diagram showing a process flow of the data processing system according to the first embodiment. FIG. [Figure 12] 11 is a sequence diagram showing a process flow of the data processing system in application example 1. FIG. [Figure 13] FIG. 11 is a sequence diagram showing the flow of processing of the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 11 is a sequence diagram showing the flow of processing in the data processing system in application example 2 when combined with an emotion engine. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0013] First, the terms used in the following description will be explained.

[0014] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), etc.

[0015] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0016] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0017] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0018] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0019] [First embodiment]

[0020] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0021] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0022] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[0023] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0024] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0025] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (e.g., voice and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs voice according to instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0026] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0027] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0028] As shown in Fig. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific process program 56 executed on the RAM 30.

[0029] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0030] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores a reception output program 60. The reception output program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads out the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0031] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0032] An embodiment for carrying out the present disclosure is embodied as a system including an extraction means, a specification means, a selection means, and a provision means. The extraction means analyzes a user's utterance to extract information related to a movie or television program that the user has previously seen. The specification means specifies an era in which the movie or television program that the user has previously seen was screened or aired based on the information extracted by the extraction means. The selection means selects content related to the movie or television program that the user has previously seen based on the information extracted by the extraction means. In this case, the selection means may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movie or television program that the user has previously seen. The selection means may also select content that indicates at least one of the background and events of the era specified by the specification means. The provision means provides the content selected by the selection means to the user.

[0033] This form allows the user to reminisce about past movies and television programs and easily find related content. As a specific example, assume that the user says, "I used to like the movie 'Memories of Youth'. Taro Yamada's acting was very good. The song 'Hikari' also suited the movie." In this case, the extraction means analyzes the user's statement and extracts information that the title of the movie the user has seen in the past is 'Memories of Youth', the cast is 'Taro Yamada', and the theme song is 'Hikari'. The identification means identifies that 'Memories of Youth' was screened in the 1980s. The selection means selects content related to 'Memories of Youth'. In this case, the selection means may select content summarizing 'Memories of Youth', may select content belonging to the same youth genre as 'Memories of Youth', may select other content in which 'Taro Yamada' appears, or may select the latest content sung by 'Hanako Sato', the singer of 'Hikari'. The selection means may also select content showing the World Expo held in the 1980s. The providing means provides the content selected in this manner to the user, for example, so that the user can not only reminisce about past content, but also enjoy other related content and the latest content.

[0034] The process flow of each embodiment will be described below.

[0035] Step 1: Extracting information

[0036] The user reminisces about the past and makes comments.

[0037] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0038] Step 2: Identify the era

[0039] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0040] Step 3: Select content

[0041] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0042] Step 4: Providing content

[0043] The server uses a providing means to provide the content selected in step 3 to the user.

[0044] Example 1

[0045] Next, a description will be given of Example 1. In the following description, the data processing device 12 is referred to as a "server" and the smart device 14 is referred to as a "terminal."

[0046] In conventional systems, it was difficult for users to efficiently find content related to movies and TV programs they had previously seen, and there was a lack of means to provide related information. In addition, it was difficult to select content that took into account the historical background and events related to past works, making it difficult to increase user satisfaction.

[0047] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0048] In this invention, the server includes an extraction means for analyzing a user's comments to extract information about a video work that the user has previously viewed, an identification means for identifying the era in which the video work was screened or broadcast based on the extracted information, a selection means for selecting content related to the video work based on the extracted information and the identified era, and a provision means for providing the selected content to the user. This makes it possible to efficiently extract information related to video works that the user has previously viewed and provide appropriate content based on that information.

[0049] The "extraction means" is a means for analyzing a user's comments and extracting information about video works that the user has seen in the past.

[0050] The "identification means" is a means for identifying the era in which a video work was screened or broadcast based on the information extracted by the extraction means.

[0051] A "selection means" is a means for selecting content related to a video work based on the extracted information and the identified era.

[0052] The "provision means" is a means for providing the content selected by the selection means to the user.

[0053] "Visual works" is a broad category of content that includes films, television programs, and other visual media.

[0054] "Information" refers to detailed data about a video work, such as the title, cast, genre, theme song, etc.

[0055] "Era" refers to the historical period in which the video work was first released or broadcast.

[0056] As an embodiment of the present invention, there is a system including a server, a terminal, and a user. This system comprises an extracting means, a specifying means, a selecting means, and a providing means.

[0057] First, the server captures the user's speech as voice data. This can be done using a voice input device or a smart speaker. The server then converts the voice data into text using voice recognition software such as the Google® Speech-to-Text API.

[0058] The server analyzes the text data using a natural language processing (NLP) library (e.g., spaCy® or NLTK®) to extract information about the video, including the title, cast, genre, theme song, etc.

[0059] Based on the extracted information, the server uses the OMDb® API or other movie databases to identify the era in which the video was screened or aired.

[0060] The server then searches for and selects content related to the video work based on the extracted information and the identified era. The selection can be made using a movie database or the YouTube® Data API. Here, related content can include other works in the same genre, other works featuring the same actors, or content related to the background or events of the identified era.

[0061] Finally, the server provides the selected content to the user's terminal via email, notification, application interface, etc.

[0062] Examples

[0063] For example, if a user says to the system, "I used to like the movie 'Memories of Youth.' The actor's performance was very good. The song 'Hikari' also suited the movie," the system will process as follows:

[0064] 1. The user's utterance is sent from the voice input device to the server.

[0065] 2. The voice data is converted into text data using voice recognition software (Google Speech-to-Text API).

[0066] 3. Using natural language processing, the information "memories of youth," "actor's name," and "light" is extracted.

[0067] 4. Using the OMDb API or similar, it is determined that "Memories of Youth" was created in the 1980s.

[0068] 5. The server consults a database of film works to select related content, such as other films starring the same actor or documentaries providing background on the 1980s.

[0069] 6. Finally, the server notifies the user's terminal of the selected content and provides it so that the user can view it.

[0070] Examples of prompt statements

[0071] "I used to love the movie Memories of Youth. The actors' performances were amazing and the song "Light" suited the movie. Please tell me some content related to this movie."

[0072] "Recommend new content related to teen movies you've watched in the past."

[0073] In this way, a system is realized that allows users to easily find new related content while reminiscing about past video works.

[0074] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0075] Step 1:

[0076] The server acquires the user's speech and receives voice data via a voice input device or a smart speaker. The user's speech is input, and voice data is generated as output.

[0077] Specifically, the user says, "I used to like the movie 'Memories of Youth'..." This speech is sent to the server.

[0078] Step 2:

[0079] The server converts the acquired voice data into text data. Speech recognition is performed using the Google Speech-to-Text API. The input is the voice data acquired in the previous step, and text data is generated as the output.

[0080] Specifically, the voice recognition software converts statements such as "I used to like the movie 'Memories of Youth'..." into text.

[0081] Step 3:

[0082] The server analyzes the text data using a natural language processing (NLP) library, for example spaCy or NLTK, to extract the required information. The input is the text data, and the output is the analyzed information (movie title, cast, etc.).

[0083] Specifically, keywords such as "memories of youth," "actor names," and "light" are extracted from the text.

[0084] Step 4:

[0085] The server identifies the era in which the video work was screened or broadcast based on the extracted information, or queries the information using the OMDb API or other movie database to identify the era, taking the extracted information as input and producing identified era data as output.

[0086] A specific action is to identify that "Memories of Youth" was screened in the 1980s.

[0087] Step 5:

[0088] The server searches and selects relevant content based on the extracted information and the identified era. It uses a movie database and YouTube Data API to get the relevant content. It has the identified information as input and generates the selected content as output.

[0089] Specific actions to be taken include other films starring the same actors as in "Memories of Youth" and documentaries about 1980s culture.

[0090] Step 6:

[0091] The server provides the selected content to the user's terminal. The server presents the content through email, notifications, application interfaces, etc. The selected content is input, and the content provided to the user terminal is generated as output.

[0092] In concrete terms, the selected content is displayed on the user terminal, making it possible for the user to view it.

[0093] This series of processing steps realizes a system that enables users to efficiently extract information related to video works they have previously seen and easily find new related content.

[0094] (Application example 1)

[0095] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0096] Conventional content recommendation systems have difficulty efficiently providing content related to movies and TV programs that users have seen in the past. They also lack the means to provide new related information and historical background based on users' memories and nostalgia. As a result, users have difficulty finding new content that matches their interests.

[0097] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0098] In this invention, the server includes an extraction means for analyzing a user's comments and extracting information related to viewing content that the user has previously viewed, a selection means for selecting related information related to the viewing content based on the information, a provision means for providing the related information to the user, and a generation AI model for generating new content based on the related information. This makes it possible to efficiently provide new content related to viewing content that the user has previously viewed and related information based on the historical background.

[0099] Definition of Terms

[0100] The "extraction means" is a means for analyzing user comments and extracting information about previously viewed content from the comments.

[0101] The "selecting means" is a means for selecting related information related to the audiovisual content based on the information extracted by the extracting means.

[0102] The "provision means" is a means for providing the related information selected by the selection means to the user.

[0103] A "generative AI model" is an artificial intelligence model that generates new content and prompts based on related information.

[0104] "Viewed content" refers to content such as movies or television programs that the user has viewed in the past.

[0105] "Related information" refers to various information related to the audiovisual content, such as the title, category, performers, musical pieces, and historical background of the audiovisual content.

[0106] The "identification means" is a means for determining the time when the audiovisual content was released or broadcast.

[0107] A "prompt" is a document generated by a generative AI model that contains recommendations and explanations of relevant content.

[0108] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0109] The present invention relates to a system that extracts information related to viewing content that a user has previously viewed, and provides new related content based on that information. Specifically, it uses an AI model that analyzes user comments, identifies related viewing content and historical background based on the extracted information, and generates new content.

[0110] System Configuration

[0111] This system has speech recognition software (e.g., Google Recognition API) to analyze what the user says, and the ability to convert the speech into text. It uses natural language processing technology (e.g., NLP library) to extract information about the content the user has previously viewed from the converted text.

[0112] extraction means

[0113] The server analyzes the user's comments and extracts information about the content the user has previously viewed (e.g., movies and TV shows), such as titles, categories, cast members, and musical works. This information is stored in a database.

[0114] Selection method

[0115] The server selects information related to the viewed content based on the extracted information, including the title, category, performers, musical pieces, or other similar content of the viewed content, allowing the user to easily find new content related to previously viewed content.

[0116] Providing means

[0117] The server provides the selected relevant information to the user, which may be displayed on the user's screen or as a recommendation list.

[0118] Generative AI Models

[0119] A feature of the present invention is a generative AI model that generates new content based on related information. The generative AI model (e.g., ChatGPT (registered trademark)) generates new content and prompts related to content previously viewed by the user.

[0120] Examples

[0121] As a specific example, if a user says, "I used to like the movie 'Memories of Youth,'" the server analyzes this statement and extracts that "Memories of Youth" is the title of the viewing content. Next, other movies related to "Memories of Youth" and documentaries set in the same historical period are selected. This information is then provided to the user. The generative AI model can also generate a prompt such as "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s," and recommend new related content to the user.

[0122] In this way, the present invention can provide new related information based on content that the user has previously viewed, thereby not only evoking the user's nostalgia but also providing a new viewing experience.

[0123] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0124] Program processing flow

[0125] Step 1:

[0126] The user speaks to the terminal. The terminal converts this speech into text data using voice recognition software (e.g., Google Recognition API). The input is the user's voice, and the output is text data. This converted text data is passed on to subsequent processing.

[0127] Step 2:

[0128] The server receives the text data and uses natural language processing techniques (e.g., NLP libraries) to extract information about the viewed content mentioned by the user during speech. This information may include movie or TV show titles, categories, actors, musical works, etc. The input is the text data, and the output is the extracted information about the viewed content. This extracted information is stored in a database.

[0129] Step 3:

[0130] The server searches and selects related information related to the viewing content based on the extracted information in the database, including the title, category, performers, musical pieces, or other similar content of the viewing content. The input is the extracted information, and the output is related information, which is used in the next step.

[0131] Step 4:

[0132] The server uses a generative AI model (e.g. ChatGPT) to generate a new prompt based on the relevant information. This prompt can be a recommendation or description of new content related to the user's interests. The input is the relevant information, and the output is the prompt.

[0133] Step 5:

[0134] The server provides new content to the user along with related information based on the generated prompt. Specifically, the related information and prompt are displayed on the screen of the user's device. The input is the prompt and related information, and the output is the display on the user's screen. Through these displays, the user can view information about new viewing content and data about the historical background.

[0135] Examples

[0136] For example, consider the case where a user says, "I used to like the movie 'Memories of Youth.'" In step 1, the device converts this statement into text data, and in step 2, the server analyzes the text data and extracts the title "Memories of Youth." In step 3, the server selects information about other movies and actors related to "Memories of Youth," and in step 4, the generative AI model generates a prompt saying, "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s." Finally, in step 5, this information is provided to the user, encouraging a new viewing experience.

[0137] Furthermore, an emotion engine that estimates the emotion of the user may be combined. That is, the identification processing unit 290 may estimate the emotion of the user using the emotion identification model 59, and perform identification processing using the emotion of the user.

[0138] An embodiment of the present disclosure may be embodied as a system further including an emotion engine and an emotion analysis means. The emotion engine recognizes the emotion of a user. The emotion analysis means analyzes information such as facial expressions, voice, and text of the user to estimate the user's emotional state.

[0139] This allows the user to reminisce about past movies and TV shows and easily find related content. As a specific example, the emotion engine recognizes the emotion of the user while watching the content. The selection means customizes the content to be selected based on the emotion of the user recognized by the emotion engine. For example, when the user's emotion indicates a "nostalgic" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are more likely to be selected. On the other hand, when the user's emotion indicates a "boring" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are less likely to be selected. This allows the user to enjoy content that matches his or her emotion.

[0140] The process flow will be explained below.

[0141] Step 1: Extracting information

[0142] The user reminisces about the past and makes comments.

[0143] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0144] Step 2: Identify the era

[0145] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0146] Step 3: Select content

[0147] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0148] Step 4: Providing content

[0149] The server uses a providing means to provide the content selected in step 3 to the user.

[0150] Step 5: Recognize emotions

[0151] The server uses an emotion engine to recognize the user's emotion.

[0152] Step 6: Customize your content

[0153] The server uses the selection means to customize the content selection based on the user's emotions recognized in step 5.

[0154] Example 2

[0155] Next, a description will be given of Example 2. In the following description, the data processing device 12 is referred to as a "server" and the smart device 14 is referred to as a "terminal."

[0156] In conventional content provision systems, content is selected without considering the emotional state of the user, so content that satisfies the user is often not provided. In addition, it is difficult to recommend content based on the content the user has previously viewed or the emotional state at the time. For this reason, there is a demand for improving user satisfaction.

[0157] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring facial expression and voice data of the user in real time, an analysis means for analyzing the facial expression and voice data to estimate the emotional state of the user, a selection means for selecting content suitable for the emotional state of the user based on the analysis means, and a provision means for providing the content selected by the selection means to the user. This makes it possible to provide content suited to the emotional state of the user.

[0158] The "means for acquiring the user's facial expression and voice data in real time" refers to a means for capturing the user's facial expression and voice data in real time using an input device such as a camera or a microphone.

[0159] The "analysis means for analyzing the facial expression and voice data to infer the user's emotional state" refers to a means for inferring the user's emotional state based on the acquired facial expression and voice data using an analysis technique such as a machine learning algorithm.

[0160] The "selection means for selecting content appropriate to the user's emotional state based on the analysis means" refers to a means for adjusting content selection criteria in accordance with the estimated emotional state of the user and selecting the most suitable content.

[0161] The "means for providing the content selected by the selection means to the user" refers to a means for distributing the selected content to the user via a network.

[0162] "Content" refers to movies, television programs, music, videos, and other media forms of entertainment or information.

[0163] This system operates by combining multiple pieces of hardware and software in order to provide suitable content based on the emotional state of a user. A specific embodiment of this system will be described below.

[0164] The system is equipped with a means for acquiring the user's facial expression and voice data in real time. The camera and microphone installed on the terminal fulfill this role. As for the hardware, a general webcam and microphone can be used. These data are sent to the server via the MQTT (registered trademark) protocol or HTTP POST request.

[0165] The server is equipped with software for analyzing the received facial and voice data. To analyze the facial data, OpenCV (registered trademark) is used as an image processing library, and Azure (registered trademark) Face API is used as a cloud service. To analyze the voice data, the voice is first converted to text using Google Cloud Speech-to-Text API, and then the tone and pitch of the voice are analyzed. From this, the analysis means estimates the user's emotional state.

[0166] Based on the estimated emotional state, the server uses a selection means to select the most suitable content. The selection means uses deep learning frameworks such as TENSORFLOW® and PyTorch® to identify content that is suitable for the user's emotion. For example, if the user's emotion is estimated to be "nostalgic," content that was popular in the past or content that the user has previously viewed is prioritized. On the other hand, if the emotion is estimated to be "boring," content from a different genre is provided.

[0167] The selected content is delivered to the user's device through content delivery networks such as AWS CloudFront and Azure Media Services, which allows for high-speed, high-quality streaming. Users can watch the content provided on their device and enjoy entertainment that matches their emotional state.

[0168] Examples of prompt statements

[0169] Emotion recognition prompts:

[0170] "Please analyze the emotions of the user currently viewing the video. Please tell us what emotions the user is feeling based on the following data: Facial expression data: {image_data}, Audio data: {audio_data}."

[0171] Content recommendation prompt:

[0172] "Sentiment analysis indicates that the user feels 'nostalgic'. Based on this user's sentiment, please provide relevant movies and TV shows from the content library below."

[0173] In this way, the present invention can improve user satisfaction by dynamically selecting and providing content based on the user's emotional state.

[0174] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0175] Program processing flow

[0176] Step 1:

[0177] The device uses a camera and a microphone to capture the user's facial expression and voice data in real time. As input, it receives image data from the camera and voice data from the microphone. The camera captures the user's face, and the microphone captures the user's speech. As output, it generates facial expression data (in image format) and voice data (in voice format).

[0178] Step 2:

[0179] The terminal transmits the acquired facial expression data and voice data to the server. As input, it uses the facial expression data and voice data generated in step 1. The data is sent to the server via the MQTT protocol or an HTTP POST request. As output, the facial expression data and voice data are transferred to the server.

[0180] Step 3:

[0181] The server analyzes the received data. First, the facial expression data is processed using an image processing library (e.g., OpenCV), and facial feature points are extracted using an analysis method (e.g., Azure Face API). The voice data is converted to text using Google Cloud Speech-to-Text API, and then the tone and pitch of the voice are analyzed. Facial expression data and voice data are received as input, and the user's emotional state (e.g., "nostalgic," "bored," etc.) is estimated as output.

[0182] Step 4:

[0183] The server selects optimal content based on the emotional state estimated by the analysis means. The selection means uses, for example, TensorFlow or PyTorch to identify content that corresponds to the user's emotional state. As input, it receives the estimated result of the emotional state, and if the user feels "nostalgic," it selects content that was popular in the past or content that the user has viewed in the past, and conversely, if the user feels "boring," it selects content of a different genre. As output, a list of selected content is generated.

[0184] Step 5:

[0185] The server delivers the selected content to the user's device. As input, it uses the list of content generated in step 4. It uses a delivery network such as AWS CloudFront or Azure Media Services to deliver the selected content at high speed and high quality. As output, the content is streamed to the user's device.

[0186] Step 6:

[0187] The user watches and enjoys the content provided on the terminal. As input, the user receives the content delivered in step 5. As output, the user can enjoy content that matches his / her emotional state.

[0188] (Application example 2)

[0189] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0190] Conventional content delivery systems provide relevant content based on the user's past viewing history, but do not take into account the user's current emotional state, making it difficult to provide optimal content that meets the user's needs in a timely manner. In addition, there is a lack of functionality to flexibly adjust content appropriate for when the user is feeling emotionally "nostalgic" or "bored." This creates a problem of limiting the user's viewing experience.

[0191] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes an extraction means for analyzing a user's utterances and facial expressions to extract information about movies or TV programs that the user has previously watched, a selection means for selecting content related to the movie or TV program based on the information and the result of the user's emotion analysis, a provision means for providing the content to the user, an adjustment means for customizing the content to be provided based on the user's emotions, and a display means for displaying content that matches the user's emotions. This makes it possible to provide content optimized for the user's emotional state.

[0192] "User comments":

[0193] This refers to information verbally spoken by a user, and is the subject of analysis using natural language processing.

[0194] "Facial expression":

[0195] It refers to the user's facial movements and expressions, and is an important indicator for identifying emotions through analysis.

[0196] "information":

[0197] This refers to all data related to movies and TV shows, including metadata such as title, genre, and cast.

[0198] “Emotional analysis results”:

[0199] This data indicates the user's emotional state and refers to the results of analysis based on facial expressions and statements.

[0200] "Extraction means":

[0201] This refers to technology and systems for obtaining relevant information from a user's statements and facial expressions.

[0202] "Selection Method":

[0203] This refers to the algorithms and processes that select the most appropriate content based on extracted information and sentiment analysis results.

[0204] "Providing means":

[0205] This refers to technology for presenting selected content to users, and includes, for example, streaming services.

[0206] "Adjustment means":

[0207] This refers to the function of dynamically changing the content provided and recommendation algorithms based on the user's emotional state.

[0208] "Display means":

[0209] It means a device or interface for visually displaying content to a user.

[0210] "content":

[0211] This refers to all media that users watch or listen to, including movies, TV programs, videos, and music.

[0212] An embodiment of the present invention will be described.

[0213] The system analyzes the user's comments and facial expressions to recognize the user's emotions, and provides optimal content based on those emotions. Specifically, the system has the following configuration and operation.

[0214] System Configuration

[0215] The system includes the following major hardware and software:

[0216] Hardware:

[0217] Smartphone Camera

[0218] Smartphone Processors

[0219] software:

[0220] OpenCV: Used as an image processing library to capture the user's face with the camera.

[0221] EmotionRecognizer (Emotion Recognition Library): A model for identifying emotions from a user's face, using a deep learning-based model.

[0222] ContentRecommender (content recommendation engine): Equipped with an algorithm for recommending appropriate content based on emotion recognition results.

[0223] Operating procedure

[0224] 1. User sentiment analysis:

[0225] The server analyzes the user's facial image acquired from the smartphone camera using OpenCV and performs face detection.

[0226] The emotion recognition library (EmotionRecognizer) analyzes facial expressions from detected facial images and identifies the user's emotional state.

[0227] 2. Content Extraction:

[0228] Based on the analysis of the user's utterances and facial expressions, the server extracts information about previously viewed content.

[0229] The extracted information includes data such as movie or TV show title, genre, and cast.

[0230] 3. Content Selection:

[0231] The content recommendation engine (ContentRecommender) selects content suitable for the user based on the sentiment analysis results and extracted information.

[0232] If the user feels "nostalgic," popular dramas and movies from the past will be recommended, and if the user feels "boring," exciting new content will be recommended.

[0233] 4. Content provider:

[0234] The selected content is provided to the user's device by the server.

[0235] An application on the device visually presents the selected content to the user.

[0236] Examples

[0237] For example, if a user expresses a "nostalgic" emotion in front of the camera, the system will recommend hit dramas and movies from the past. Conversely, if the user feels "bored," the system will recommend exciting new movies and TV shows.

[0238] Examples of prompt statements

[0239] If users feel nostalgic, develop a video streaming app that recommends hit dramas and movies from the past.

[0240] As described above, the present invention makes it possible to provide content optimized for the emotional state of the user.

[0241] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0242] Processing steps and detailed explanations

[0243] Step 1:

[0244] The server acquires the user's face images in real time from the smartphone's camera. The input is the video stream from the camera. The output is frames of face images. Using OpenCV, image processing is performed to detect the face. Specifically, the video stream acquired from the camera is divided into frames, and a face detection algorithm is applied to identify the user's face.

[0245] Step 2:

[0246] The server passes the face-detected frames to EmotionRecognizer for emotion analysis. The input is the face image with the detected face. The output is the emotional state (e.g. happy, sad, nostalgic, bored). EmotionRecognizer uses a deep learning model to identify the user's emotion from the face image. Specifically, the face image is input to the model, and emotion labels are obtained as the output.

[0247] Step 3:

[0248] The server converts the user's speech into text using speech recognition technology. The input is speech data from a microphone. The output is text data recognized from the speech. Specifically, a speech recognition engine is used to convert the user's speech into text data in real time.

[0249] Step 4:

[0250] The server extracts information related to the user's past viewing history from the text data of the speech recognition results. The input is the recognized text data. The output is information about movies and TV shows that have been viewed in the past. Specifically, it uses natural language processing (NLP) technology to analyze the text data and extract keywords and metadata related to the past viewing history.

[0251] Step 5:

[0252] The server selects appropriate content based on the emotion analysis results and the extracted viewing history information. The input is the emotion analysis results and viewing history information. The output is a list of recommended content. Using the ContentRecommender engine, a recommendation algorithm based on emotion and past viewing history is applied. Specifically, if the emotional state is "nostalgic," hit dramas and movies from the past are selected, and if the emotional state is "boring," exciting new content is selected.

[0253] Step 6:

[0254] The server provides the selected content to the user's smartphone device. The input is a list of recommended content. The output is the content to be displayed on the device. Specifically, the server sends the URL and metadata of the selected content to the device so that the user can view it.

[0255] Step 7:

[0256] The terminal visually displays the provided content to the user. The input is the content information provided by the server. The output is the content displayed on the terminal screen. Specifically, the application UI is updated to allow the user to easily view the recommended content.

[0257] Through the above processing steps, content that is optimized for the user's emotional state is provided.

[0258] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0259] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0260] In the above embodiment, an example was given in which the specific process was performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0261] [Second embodiment]

[0262] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0263] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0264] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[0265] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0266] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0267] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0268] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0269] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0270] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0271] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0272] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0273] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0274] An embodiment for carrying out the present disclosure is embodied as a system including an extraction means, a specification means, a selection means, and a provision means. The extraction means analyzes a user's utterance to extract information related to a movie or television program that the user has previously seen. The specification means specifies an era in which the movie or television program that the user has previously seen was screened or aired based on the information extracted by the extraction means. The selection means selects content related to the movie or television program that the user has previously seen based on the information extracted by the extraction means. In this case, the selection means may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movie or television program that the user has previously seen. The selection means may also select content that indicates at least one of the background and events of the era specified by the specification means. The provision means provides the content selected by the selection means to the user.

[0275] This form allows the user to reminisce about past movies and television programs and easily find related content. As a specific example, assume that the user says, "I used to like the movie 'Memories of Youth'. Taro Yamada's acting was very good. The song 'Hikari' also suited the movie." In this case, the extraction means analyzes the user's statement and extracts information that the title of the movie the user has seen in the past is 'Memories of Youth', the cast is 'Taro Yamada', and the theme song is 'Hikari'. The identification means identifies that 'Memories of Youth' was screened in the 1980s. The selection means selects content related to 'Memories of Youth'. In this case, the selection means may select content summarizing 'Memories of Youth', may select content belonging to the same youth genre as 'Memories of Youth', may select other content in which 'Taro Yamada' appears, or may select the latest content sung by 'Hanako Sato', the singer of 'Hikari'. The selection means may also select content showing the World Expo held in the 1980s. The providing means provides the content selected in this manner to the user, for example, so that the user can not only reminisce about past content, but also enjoy other related content and the latest content.

[0276] The process flow will be explained below.

[0277] Step 1: Extracting information

[0278] The user reminisces about the past and makes comments.

[0279] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0280] Step 2: Identify the era

[0281] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0282] Step 3: Select content

[0283] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0284] Step 4: Providing content

[0285] The server uses a providing means to provide the content selected in step 3 to the user.

[0286] Example 1

[0287] Next, a description will be given of Example 1. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[0288] In conventional systems, it was difficult for users to efficiently find content related to movies and TV programs they had previously seen, and there was a lack of means to provide related information. In addition, it was difficult to select content that took into account the historical background and events related to past works, making it difficult to increase user satisfaction.

[0289] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0290] In this invention, the server includes an extraction means for analyzing a user's comments to extract information about a video work that the user has previously viewed, an identification means for identifying the era in which the video work was screened or broadcast based on the extracted information, a selection means for selecting content related to the video work based on the extracted information and the identified era, and a provision means for providing the selected content to the user. This makes it possible to efficiently extract information related to video works that the user has previously viewed and provide appropriate content based on that information.

[0291] The "extraction means" is a means for analyzing a user's comments and extracting information about video works that the user has seen in the past.

[0292] The "identification means" is a means for identifying the era in which a video work was screened or broadcast based on the information extracted by the extraction means.

[0293] A "selection means" is a means for selecting content related to a video work based on the extracted information and the identified era.

[0294] The "provision means" is a means for providing the content selected by the selection means to the user.

[0295] "Visual works" is a broad category of content that includes films, television programs, and other visual media.

[0296] "Information" refers to detailed data about a video work, such as the title, cast, genre, theme song, etc.

[0297] "Era" refers to the historical period in which the video work was first released or broadcast.

[0298] As an embodiment of the present invention, there is a system including a server, a terminal, and a user. This system comprises an extracting means, a specifying means, a selecting means, and a providing means.

[0299] First, the server captures the user's speech as voice data. This can be done using a voice input device or a smart speaker. Then, the server converts the voice data into text using voice recognition software such as the Google Speech-to-Text API.

[0300] The server parses the text data using a natural language processing (NLP) library (e.g., spaCy or NLTK) to extract information about the video, including the title, cast, genre, theme song, etc.

[0301] Based on the extracted information, the server uses the OMDb API and other movie databases to identify the era in which the footage was screened or aired.

[0302] The server then searches for and selects content related to the video work based on the extracted information and the identified era, such as by using a movie database or the YouTube Data API, where related content may include other works in the same genre, other works featuring the same cast, or content related to the background or events of the identified era.

[0303] Finally, the server provides the selected content to the user's terminal via email, notification, application interface, etc.

[0304] Examples

[0305] For example, if a user says to the system, "I used to like the movie 'Memories of Youth'. The actor's performance was very good. The song 'Hikari' also suited the movie," the system will process as follows:

[0306] 1. The user's utterance is sent from the voice input device to the server.

[0307] 2. The voice data is converted into text data using voice recognition software (Google Speech-to-Text API).

[0308] 3. Using natural language processing, the information "memories of youth," "actor's name," and "light" is extracted.

[0309] 4. Using the OMDb API or similar, it is determined that "Memories of Youth" was created in the 1980s.

[0310] 5. The server consults a database of film works to select related content, such as other films starring the same actor or documentaries providing background on the 1980s.

[0311] 6. Finally, the server notifies the user's terminal of the selected content and provides it so that the user can view it.

[0312] Examples of prompt statements

[0313] "I used to love the movie Memories of Youth. The actors' performances were amazing and the song "Light" suited the movie. Please tell me some content related to this movie."

[0314] "Recommend new content related to teen movies you've watched in the past."

[0315] In this way, a system is realized that allows users to easily find new related content while reminiscing about past video works.

[0316] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0317] Step 1:

[0318] The server acquires the user's speech and receives voice data via a voice input device or a smart speaker. The user's speech is input, and voice data is generated as output.

[0319] Specifically, the user says, "I used to like the movie 'Memories of Youth'..." This speech is sent to the server.

[0320] Step 2:

[0321] The server converts the acquired voice data into text data. Speech recognition is performed using the Google Speech-to-Text API. The input is the voice data acquired in the previous step, and text data is generated as the output.

[0322] Specifically, the voice recognition software converts statements such as "I used to like the movie 'Memories of Youth'..." into text.

[0323] Step 3:

[0324] The server analyzes the text data using a natural language processing (NLP) library, for example spaCy or NLTK, to extract the required information. The input is the text data, and the output is the analyzed information (movie title, cast, etc.).

[0325] Specifically, keywords such as "memories of youth," "actor names," and "light" are extracted from the text.

[0326] Step 4:

[0327] The server identifies the era in which the video work was screened or broadcast based on the extracted information, or queries the information using the OMDb API or other movie database to identify the era, taking the extracted information as input and producing identified era data as output.

[0328] A specific action is to identify that "Memories of Youth" was screened in the 1980s.

[0329] Step 5:

[0330] The server searches and selects relevant content based on the extracted information and the identified era. It uses a movie database and YouTube Data API to get the relevant content. It has the identified information as input and generates the selected content as output.

[0331] Specific actions to be taken include other films starring the same actors as in "Memories of Youth" and documentaries about 1980s culture.

[0332] Step 6:

[0333] The server provides the selected content to the user's terminal. The content is presented through email, notifications, application interfaces, etc. The selected content is taken as input, and the content provided to the user terminal is generated as output.

[0334] In concrete terms, the selected content is displayed on the user terminal, making it possible for the user to view it.

[0335] This series of processing steps realizes a system that enables users to efficiently extract information related to video works they have previously seen and easily find new related content.

[0336] (Application example 1)

[0337] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0338] Conventional content recommendation systems have difficulty efficiently providing content related to movies and TV programs that users have seen in the past. They also lack the means to provide new related information and historical background based on users' memories and nostalgia. As a result, users have difficulty finding new content that matches their interests.

[0339] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0340] In this invention, the server includes an extraction means for analyzing a user's comments and extracting information related to viewing content that the user has previously viewed, a selection means for selecting related information related to the viewing content based on the information, a provision means for providing the related information to the user, and a generation AI model for generating new content based on the related information. This makes it possible to efficiently provide new content related to viewing content that the user has previously viewed and related information based on the historical background.

[0341] Definition of Terms

[0342] The "extraction means" is a means for analyzing user comments and extracting information about previously viewed content from the comments.

[0343] The "selecting means" is a means for selecting related information related to the audiovisual content based on the information extracted by the extracting means.

[0344] The "provision means" is a means for providing the related information selected by the selection means to the user.

[0345] A "generative AI model" is an artificial intelligence model that generates new content and prompts based on related information.

[0346] "Viewed content" refers to content such as movies or television programs that the user has viewed in the past.

[0347] "Related information" refers to various information related to the audiovisual content, such as the title, category, performers, musical pieces, and historical background of the audiovisual content.

[0348] The "identification means" is a means for determining the time when the audiovisual content was released or broadcast.

[0349] A "prompt" is a document generated by a generative AI model that contains recommendations and explanations of relevant content.

[0350] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0351] The present invention relates to a system that extracts information related to viewing content that a user has previously viewed, and provides new related content based on that information. Specifically, it uses an AI model that analyzes user comments, identifies related viewing content and historical background based on the extracted information, and generates new content.

[0352] System Configuration

[0353] This system has speech recognition software (e.g., Google Recognition API) to analyze what the user says, and the ability to convert the speech into text. It uses natural language processing technology (e.g., NLP library) to extract information about the content the user has previously viewed from the converted text.

[0354] extraction means

[0355] The server analyzes the user's comments and extracts information about the content the user has previously viewed (e.g., movies and TV shows), such as titles, categories, cast members, and musical works. This information is stored in a database.

[0356] Selection method

[0357] The server selects information related to the viewed content based on the extracted information, including the title, category, performers, musical pieces, or other similar content of the viewed content, allowing the user to easily find new content related to previously viewed content.

[0358] Providing means

[0359] The server provides the selected relevant information to the user, which may be displayed on the user's screen or as a recommendation list.

[0360] Generative AI Models

[0361] A feature of the present invention is a generative AI model that generates new content based on related information. The generative AI model (e.g., ChatGPT) generates new content and prompts related to content that the user has previously viewed.

[0362] Examples

[0363] As a specific example, if a user says, "I used to like the movie 'Memories of Youth,'" the server analyzes this statement and extracts that "Memories of Youth" is the title of the viewing content. Next, other movies related to "Memories of Youth" and documentaries set in the same historical period are selected. This information is then provided to the user. The generative AI model can also generate a prompt such as "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s," and recommend new related content to the user.

[0364] In this way, the present invention can provide new related information based on content that the user has previously viewed, not only evoking the user's nostalgia but also providing a new viewing experience.

[0365] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0366] Program processing flow

[0367] Step 1:

[0368] The user speaks to the terminal. The terminal converts this speech into text data using voice recognition software (e.g., Google Recognition API). The input is the user's voice, and the output is text data. This converted text data is passed on to subsequent processing.

[0369] Step 2:

[0370] The server receives the text data and uses natural language processing techniques (e.g., NLP libraries) to extract information about the viewed content mentioned by the user during speech. This information may include movie or TV show titles, categories, actors, musical works, etc. The input is the text data, and the output is the extracted information about the viewed content. This extracted information is stored in a database.

[0371] Step 3:

[0372] The server searches and selects related information related to the viewing content based on the extracted information in the database, including the title, category, performers, musical pieces, or other similar content of the viewing content. The input is the extracted information, and the output is related information, which is used in the next step.

[0373] Step 4:

[0374] The server uses a generative AI model (e.g. ChatGPT) to generate a new prompt based on the relevant information. This prompt can be a recommendation or description of new content related to the user's interests. The input is the relevant information, and the output is the prompt.

[0375] Step 5:

[0376] The server provides new content to the user along with related information based on the generated prompt. Specifically, the related information and prompt are displayed on the screen of the user's device. The input is the prompt and related information, and the output is the display on the user's screen. Through these displays, the user can view information about new viewing content and data about the historical background.

[0377] Examples

[0378] For example, consider the case where a user says, "I used to like the movie 'Memories of Youth.'" In step 1, the device converts this statement into text data, and in step 2, the server analyzes the text data and extracts the title "Memories of Youth." In step 3, the server selects information about other movies and actors related to "Memories of Youth," and in step 4, the generative AI model generates a prompt saying, "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s." Finally, in step 5, this information is provided to the user, encouraging a new viewing experience.

[0379] In addition, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0380] An embodiment of the present disclosure may be embodied as a system further including an emotion engine and an emotion analysis means. The emotion engine recognizes the emotion of a user. The emotion analysis means analyzes information such as facial expressions, voice, and text of the user to estimate the user's emotional state.

[0381] This allows the user to reminisce about past movies and TV shows and easily find related content. As a specific example, the emotion engine recognizes the emotion of the user while watching the content. The selection means customizes the content to be selected based on the emotion of the user recognized by the emotion engine. For example, when the user's emotion indicates a "nostalgic" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are more likely to be selected. On the other hand, when the user's emotion indicates a "boring" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are less likely to be selected. This allows the user to enjoy content that matches his or her emotion.

[0382] The process flow will be explained below.

[0383] Step 1: Extracting information

[0384] The user reminisces about the past and makes comments.

[0385] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0386] Step 2: Identify the era

[0387] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0388] Step 3: Select content

[0389] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0390] Step 4: Providing content

[0391] The server uses a providing means to provide the content selected in step 3 to the user.

[0392] Step 5: Recognize emotions

[0393] The server uses an emotion engine to recognize the user's emotion.

[0394] Step 6: Customize your content

[0395] The server uses the selection means to customize the content selection based on the user's emotions recognized in step 5.

[0396] Example 2

[0397] Next, a description will be given of Example 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[0398] In conventional content provision systems, content is selected without considering the emotional state of the user, so content that satisfies the user is often not provided. In addition, it is difficult to recommend content based on the content the user has previously viewed or the emotional state at the time. For this reason, there is a demand for improving user satisfaction.

[0399] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring facial expression and voice data of the user in real time, an analysis means for analyzing the facial expression and voice data to estimate the emotional state of the user, a selection means for selecting content suitable for the emotional state of the user based on the analysis means, and a provision means for providing the content selected by the selection means to the user. This makes it possible to provide content suited to the emotional state of the user.

[0400] The "means for acquiring the user's facial expression and voice data in real time" refers to a means for capturing the user's facial expression and voice data in real time using an input device such as a camera or a microphone.

[0401] The "analysis means for analyzing the facial expression and voice data to infer the user's emotional state" refers to a means for inferring the user's emotional state based on the acquired facial expression and voice data using an analysis technique such as a machine learning algorithm.

[0402] The "selection means for selecting content appropriate to the user's emotional state based on the analysis means" refers to a means for adjusting content selection criteria in accordance with the estimated emotional state of the user and selecting the most suitable content.

[0403] The "means for providing the content selected by the selection means to the user" refers to a means for distributing the selected content to the user via a network.

[0404] "Content" refers to movies, television programs, music, videos, and other media forms of entertainment or information.

[0405] This system operates by combining multiple pieces of hardware and software in order to provide suitable content based on the emotional state of a user. A specific embodiment of this system will be described below.

[0406] The system is equipped with a means to capture the user's facial expression and voice data in real time. The camera and microphone installed on the terminal perform this role. As hardware, a general webcam and microphone can be used. These data are sent to the server via the MQTT protocol or HTTP POST request.

[0407] The server is equipped with software to analyze the received facial and voice data. To analyze the facial data, it uses OpenCV as an image processing library and Azure Face API as a cloud service. To analyze the voice data, it first converts the voice to text using Google Cloud Speech-to-Text API, and then analyzes the tone and pitch of the voice. From this, the analysis means estimates the user's emotional state.

[0408] Based on the estimated emotional state, the server uses a selection mechanism to choose the most suitable content. The selection mechanism uses deep learning frameworks such as TensorFlow and PyTorch to identify content that is suitable for the user's emotion. For example, if the user's emotion is estimated to be "nostalgic," content that was popular in the past or that the user has previously viewed is prioritized. On the other hand, if the emotion is estimated to be "boring," content from a different genre is provided.

[0409] The selected content is delivered to the user's device through content delivery networks such as AWS CloudFront and Azure Media Services, which allows for high-speed, high-quality streaming. Users can watch the content provided on their device and enjoy entertainment that matches their emotional state.

[0410] Examples of prompt statements

[0411] Emotion recognition prompts:

[0412] "Please analyze the emotions of the user currently viewing the video. Please tell us what emotions the user is feeling based on the following data: Facial expression data: {image_data}, Audio data: {audio_data}."

[0413] Content recommendation prompt:

[0414] "Sentiment analysis indicates that the user feels 'nostalgic'. Based on this user's sentiment, please provide relevant movies and TV shows from the content library below."

[0415] In this way, the present invention can improve user satisfaction by dynamically selecting and providing content based on the user's emotional state.

[0416] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0417] Program processing flow

[0418] Step 1:

[0419] The device uses a camera and a microphone to capture the user's facial expression and voice data in real time. As input, it receives image data from the camera and voice data from the microphone. The camera captures the user's face, and the microphone captures the user's speech. As output, it generates facial expression data (in image format) and voice data (in voice format).

[0420] Step 2:

[0421] The terminal transmits the acquired facial expression data and voice data to the server. As input, it uses the facial expression data and voice data generated in step 1. The data is sent to the server via the MQTT protocol or an HTTP POST request. As output, the facial expression data and voice data are transferred to the server.

[0422] Step 3:

[0423] The server analyzes the received data. First, the facial expression data is processed using an image processing library (e.g., OpenCV), and facial feature points are extracted using an analysis method (e.g., Azure Face API). The voice data is converted to text using Google Cloud Speech-to-Text API, and then the tone and pitch of the voice are analyzed. Facial expression data and voice data are received as input, and the user's emotional state (e.g., "nostalgic," "bored," etc.) is estimated as output.

[0424] Step 4:

[0425] The server selects optimal content based on the emotional state estimated by the analysis means. The selection means uses, for example, TensorFlow or PyTorch to identify content that corresponds to the user's emotional state. As input, it receives the estimated result of the emotional state, and if the user feels "nostalgic," it selects content that was popular in the past or content that the user has viewed in the past, and conversely, if the user feels "boring," it selects content of a different genre. As output, a list of selected content is generated.

[0426] Step 5:

[0427] The server delivers the selected content to the user's device. As input, it uses the list of content generated in step 4. It uses a delivery network such as AWS CloudFront or Azure Media Services to deliver the selected content at high speed and high quality. As output, the content is streamed to the user's device.

[0428] Step 6:

[0429] The user watches and enjoys the content provided on the terminal. As input, the user receives the content delivered in step 5. As output, the user can enjoy content that matches his / her emotional state.

[0430] (Application example 2)

[0431] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0432] Conventional content delivery systems provide relevant content based on the user's past viewing history, but do not take into account the user's current emotional state, making it difficult to provide optimal content that meets the user's needs in a timely manner. In addition, there is a lack of functionality to flexibly adjust content appropriate for when the user is feeling emotionally "nostalgic" or "bored." This creates a problem of limiting the user's viewing experience.

[0433] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes an extraction means for analyzing a user's utterances and facial expressions to extract information about movies or TV programs that the user has previously watched, a selection means for selecting content related to the movie or TV program based on the information and the result of the user's emotion analysis, a provision means for providing the content to the user, an adjustment means for customizing the content to be provided based on the user's emotions, and a display means for displaying content that matches the user's emotions. This makes it possible to provide content optimized for the user's emotional state.

[0434] "User comments":

[0435] This refers to information verbally spoken by a user, and is the subject of analysis using natural language processing.

[0436] "Facial expression":

[0437] It refers to the user's facial movements and expressions, and is an important indicator for identifying emotions through analysis.

[0438] "information":

[0439] This refers to all data related to movies and TV shows, including metadata such as title, genre, and cast.

[0440] “Emotional analysis results”:

[0441] This data indicates the user's emotional state and refers to the results of analysis based on facial expressions and statements.

[0442] "Extraction means":

[0443] This refers to technology and systems for obtaining relevant information from a user's statements and facial expressions.

[0444] "Selection Method":

[0445] This refers to the algorithms and processes that select the most appropriate content based on extracted information and sentiment analysis results.

[0446] "Providing means":

[0447] This refers to technology for presenting selected content to users, and includes, for example, streaming services.

[0448] "Adjustment means":

[0449] This refers to the function of dynamically changing the content provided and recommendation algorithms based on the user's emotional state.

[0450] "Display means":

[0451] It means a device or interface for visually displaying content to a user.

[0452] "content":

[0453] This refers to all media that users watch or listen to, including movies, TV programs, videos, and music.

[0454] An embodiment of the present invention will be described.

[0455] The system analyzes the user's comments and facial expressions to recognize the user's emotions, and provides optimal content based on those emotions. Specifically, the system has the following configuration and operation.

[0456] System Configuration

[0457] The system includes the following major hardware and software:

[0458] Hardware:

[0459] Smartphone Camera

[0460] Smartphone Processors

[0461] software:

[0462] OpenCV: Used as an image processing library to capture the user's face with the camera.

[0463] EmotionRecognizer (Emotion Recognition Library): A model for identifying emotions from a user's face, using a deep learning-based model.

[0464] ContentRecommender (content recommendation engine): Equipped with an algorithm for recommending appropriate content based on emotion recognition results.

[0465] Operating procedure

[0466] 1. User sentiment analysis:

[0467] The server analyzes the user's facial image acquired from the smartphone camera using OpenCV and performs face detection.

[0468] The emotion recognition library (EmotionRecognizer) analyzes facial expressions from detected facial images and identifies the user's emotional state.

[0469] 2. Content Extraction:

[0470] Based on the analysis of the user's utterances and facial expressions, the server extracts information about previously viewed content.

[0471] The extracted information includes data such as movie or TV show title, genre, and cast.

[0472] 3. Content Selection:

[0473] The content recommendation engine (ContentRecommender) selects content suitable for the user based on the sentiment analysis results and extracted information.

[0474] If the user feels "nostalgic," popular dramas and movies from the past will be recommended, and if the user feels "boring," exciting new content will be recommended.

[0475] 4. Content provider:

[0476] The selected content is provided to the user's device by the server.

[0477] An application on the device visually presents the selected content to the user.

[0478] Examples

[0479] For example, if a user expresses a "nostalgic" emotion in front of the camera, the system will recommend hit dramas and movies from the past. Conversely, if the user feels "bored," the system will recommend exciting new movies and TV shows.

[0480] Examples of prompt statements

[0481] If users feel nostalgic, develop a video streaming app that recommends hit dramas and movies from the past.

[0482] As described above, the present invention makes it possible to provide content optimized for the emotional state of the user.

[0483] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0484] Processing steps and detailed explanations

[0485] Step 1:

[0486] The server acquires the user's face images in real time from the smartphone's camera. The input is the video stream from the camera. The output is frames of face images. Using OpenCV, image processing is performed to detect the face. Specifically, the video stream acquired from the camera is divided into frames, and a face detection algorithm is applied to identify the user's face.

[0487] Step 2:

[0488] The server passes the face-detected frames to EmotionRecognizer for emotion analysis. The input is the face image with the detected face. The output is the emotional state (e.g. happy, sad, nostalgic, bored). EmotionRecognizer uses a deep learning model to identify the user's emotion from the face image. Specifically, the face image is input to the model, and emotion labels are obtained as the output.

[0489] Step 3:

[0490] The server converts the user's speech into text using speech recognition technology. The input is speech data from a microphone. The output is text data recognized from the speech. Specifically, a speech recognition engine is used to convert the user's speech into text data in real time.

[0491] Step 4:

[0492] The server extracts information related to the user's past viewing history from the text data of the speech recognition results. The input is the recognized text data. The output is information about movies and TV shows that have been viewed in the past. Specifically, it uses natural language processing (NLP) technology to analyze the text data and extract keywords and metadata related to the past viewing history.

[0493] Step 5:

[0494] The server selects appropriate content based on the emotion analysis results and the extracted viewing history information. The input is the emotion analysis results and viewing history information. The output is a list of recommended content. Using the ContentRecommender engine, a recommendation algorithm based on emotion and past viewing history is applied. Specifically, if the emotional state is "nostalgic," hit dramas and movies from the past are selected, and if the emotional state is "boring," exciting new content is selected.

[0495] Step 6:

[0496] The server provides the selected content to the user's smartphone device. The input is a list of recommended content. The output is the content to be displayed on the device. Specifically, the server sends the URL and metadata of the selected content to the device so that the user can view it.

[0497] Step 7:

[0498] The terminal visually displays the provided content to the user. The input is the content information provided by the server. The output is the content displayed on the terminal screen. Specifically, the application UI is updated to allow the user to easily view the recommended content.

[0499] Through the above processing steps, content that is optimized for the user's emotional state is provided.

[0500] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0501] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0502] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0503] [Third embodiment]

[0504] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0505] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0506] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[0507] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0508] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0509] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0510] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0511] Fig. 6 shows an example of main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0512] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0513] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0514] In the headset type terminal 314, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0515] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server", and the headset type terminal 314 will be referred to as the "terminal".

[0516] An embodiment for carrying out the present disclosure is embodied as a system including an extraction means, a specification means, a selection means, and a provision means. The extraction means analyzes a user's utterance to extract information related to a movie or television program that the user has previously seen. The specification means specifies an era in which the movie or television program that the user has previously seen was screened or aired based on the information extracted by the extraction means. The selection means selects content related to the movie or television program that the user has previously seen based on the information extracted by the extraction means. In this case, the selection means may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movie or television program that the user has previously seen. The selection means may also select content that indicates at least one of the background and events of the era specified by the specification means. The provision means provides the content selected by the selection means to the user.

[0517] This form allows the user to reminisce about past movies and television programs and easily find related content. As a specific example, assume that the user says, "I used to like the movie 'Memories of Youth'. Taro Yamada's acting was very good. The song 'Hikari' also suited the movie." In this case, the extraction means analyzes the user's statement and extracts information that the title of the movie the user has seen in the past is 'Memories of Youth', the cast is 'Taro Yamada', and the theme song is 'Hikari'. The identification means identifies that 'Memories of Youth' was screened in the 1980s. The selection means selects content related to 'Memories of Youth'. In this case, the selection means may select content summarizing 'Memories of Youth', may select content belonging to the same youth genre as 'Memories of Youth', may select other content in which 'Taro Yamada' appears, or may select the latest content sung by 'Hanako Sato', the singer of 'Hikari'. The selection means may also select content showing the World Expo held in the 1980s. The providing means provides the content selected in this manner to the user, for example, so that the user can not only reminisce about past content, but also enjoy other related content and the latest content.

[0518] The process flow will be explained below.

[0519] Step 1: Extracting information

[0520] The user reminisces about the past and makes comments.

[0521] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0522] Step 2: Identify the era

[0523] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0524] Step 3: Select content

[0525] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0526] Step 4: Providing content

[0527] The server uses a providing means to provide the content selected in step 3 to the user.

[0528] Example 1

[0529] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal".

[0530] In conventional systems, it was difficult for users to efficiently find content related to movies and TV programs they had previously seen, and there was a lack of means to provide related information. In addition, it was difficult to select content that took into account the historical background and events related to past works, making it difficult to increase user satisfaction.

[0531] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0532] In this invention, the server includes an extraction means for analyzing a user's comments to extract information about a video work that the user has previously viewed, an identification means for identifying the era in which the video work was screened or broadcast based on the extracted information, a selection means for selecting content related to the video work based on the extracted information and the identified era, and a provision means for providing the selected content to the user. This makes it possible to efficiently extract information related to video works that the user has previously viewed and provide appropriate content based on that information.

[0533] The "extraction means" is a means for analyzing a user's comments and extracting information about video works that the user has seen in the past.

[0534] The "identification means" is a means for identifying the era in which a video work was screened or broadcast based on the information extracted by the extraction means.

[0535] A "selection means" is a means for selecting content related to a video work based on the extracted information and the identified era.

[0536] The "provision means" is a means for providing the content selected by the selection means to the user.

[0537] "Visual works" is a broad category of content that includes films, television programs, and other visual media.

[0538] "Information" refers to detailed data about a video work, such as the title, cast, genre, theme song, etc.

[0539] "Era" refers to the historical period in which the video work was first released or broadcast.

[0540] As an embodiment of the present invention, there is a system including a server, a terminal, and a user. This system comprises an extracting means, a specifying means, a selecting means, and a providing means.

[0541] First, the server captures the user's speech as voice data. This can be done using a voice input device or a smart speaker. Then, the server converts the voice data into text using voice recognition software such as the Google Speech-to-Text API.

[0542] The server parses the text data using a natural language processing (NLP) library (e.g., spaCy or NLTK) to extract information about the video, including the title, cast, genre, theme song, etc.

[0543] Based on the extracted information, the server uses the OMDb API and other movie databases to identify the era in which the footage was screened or aired.

[0544] The server then searches for and selects content related to the video work based on the extracted information and the identified era, such as by using a movie database or the YouTube Data API, where related content may include other works in the same genre, other works featuring the same cast, or content related to the background or events of the identified era.

[0545] Finally, the server provides the selected content to the user's terminal via email, notification, application interface, etc.

[0546] Examples

[0547] For example, if a user says to the system, "I used to like the movie 'Memories of Youth'. The actor's performance was very good. The song 'Hikari' also suited the movie," the system will process as follows:

[0548] 1. The user's utterance is sent from the voice input device to the server.

[0549] 2. The voice data is converted into text data using voice recognition software (Google Speech-to-Text API).

[0550] 3. Using natural language processing, the information "memories of youth," "actor's name," and "light" is extracted.

[0551] 4. Using the OMDb API or similar, it is determined that "Memories of Youth" was created in the 1980s.

[0552] 5. The server consults a database of film works to select related content, such as other films starring the same actor or documentaries providing background on the 1980s.

[0553] 6. Finally, the server notifies the user's terminal of the selected content and provides it so that the user can view it.

[0554] Examples of prompt statements

[0555] "I used to love the movie Memories of Youth. The actors' performances were amazing and the song "Light" suited the movie. Please tell me some content related to this movie."

[0556] "Recommend new content related to teen movies you've watched in the past."

[0557] In this way, a system is realized that allows users to easily find new related content while reminiscing about past video works.

[0558] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0559] Step 1:

[0560] The server acquires the user's utterance and receives voice data via a voice input device or a smart speaker. The user's utterance is the input, and voice data is generated as the output.

[0561] Specifically, the user says, "I used to like the movie 'Memories of Youth'..." This speech is sent to the server.

[0562] Step 2:

[0563] The server converts the acquired voice data into text data. Speech recognition is performed using the Google Speech-to-Text API. The input is the voice data acquired in the previous step, and text data is generated as the output.

[0564] Specifically, the voice recognition software converts statements such as "I used to like the movie 'Memories of Youth'..." into text.

[0565] Step 3:

[0566] The server analyzes the text data using a natural language processing (NLP) library, for example spaCy or NLTK, to extract the required information. The input is the text data, and the output is the analyzed information (movie title, cast, etc.).

[0567] Specifically, keywords such as "memories of youth," "actor names," and "light" are extracted from the text.

[0568] Step 4:

[0569] The server identifies the era in which the video work was screened or broadcast based on the extracted information, or queries the information using the OMDb API or other movie database to identify the era, taking the extracted information as input and producing identified era data as output.

[0570] A specific action is to identify that "Memories of Youth" was screened in the 1980s.

[0571] Step 5:

[0572] The server searches and selects relevant content based on the extracted information and the identified era. It uses a movie database and YouTube Data API to get the relevant content. It has the identified information as input and generates the selected content as output.

[0573] Specific actions to be taken include other films starring the same actors as in "Memories of Youth" and documentaries about 1980s culture.

[0574] Step 6:

[0575] The server provides the selected content to the user's terminal. The content is presented through email, notifications, application interfaces, etc. The selected content is taken as input, and the content provided to the user terminal is generated as output.

[0576] In concrete terms, the selected content is displayed on the user terminal, making it possible for the user to view it.

[0577] This series of processing steps realizes a system that enables users to efficiently extract information related to video works they have previously seen and easily find new related content.

[0578] (Application example 1)

[0579] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0580] Conventional content recommendation systems have difficulty efficiently providing content related to movies and TV programs that users have seen in the past. They also lack the means to provide new related information and historical background based on users' memories and nostalgia. As a result, users have difficulty finding new content that matches their interests.

[0581] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0582] In this invention, the server includes an extraction means for analyzing a user's comments and extracting information related to viewing content that the user has previously viewed, a selection means for selecting related information related to the viewing content based on the information, a provision means for providing the related information to the user, and a generation AI model for generating new content based on the related information. This makes it possible to efficiently provide new content related to viewing content that the user has previously viewed and related information based on the historical background.

[0583] Definition of Terms

[0584] The "extraction means" is a means for analyzing user comments and extracting information about previously viewed content from the comments.

[0585] The "selecting means" is a means for selecting related information related to the audiovisual content based on the information extracted by the extracting means.

[0586] The "provision means" is a means for providing the related information selected by the selection means to the user.

[0587] A "generative AI model" is an artificial intelligence model that generates new content and prompts based on related information.

[0588] "Viewed content" refers to content such as movies or television programs that the user has viewed in the past.

[0589] "Related information" refers to various information related to the audiovisual content, such as the title, category, performers, musical pieces, and historical background of the audiovisual content.

[0590] The "identification means" is a means for determining the time when the audiovisual content was released or broadcast.

[0591] A "prompt" is a document generated by a generative AI model that contains recommendations and explanations of relevant content.

[0592] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0593] The present invention relates to a system that extracts information related to viewing content that a user has previously viewed, and provides new related content based on that information. Specifically, it uses an AI model that analyzes user comments, identifies related viewing content and historical background based on the extracted information, and generates new content.

[0594] System Configuration

[0595] This system has speech recognition software (e.g., Google Recognition API) to analyze what the user says, and the ability to convert the speech into text. It uses natural language processing technology (e.g., NLP library) to extract information about the content the user has previously viewed from the converted text.

[0596] extraction means

[0597] The server analyzes the user's comments and extracts information about the content the user has previously viewed (e.g., movies and TV shows), such as titles, categories, cast members, and musical works. This information is stored in a database.

[0598] Selection method

[0599] The server selects information related to the viewed content based on the extracted information, including the title, category, performers, musical pieces, or other similar content of the viewed content, allowing the user to easily find new content related to previously viewed content.

[0600] Providing means

[0601] The server provides the selected relevant information to the user, which may be displayed on the user's screen or as a recommendation list.

[0602] Generative AI Models

[0603] A feature of the present invention is a generative AI model that generates new content based on related information. The generative AI model (e.g., ChatGPT) generates new content and prompts related to content that the user has previously viewed.

[0604] Examples

[0605] As a specific example, if a user says, "I used to like the movie 'Memories of Youth,'" the server analyzes this statement and extracts that "Memories of Youth" is the title of the viewing content. Next, other movies related to "Memories of Youth" and documentaries set in the same historical period are selected. This information is then provided to the user. The generative AI model can also generate a prompt such as "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s," and recommend new related content to the user.

[0606] In this way, the present invention can provide new related information based on content that the user has previously viewed, thereby not only evoking the user's nostalgia but also providing a new viewing experience.

[0607] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0608] Program processing flow

[0609] Step 1:

[0610] The user speaks to the terminal. The terminal converts this speech into text data using voice recognition software (e.g., Google Recognition API). The input is the user's voice, and the output is text data. This converted text data is passed on to subsequent processing.

[0611] Step 2:

[0612] The server receives the text data and uses natural language processing techniques (e.g., NLP libraries) to extract information about the viewed content mentioned by the user during speech. This information may include movie or TV show titles, categories, actors, musical works, etc. The input is the text data, and the output is the extracted information about the viewed content. This extracted information is stored in a database.

[0613] Step 3:

[0614] The server searches and selects related information related to the viewing content based on the extracted information in the database, including the title, category, performers, musical pieces, or other similar content of the viewing content. The input is the extracted information, and the output is related information, which is used in the next step.

[0615] Step 4:

[0616] The server uses a generative AI model (e.g. ChatGPT) to generate a new prompt based on the relevant information. This prompt can be a recommendation or description of new content related to the user's interests. The input is the relevant information, and the output is the prompt.

[0617] Step 5:

[0618] The server provides new content to the user along with related information based on the generated prompt. Specifically, the related information and prompt are displayed on the screen of the user's device. The input is the prompt and related information, and the output is the display on the user's screen. Through these displays, the user can view information about new viewing content and data about the historical background.

[0619] Examples

[0620] For example, consider the case where a user says, "I used to like the movie 'Memories of Youth.'" In step 1, the device converts this statement into text data, and in step 2, the server analyzes the text data and extracts the title "Memories of Youth." In step 3, the server selects information about other movies and actors related to "Memories of Youth," and in step 4, the generative AI model generates a prompt saying, "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s." Finally, in step 5, this information is provided to the user, encouraging a new viewing experience.

[0621] In addition, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0622] An embodiment of the present disclosure may be embodied as a system further including an emotion engine and an emotion analysis means. The emotion engine recognizes the emotion of a user. The emotion analysis means analyzes information such as facial expressions, voice, and text of the user to estimate the user's emotional state.

[0623] This allows the user to reminisce about past movies and TV shows and easily find related content. As a specific example, the emotion engine recognizes the emotion of the user while watching the content. The selection means customizes the content to be selected based on the emotion of the user recognized by the emotion engine. For example, when the user's emotion indicates a "nostalgic" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are more likely to be selected. On the other hand, when the user's emotion indicates a "boring" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are less likely to be selected. This allows the user to enjoy content that matches his or her emotion.

[0624] The process flow will be explained below.

[0625] Step 1: Extracting information

[0626] The user reminisces about the past and makes comments.

[0627] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0628] Step 2: Identify the era

[0629] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0630] Step 3: Select content

[0631] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0632] Step 4: Providing content

[0633] The server uses a providing means to provide the content selected in step 3 to the user.

[0634] Step 5: Recognize emotions

[0635] The server uses an emotion engine to recognize the user's emotion.

[0636] Step 6: Customize your content

[0637] The server uses the selection means to customize the content selection based on the user's emotions recognized in step 5.

[0638] Example 2

[0639] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal".

[0640] In conventional content provision systems, content is selected without considering the emotional state of the user, so content that satisfies the user is often not provided. In addition, it is difficult to recommend content based on the content the user has previously viewed or the emotional state at the time. For this reason, there is a demand for improving user satisfaction.

[0641] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring facial expression and voice data of the user in real time, an analysis means for analyzing the facial expression and voice data to estimate the emotional state of the user, a selection means for selecting content suitable for the emotional state of the user based on the analysis means, and a provision means for providing the content selected by the selection means to the user. This makes it possible to provide content suited to the emotional state of the user.

[0642] The "means for acquiring the user's facial expression and voice data in real time" refers to a means for capturing the user's facial expression and voice data in real time using an input device such as a camera or a microphone.

[0643] The "analysis means for analyzing the facial expression and voice data to infer the user's emotional state" refers to a means for inferring the user's emotional state based on the acquired facial expression and voice data using an analysis technique such as a machine learning algorithm.

[0644] The "selection means for selecting content appropriate to the user's emotional state based on the analysis means" refers to a means for adjusting content selection criteria in accordance with the estimated emotional state of the user and selecting the most suitable content.

[0645] The "means for providing the content selected by the selection means to the user" refers to a means for distributing the selected content to the user via a network.

[0646] "Content" refers to movies, television programs, music, videos, and other media forms of entertainment or information.

[0647] This system operates by combining multiple pieces of hardware and software in order to provide suitable content based on the emotional state of a user. A specific embodiment of this system will be described below.

[0648] The system is equipped with a means to capture the user's facial expression and voice data in real time. The camera and microphone installed on the terminal perform this role. As hardware, a general webcam and microphone can be used. These data are sent to the server via the MQTT protocol or HTTP POST request.

[0649] The server is equipped with software to analyze the received facial and voice data. To analyze the facial data, it uses OpenCV as an image processing library and Azure Face API as a cloud service. To analyze the voice data, it first converts the voice to text using Google Cloud Speech-to-Text API, and then analyzes the tone and pitch of the voice. From this, the analysis means estimates the user's emotional state.

[0650] Based on the estimated emotional state, the server uses a selection mechanism to choose the most suitable content. The selection mechanism uses deep learning frameworks such as TensorFlow and PyTorch to identify content that is suitable for the user's emotion. For example, if the user's emotion is estimated to be "nostalgic," content that was popular in the past or that the user has previously viewed is prioritized. On the other hand, if the emotion is estimated to be "boring," content from a different genre is provided.

[0651] The selected content is delivered to the user's device through content delivery networks such as AWS CloudFront and Azure Media Services, which allows for high-speed, high-quality streaming. Users can watch the content provided on their device and enjoy entertainment that matches their emotional state.

[0652] Examples of prompt statements

[0653] Emotion recognition prompts:

[0654] "Please analyze the emotions of the user currently viewing the video. Please tell us what emotions the user is feeling based on the following data: Facial expression data: {image_data}, Audio data: {audio_data}."

[0655] Content recommendation prompt:

[0656] "Sentiment analysis indicates that the user feels 'nostalgic'. Based on this user's sentiment, please provide relevant movies and TV shows from the content library below."

[0657] In this way, the present invention can improve user satisfaction by dynamically selecting and providing content based on the user's emotional state.

[0658] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0659] Program processing flow

[0660] Step 1:

[0661] The device uses a camera and a microphone to capture the user's facial expression and voice data in real time. As input, it receives image data from the camera and voice data from the microphone. The camera captures the user's face, and the microphone captures the user's speech. As output, it generates facial expression data (in image format) and voice data (in voice format).

[0662] Step 2:

[0663] The terminal transmits the acquired facial expression data and voice data to the server. As input, it uses the facial expression data and voice data generated in step 1. The data is sent to the server via the MQTT protocol or an HTTP POST request. As output, the facial expression data and voice data are transferred to the server.

[0664] Step 3:

[0665] The server analyzes the received data. First, the facial expression data is processed using an image processing library (e.g., OpenCV), and facial feature points are extracted using an analysis method (e.g., Azure Face API). The voice data is converted to text using Google Cloud Speech-to-Text API, and then the tone and pitch of the voice are analyzed. Facial expression data and voice data are received as input, and the user's emotional state (e.g., "nostalgic," "bored," etc.) is estimated as output.

[0666] Step 4:

[0667] The server selects optimal content based on the emotional state estimated by the analysis means. The selection means uses, for example, TensorFlow or PyTorch to identify content that corresponds to the user's emotional state. As input, it receives the estimated result of the emotional state, and if the user feels "nostalgic," it selects content that was popular in the past or content that the user has viewed in the past, and conversely, if the user feels "boring," it selects content of a different genre. As output, a list of selected content is generated.

[0668] Step 5:

[0669] The server delivers the selected content to the user's device. As input, it uses the list of content generated in step 4. It uses a delivery network such as AWS CloudFront or Azure Media Services to deliver the selected content at high speed and high quality. As output, the content is streamed to the user's device.

[0670] Step 6:

[0671] The user watches and enjoys the content provided on the terminal. As input, the user receives the content delivered in step 5. As output, the user can enjoy content that matches his / her emotional state.

[0672] (Application example 2)

[0673] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server", and the headset type terminal 314 will be referred to as a "terminal".

[0674] Conventional content delivery systems provide relevant content based on the user's past viewing history, but do not take into account the user's current emotional state, making it difficult to provide optimal content that meets the user's needs in a timely manner. In addition, there is a lack of functionality to flexibly adjust content appropriate for when the user is feeling emotionally "nostalgic" or "bored." This creates a problem of limiting the user's viewing experience.

[0675] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes an extraction means for analyzing a user's utterances and facial expressions to extract information about movies or TV programs that the user has previously watched, a selection means for selecting content related to the movie or TV program based on the information and the result of the user's emotion analysis, a provision means for providing the content to the user, an adjustment means for customizing the content to be provided based on the user's emotions, and a display means for displaying content that matches the user's emotions. This makes it possible to provide content optimized for the user's emotional state.

[0676] "User comments":

[0677] This refers to information verbally spoken by a user, and is the subject of analysis using natural language processing.

[0678] "Facial expression":

[0679] It refers to the user's facial movements and expressions, and is an important indicator for identifying emotions through analysis.

[0680] "information":

[0681] This refers to all data related to movies and TV shows, including metadata such as title, genre, and cast.

[0682] “Emotional analysis results”:

[0683] This data indicates the user's emotional state and refers to the results of analysis based on facial expressions and statements.

[0684] "Extraction means":

[0685] This refers to technology and systems for obtaining relevant information from a user's statements and facial expressions.

[0686] "Selection Method":

[0687] This refers to the algorithms and processes that select the most appropriate content based on extracted information and sentiment analysis results.

[0688] "Providing means":

[0689] This refers to technology for presenting selected content to users, and includes, for example, streaming services.

[0690] "Adjustment means":

[0691] This refers to the function of dynamically changing the content provided and recommendation algorithms based on the user's emotional state.

[0692] "Display means":

[0693] It means a device or interface for visually displaying content to a user.

[0694] "content":

[0695] This refers to all media that users watch or listen to, including movies, TV programs, videos, and music.

[0696] An embodiment of the present invention will be described.

[0697] The system analyzes the user's comments and facial expressions to recognize the user's emotions, and provides optimal content based on those emotions. Specifically, the system has the following configuration and operation.

[0698] System Configuration

[0699] The system includes the following major hardware and software:

[0700] Hardware:

[0701] Smartphone Camera

[0702] Smartphone Processors

[0703] software:

[0704] OpenCV: Used as an image processing library to capture the user's face with the camera.

[0705] EmotionRecognizer (Emotion Recognition Library): A model for identifying emotions from a user's face, using a deep learning-based model.

[0706] ContentRecommender (content recommendation engine): Equipped with an algorithm for recommending appropriate content based on emotion recognition results.

[0707] Operating procedure

[0708] 1. User sentiment analysis:

[0709] The server analyzes the user's facial image acquired from the smartphone camera using OpenCV and performs face detection.

[0710] The emotion recognition library (EmotionRecognizer) analyzes facial expressions from detected facial images and identifies the user's emotional state.

[0711] 2. Content Extraction:

[0712] Based on the analysis of the user's utterances and facial expressions, the server extracts information about previously viewed content.

[0713] The extracted information includes data such as movie or TV show title, genre, and cast.

[0714] 3. Content Selection:

[0715] The content recommendation engine (ContentRecommender) selects content suitable for the user based on the sentiment analysis results and extracted information.

[0716] If the user feels "nostalgic," popular dramas and movies from the past will be recommended, and if the user feels "boring," exciting new content will be recommended.

[0717] 4. Content provider:

[0718] The selected content is provided to the user's device by the server.

[0719] An application on the device visually presents the selected content to the user.

[0720] Examples

[0721] For example, if a user expresses a "nostalgic" emotion in front of the camera, the system will recommend hit dramas and movies from the past. Conversely, if the user feels "bored," the system will recommend exciting new movies and TV shows.

[0722] Examples of prompt statements

[0723] If users feel nostalgic, develop a video streaming app that recommends hit dramas and movies from the past.

[0724] As described above, the present invention makes it possible to provide content optimized for the emotional state of the user.

[0725] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0726] Processing steps and detailed explanations

[0727] Step 1:

[0728] The server acquires the user's face images in real time from the smartphone's camera. The input is the video stream from the camera. The output is frames of face images. Using OpenCV, image processing is performed to detect the face. Specifically, the video stream acquired from the camera is divided into frames, and a face detection algorithm is applied to identify the user's face.

[0729] Step 2:

[0730] The server passes the face-detected frames to EmotionRecognizer for emotion analysis. The input is the face image with the detected face. The output is the emotional state (e.g. happy, sad, nostalgic, bored). EmotionRecognizer uses a deep learning model to identify the user's emotion from the face image. Specifically, the face image is input to the model, and emotion labels are obtained as the output.

[0731] Step 3:

[0732] The server converts the user's speech into text using speech recognition technology. The input is speech data from a microphone. The output is text data recognized from the speech. Specifically, a speech recognition engine is used to convert the user's speech into text data in real time.

[0733] Step 4:

[0734] The server extracts information related to the user's past viewing history from the text data of the speech recognition results. The input is the recognized text data. The output is information about movies and TV shows that have been viewed in the past. Specifically, it uses natural language processing (NLP) technology to analyze the text data and extract keywords and metadata related to the past viewing history.

[0735] Step 5:

[0736] The server selects appropriate content based on the emotion analysis results and the extracted viewing history information. The input is the emotion analysis results and viewing history information. The output is a list of recommended content. Using the ContentRecommender engine, a recommendation algorithm based on emotion and past viewing history is applied. Specifically, if the emotional state is "nostalgic," hit dramas and movies from the past are selected, and if the emotional state is "boring," exciting new content is selected.

[0737] Step 6:

[0738] The server provides the selected content to the user's smartphone device. The input is a list of recommended content. The output is the content to be displayed on the device. Specifically, the server sends the URL and metadata of the selected content to the device so that the user can view it.

[0739] Step 7:

[0740] The terminal visually displays the provided content to the user. The input is the content information provided by the server. The output is the content displayed on the terminal screen. Specifically, the application UI is updated to allow the user to easily view the recommended content.

[0741] Through the above processing steps, content that is optimized for the user's emotional state is provided.

[0742] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0743] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0744] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0745] [Fourth embodiment]

[0746] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0747] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0748] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[0749] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. In addition, the microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0750] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0751] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0752] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0753] The control target 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, legs, etc. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, legs, etc. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0754] Fig. 8 shows an example of main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0755] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0756] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0757] In the robot 414, the reception and output process is performed by the processor 46. A reception and output program 60 is stored in the storage 50. The processor 46 reads the reception and output program 60 from the storage 50, and executes the read reception and output program 60 on the RAM 48. The reception and output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception and output program 60 executed on the RAM 48.

[0758] Next, a description will be given of the specific processing by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[0759] An embodiment for carrying out the present disclosure is embodied as a system including an extraction means, a specification means, a selection means, and a provision means. The extraction means analyzes a user's utterance to extract information related to a movie or television program that the user has previously seen. The specification means specifies an era in which the movie or television program that the user has previously seen was screened or aired based on the information extracted by the extraction means. The selection means selects content related to the movie or television program that the user has previously seen based on the information extracted by the extraction means. In this case, the selection means may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movie or television program that the user has previously seen. The selection means may also select content that indicates at least one of the background and events of the era specified by the specification means. The provision means provides the content selected by the selection means to the user.

[0760] This form allows the user to reminisce about past movies and television programs and easily find related content. As a specific example, assume that the user says, "I used to like the movie 'Memories of Youth'. Taro Yamada's acting was very good. The song 'Hikari' also suited the movie." In this case, the extraction means analyzes the user's statement and extracts information that the title of the movie the user has seen in the past is 'Memories of Youth', the cast is 'Taro Yamada', and the theme song is 'Hikari'. The identification means identifies that 'Memories of Youth' was screened in the 1980s. The selection means selects content related to 'Memories of Youth'. In this case, the selection means may select content summarizing 'Memories of Youth', may select content belonging to the same youth genre as 'Memories of Youth', may select other content in which 'Taro Yamada' appears, or may select the latest content sung by 'Hanako Sato', the singer of 'Hikari'. The selection means may also select content showing the World Expo held in the 1980s. The providing means provides the content selected in this manner to the user, for example, so that the user can not only reminisce about past content, but also enjoy other related content and the latest content.

[0761] The process flow will be explained below.

[0762] Step 1: Extracting information

[0763] The user reminisces about the past and makes comments.

[0764] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0765] Step 2: Identify the era

[0766] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0767] Step 3: Select content

[0768] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0769] Step 4: Providing content

[0770] The server uses a providing means to provide the content selected in step 3 to the user.

[0771] Example 1

[0772] Next, a description will be given of Example 1. In the following description, the data processing device 12 is referred to as a "server" and the robot 414 is referred to as a "terminal."

[0773] In conventional systems, it was difficult for users to efficiently find content related to movies and TV programs they had previously seen, and there was a lack of means to provide related information. In addition, it was difficult to select content that took into account the historical background and events related to past works, making it difficult to increase user satisfaction.

[0774] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0775] In this invention, the server includes an extraction means for analyzing a user's comments to extract information about a video work that the user has previously viewed, an identification means for identifying the era in which the video work was screened or broadcast based on the extracted information, a selection means for selecting content related to the video work based on the extracted information and the identified era, and a provision means for providing the selected content to the user. This makes it possible to efficiently extract information related to video works that the user has previously viewed and provide appropriate content based on that information.

[0776] The "extraction means" is a means for analyzing a user's comments and extracting information about video works that the user has seen in the past.

[0777] The "identification means" is a means for identifying the era in which a video work was screened or broadcast based on the information extracted by the extraction means.

[0778] A "selection means" is a means for selecting content related to a video work based on the extracted information and the identified era.

[0779] The "provision means" is a means for providing the content selected by the selection means to the user.

[0780] "Visual works" is a broad category of content that includes films, television programs, and other visual media.

[0781] "Information" refers to detailed data about a video work, such as the title, cast, genre, theme song, etc.

[0782] "Era" refers to the historical period in which the video work was first released or broadcast.

[0783] As an embodiment of the present invention, there is a system including a server, a terminal, and a user. This system comprises an extracting means, a specifying means, a selecting means, and a providing means.

[0784] First, the server captures the user's speech as voice data. This can be done using a voice input device or a smart speaker. Then, the server converts the voice data into text using voice recognition software such as the Google Speech-to-Text API.

[0785] The server analyzes the text data using a natural language processing (NLP) library (e.g., spaCy or NLTK) to extract information about the video, including the title, cast, genre, theme song, etc.

[0786] Based on the extracted information, the server uses the OMDb API and other movie databases to identify the era in which the footage was screened or aired.

[0787] The server then searches for and selects content related to the video work based on the extracted information and the identified era, such as by using a movie database or the YouTube Data API, where related content may include other works in the same genre, other works featuring the same cast, or content related to the background or events of the identified era.

[0788] Finally, the server provides the selected content to the user's terminal via email, notification, application interface, etc.

[0789] Examples

[0790] For example, if a user says to the system, "I used to like the movie 'Memories of Youth.' The actor's performance was very good. The song 'Hikari' also suited the movie," the system will process as follows:

[0791] 1. The user's utterance is sent from the voice input device to the server.

[0792] 2. The voice data is converted into text data using voice recognition software (Google Speech-to-Text API).

[0793] 3. Using natural language processing, the information "memories of youth," "actor's name," and "light" is extracted.

[0794] 4. Using the OMDb API or similar, it is determined that "Memories of Youth" was created in the 1980s.

[0795] 5. The server consults a database of film works to select related content, such as other films starring the same actor or documentaries providing background on the 1980s.

[0796] 6. Finally, the server notifies the user's terminal of the selected content and provides it so that the user can view it.

[0797] Examples of prompt statements

[0798] "I used to love the movie Memories of Youth. The actors' performances were amazing and the song "Light" suited the movie. Please tell me some content related to this movie."

[0799] "Recommend new content related to teen movies you've watched in the past."

[0800] In this way, a system is realized that allows users to easily find new related content while reminiscing about past video works.

[0801] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0802] Step 1:

[0803] The server acquires the user's speech and receives voice data via a voice input device or a smart speaker. The user's speech is input, and voice data is generated as output.

[0804] Specifically, the user says, "I used to like the movie 'Memories of Youth'..." This speech is sent to the server.

[0805] Step 2:

[0806] The server converts the acquired voice data into text data. Speech recognition is performed using the Google Speech-to-Text API. The input is the voice data acquired in the previous step, and text data is generated as the output.

[0807] Specifically, the voice recognition software converts statements such as "I used to like the movie 'Memories of Youth'..." into text.

[0808] Step 3:

[0809] The server analyzes the text data using a natural language processing (NLP) library, for example spaCy or NLTK, to extract the required information. The input is the text data, and the output is the analyzed information (movie title, cast, etc.).

[0810] Specifically, keywords such as "memories of youth," "actor names," and "light" are extracted from the text.

[0811] Step 4:

[0812] The server identifies the era in which the video work was screened or broadcast based on the extracted information, or queries the information using the OMDb API or other movie database to identify the era, taking the extracted information as input and producing identified era data as output.

[0813] A specific action is to identify that "Memories of Youth" was screened in the 1980s.

[0814] Step 5:

[0815] The server searches and selects relevant content based on the extracted information and the identified era. It uses a movie database and YouTube Data API to get the relevant content. It has the identified information as input and generates the selected content as output.

[0816] Specific actions to be taken include other films starring the same actors as in "Memories of Youth" and documentaries about 1980s culture.

[0817] Step 6:

[0818] The server provides the selected content to the user's terminal. The server presents the content through email, notifications, application interfaces, etc. The selected content is input, and the content provided to the user terminal is generated as output.

[0819] In concrete terms, the selected content is displayed on the user terminal, making it possible for the user to view it.

[0820] This series of processing steps realizes a system that enables users to efficiently extract information related to video works they have previously seen and easily find new related content.

[0821] (Application example 1)

[0822] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0823] Conventional content recommendation systems have difficulty efficiently providing content related to movies and TV programs that users have seen in the past. They also lack the means to provide new related information and historical background based on users' memories and nostalgia. As a result, users have difficulty finding new content that matches their interests.

[0824] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0825] In this invention, the server includes an extraction means for analyzing a user's comments and extracting information related to viewing content that the user has previously viewed, a selection means for selecting related information related to the viewing content based on the information, a provision means for providing the related information to the user, and a generation AI model for generating new content based on the related information. This makes it possible to efficiently provide new content related to viewing content that the user has previously viewed and related information based on the historical background.

[0826] Definition of Terms

[0827] The "extraction means" is a means for analyzing user comments and extracting information about previously viewed content from the comments.

[0828] The "selecting means" is a means for selecting related information related to the audiovisual content based on the information extracted by the extracting means.

[0829] The "provision means" is a means for providing the related information selected by the selection means to the user.

[0830] A "generative AI model" is an artificial intelligence model that generates new content and prompts based on related information.

[0831] "Viewed content" refers to content such as movies or television programs that the user has viewed in the past.

[0832] "Related information" refers to various information related to the audiovisual content, such as the title, category, performers, musical pieces, and historical background of the audiovisual content.

[0833] The "identification means" is a means for determining the time when the audiovisual content was released or broadcast.

[0834] A "prompt" is a document generated by a generative AI model that contains recommendations and explanations of relevant content.

[0835] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0836] The present invention relates to a system that extracts information related to viewing content that a user has previously viewed, and provides new related content based on that information. Specifically, it uses an AI model that analyzes user comments, identifies related viewing content and historical background based on the extracted information, and generates new content.

[0837] System Configuration

[0838] This system has speech recognition software (e.g., Google Recognition API) to analyze what the user says, and the ability to convert the speech into text. It uses natural language processing technology (e.g., NLP library) to extract information about the content the user has previously viewed from the converted text.

[0839] extraction means

[0840] The server analyzes the user's comments and extracts information about the content the user has previously viewed (e.g., movies and TV shows), such as titles, categories, cast members, and musical works. This information is stored in a database.

[0841] Selection method

[0842] The server selects information related to the viewed content based on the extracted information, including the title, category, performers, musical pieces, or other similar content of the viewed content, allowing the user to easily find new content related to previously viewed content.

[0843] Providing means

[0844] The server provides the selected relevant information to the user, which may be displayed on the user's screen or as a recommendation list.

[0845] Generative AI Models

[0846] A feature of the present invention is a generative AI model that generates new content based on related information. The generative AI model (e.g., ChatGPT) generates new content and prompts related to content that the user has previously viewed.

[0847] Examples

[0848] As a specific example, if a user says, "I used to like the movie 'Memories of Youth,'" the server analyzes this statement and extracts that "Memories of Youth" is the title of the viewing content. Next, other movies related to "Memories of Youth" and documentaries set in the same historical period are selected. This information is then provided to the user. The generative AI model can also generate a prompt such as "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s," and recommend new related content to the user.

[0849] In this way, the present invention can provide new related information based on content that the user has previously viewed, thereby not only evoking the user's nostalgia but also providing a new viewing experience.

[0850] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0851] Program processing flow

[0852] Step 1:

[0853] The user speaks to the terminal. The terminal converts this speech into text data using voice recognition software (e.g., Google Recognition API). The input is the user's voice, and the output is text data. This converted text data is passed on to subsequent processing.

[0854] Step 2:

[0855] The server receives the text data and uses natural language processing techniques (e.g., NLP libraries) to extract information about the viewed content mentioned by the user during speech. This information may include movie or TV show titles, categories, actors, musical works, etc. The input is the text data, and the output is the extracted information about the viewed content. This extracted information is stored in a database.

[0856] Step 3:

[0857] The server searches and selects related information related to the viewing content based on the extracted information in the database, including the title, category, performers, musical pieces, or other similar content of the viewing content. The input is the extracted information, and the output is related information, which is used in the next step.

[0858] Step 4:

[0859] The server uses a generative AI model (e.g. ChatGPT) to generate a new prompt based on the relevant information. This prompt can be a recommendation or description of new content related to the user's interests. The input is the relevant information, and the output is the prompt.

[0860] Step 5:

[0861] The server provides new content to the user along with related information based on the generated prompt. Specifically, the related information and prompt are displayed on the screen of the user's device. The input is the prompt and related information, and the output is the display on the user's screen. Through these displays, the user can view information about new viewing content and data about the historical background.

[0862] Examples

[0863] For example, consider the case where a user says, "I used to like the movie 'Memories of Youth.'" In step 1, the device converts this statement into text data, and in step 2, the server analyzes the text data and extracts the title "Memories of Youth." In step 3, the server selects information about other movies and actors related to "Memories of Youth," and in step 4, the generative AI model generates a prompt saying, "Based on the information about the movie 'Memories of Youth,' please generate a list of movies, TV shows, and documentaries related to the 1980s." Finally, in step 5, this information is provided to the user, encouraging a new viewing experience.

[0864] In addition, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0865] An embodiment of the present disclosure may be embodied as a system further including an emotion engine and an emotion analysis means. The emotion engine recognizes the emotion of a user. The emotion analysis means analyzes information such as facial expressions, voice, and text of the user to estimate the user's emotional state.

[0866] This allows the user to reminisce about past movies and TV shows and easily find related content. As a specific example, the emotion engine recognizes the emotion of the user while watching the content. The selection means customizes the content to be selected based on the emotion of the user recognized by the emotion engine. For example, when the user's emotion indicates a "nostalgic" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are more likely to be selected. On the other hand, when the user's emotion indicates a "boring" emotion, the selection means adjusts the parameters that are the selection criteria so that the currently provided content and similar content are less likely to be selected. This allows the user to enjoy content that matches his or her emotion.

[0867] The process flow will be explained below.

[0868] Step 1: Extracting information

[0869] The user reminisces about the past and makes comments.

[0870] The server uses an extraction means to analyze the user's comments and extract information about movies and television programs that the user has previously seen.

[0871] Step 2: Identify the era

[0872] The server uses the identification means to identify the eras in which movies and television programs that the user has previously seen were screened or broadcast, based on the information extracted in step 1.

[0873] Step 3: Select content

[0874] The server uses a selection means to select content related to movies and television programs the user has previously viewed, based on the information extracted in step 1. In this case, the server may select content that is the same or similar in at least one of the title, genre, cast, and theme song of the movies and television programs the user has previously viewed. The server may also select content that shows at least one of the background and events of the era identified in step 2.

[0875] Step 4: Providing content

[0876] The server uses a providing means to provide the content selected in step 3 to the user.

[0877] Step 5: Recognize emotions

[0878] The server uses an emotion engine to recognize the user's emotion.

[0879] Step 6: Customize your content

[0880] The server uses the selection means to customize the content selection based on the user's emotions recognized in step 5.

[0881] Example 2

[0882] Next, a description will be given of Example 2. In the following description, the data processing device 12 is referred to as a "server" and the robot 414 is referred to as a "terminal."

[0883] In conventional content provision systems, content is selected without considering the emotional state of the user, so content that satisfies the user is often not provided. In addition, it is difficult to recommend content based on the content the user has previously viewed or the emotional state at the time. For this reason, there is a demand for improving user satisfaction.

[0884] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring facial expression and voice data of the user in real time, an analysis means for analyzing the facial expression and voice data to estimate the emotional state of the user, a selection means for selecting content suitable for the emotional state of the user based on the analysis means, and a provision means for providing the content selected by the selection means to the user. This makes it possible to provide content suited to the emotional state of the user.

[0885] The "means for acquiring the user's facial expression and voice data in real time" refers to a means for capturing the user's facial expression and voice data in real time using an input device such as a camera or a microphone.

[0886] The "analysis means for analyzing the facial expression and voice data to infer the user's emotional state" refers to a means for inferring the user's emotional state based on the acquired facial expression and voice data using an analysis technique such as a machine learning algorithm.

[0887] The "selection means for selecting content appropriate to the user's emotional state based on the analysis means" refers to a means for adjusting content selection criteria in accordance with the estimated emotional state of the user and selecting the most suitable content.

[0888] The "means for providing the content selected by the selection means to the user" refers to a means for distributing the selected content to the user via a network.

[0889] "Content" refers to movies, television programs, music, videos, and other media forms of entertainment or information.

[0890] This system operates by combining multiple pieces of hardware and software in order to provide suitable content based on the emotional state of a user. A specific embodiment of this system will be described below.

[0891] The system is equipped with a means to capture the user's facial expression and voice data in real time. The camera and microphone installed on the terminal perform this role. As hardware, a general webcam and microphone can be used. These data are sent to the server via the MQTT protocol or HTTP POST request.

[0892] The server is equipped with software to analyze the received facial and voice data. To analyze the facial data, it uses OpenCV as an image processing library and Azure Face API as a cloud service. To analyze the voice data, it first converts the voice to text using Google Cloud Speech-to-Text API, and then analyzes the tone and pitch of the voice. From this, the analysis means estimates the user's emotional state.

[0893] Based on the estimated emotional state, the server uses a selection mechanism to choose the most suitable content. The selection mechanism uses deep learning frameworks such as TensorFlow and PyTorch to identify content that is suitable for the user's emotion. For example, if the user's emotion is estimated to be "nostalgic," content that was popular in the past or that the user has previously viewed is prioritized. On the other hand, if the emotion is estimated to be "boring," content from a different genre is provided.

[0894] The selected content is delivered to the user's device through content delivery networks such as AWS CloudFront and Azure Media Services, which allows for high-speed, high-quality streaming. Users can watch the content provided on their device and enjoy entertainment that matches their emotional state.

[0895] Examples of prompt statements

[0896] Emotion recognition prompts:

[0897] "Please analyze the emotions of the user currently viewing the video. Please tell us what emotions the user is feeling based on the following data: Facial expression data: {image_data}, Audio data: {audio_data}."

[0898] Content recommendation prompt:

[0899] "Sentiment analysis indicates that the user feels 'nostalgic'. Based on this user's sentiment, please provide relevant movies and TV shows from the content library below."

[0900] In this way, the present invention can improve user satisfaction by dynamically selecting and providing content based on the user's emotional state.

[0901] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0902] Program processing flow

[0903] Step 1:

[0904] The device uses a camera and a microphone to capture the user's facial expression and voice data in real time. As input, it receives image data from the camera and voice data from the microphone. The camera captures the user's face, and the microphone captures the user's speech. As output, it generates facial expression data (in image format) and voice data (in voice format).

[0905] Step 2:

[0906] The terminal transmits the acquired facial expression data and voice data to the server. As input, it uses the facial expression data and voice data generated in step 1. The data is sent to the server via the MQTT protocol or an HTTP POST request. As output, the facial expression data and voice data are transferred to the server.

[0907] Step 3:

[0908] The server analyzes the received data. First, the facial expression data is processed using an image processing library (e.g., OpenCV), and facial feature points are extracted using an analysis method (e.g., Azure Face API). The voice data is converted to text using Google Cloud Speech-to-Text API, and then the tone and pitch of the voice are analyzed. Facial expression data and voice data are received as input, and the user's emotional state (e.g., "nostalgic," "bored," etc.) is estimated as output.

[0909] Step 4:

[0910] The server selects optimal content based on the emotional state estimated by the analysis means. The selection means uses, for example, TensorFlow or PyTorch to identify content that corresponds to the user's emotional state. As input, it receives the estimated result of the emotional state, and if the user feels "nostalgic," it selects content that was popular in the past or content that the user has viewed in the past, and conversely, if the user feels "boring," it selects content of a different genre. As output, a list of selected content is generated.

[0911] Step 5:

[0912] The server delivers the selected content to the user's device. As input, it uses the list of content generated in step 4. It uses a delivery network such as AWS CloudFront or Azure Media Services to deliver the selected content at high speed and high quality. As output, the content is streamed to the user's device.

[0913] Step 6:

[0914] The user watches and enjoys the content provided on the terminal. As input, the user receives the content delivered in step 5. As output, the user can enjoy content that matches his / her emotional state.

[0915] (Application example 2)

[0916] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[0917] Conventional content delivery systems provide relevant content based on the user's past viewing history, but do not take into account the user's current emotional state, making it difficult to provide optimal content that meets the user's needs in a timely manner. In addition, there is a lack of functionality to flexibly adjust content appropriate for when the user is feeling emotionally "nostalgic" or "bored." This creates a problem of limiting the user's viewing experience.

[0918] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes an extraction means for analyzing a user's utterances and facial expressions to extract information about movies or TV programs that the user has previously watched, a selection means for selecting content related to the movie or TV program based on the information and the result of the user's emotion analysis, a provision means for providing the content to the user, an adjustment means for customizing the content to be provided based on the user's emotions, and a display means for displaying content that matches the user's emotions. This makes it possible to provide content optimized for the user's emotional state.

[0919] "User comments":

[0920] This refers to information verbally spoken by a user, and is the subject of analysis using natural language processing.

[0921] "Facial expression":

[0922] It refers to the user's facial movements and expressions, and is an important indicator for identifying emotions through analysis.

[0923] "information":

[0924] This refers to all data related to movies and TV shows, including metadata such as title, genre, and cast.

[0925] “Emotional analysis results”:

[0926] This data indicates the user's emotional state and refers to the results of analysis based on facial expressions and statements.

[0927] "Extraction means":

[0928] This refers to technology and systems for obtaining relevant information from a user's statements and facial expressions.

[0929] "Selection Method":

[0930] This refers to the algorithms and processes that select the most appropriate content based on extracted information and sentiment analysis results.

[0931] "Providing means":

[0932] This refers to technology for presenting selected content to users, and includes, for example, streaming services.

[0933] "Adjustment means":

[0934] This refers to the function of dynamically changing the content provided and recommendation algorithms based on the user's emotional state.

[0935] "Display means":

[0936] It means a device or interface for visually displaying content to a user.

[0937] "content":

[0938] This refers to all media that users watch or listen to, including movies, TV programs, videos, and music.

[0939] An embodiment of the present invention will be described.

[0940] The system analyzes the user's comments and facial expressions to recognize the user's emotions, and provides optimal content based on those emotions. Specifically, the system has the following configuration and operation.

[0941] System Configuration

[0942] The system includes the following major hardware and software:

[0943] Hardware:

[0944] Smartphone Camera

[0945] Smartphone Processors

[0946] software:

[0947] OpenCV: Used as an image processing library to capture the user's face with the camera.

[0948] EmotionRecognizer (Emotion Recognition Library): A model for identifying emotions from a user's face, using a deep learning-based model.

[0949] ContentRecommender (content recommendation engine): Equipped with an algorithm for recommending appropriate content based on emotion recognition results.

[0950] Operating procedure

[0951] 1. User sentiment analysis:

[0952] The server analyzes the user's facial image acquired from the smartphone camera using OpenCV and performs face detection.

[0953] The emotion recognition library (EmotionRecognizer) analyzes facial expressions from detected facial images and identifies the user's emotional state.

[0954] 2. Content Extraction:

[0955] Based on the analysis of the user's utterances and facial expressions, the server extracts information about previously viewed content.

[0956] The extracted information includes data such as movie or TV show title, genre, and cast.

[0957] 3. Content Selection:

[0958] The content recommendation engine (ContentRecommender) selects content suitable for the user based on the sentiment analysis results and extracted information.

[0959] If the user feels "nostalgic," popular dramas and movies from the past will be recommended, and if the user feels "boring," exciting new content will be recommended.

[0960] 4. Content provider:

[0961] The selected content is provided to the user's device by the server.

[0962] An application on the device visually presents the selected content to the user.

[0963] Examples

[0964] For example, if a user expresses a "nostalgic" emotion in front of the camera, the system will recommend hit dramas and movies from the past. Conversely, if the user feels "bored," the system will recommend exciting new movies and TV shows.

[0965] Examples of prompt statements

[0966] If users feel nostalgic, develop a video streaming app that recommends hit dramas and movies from the past.

[0967] As described above, the present invention makes it possible to provide content optimized for the emotional state of the user.

[0968] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0969] Processing steps and detailed explanations

[0970] Step 1:

[0971] The server acquires the user's face images in real time from the smartphone's camera. The input is the video stream from the camera. The output is frames of face images. Using OpenCV, image processing is performed to detect the face. Specifically, the video stream acquired from the camera is divided into frames, and a face detection algorithm is applied to identify the user's face.

[0972] Step 2:

[0973] The server passes the face-detected frames to EmotionRecognizer for emotion analysis. The input is the face image with the detected face. The output is the emotional state (e.g. happy, sad, nostalgic, bored). EmotionRecognizer uses a deep learning model to identify the user's emotion from the face image. Specifically, the face image is input to the model, and emotion labels are obtained as the output.

[0974] Step 3:

[0975] The server converts the user's speech into text using speech recognition technology. The input is speech data from a microphone. The output is text data recognized from the speech. Specifically, a speech recognition engine is used to convert the user's speech into text data in real time.

[0976] Step 4:

[0977] The server extracts information related to the user's past viewing history from the text data of the speech recognition results. The input is the recognized text data. The output is information about movies and TV shows that have been viewed in the past. Specifically, it uses natural language processing (NLP) technology to analyze the text data and extract keywords and metadata related to the past viewing history.

[0978] Step 5:

[0979] The server selects appropriate content based on the emotion analysis results and the extracted viewing history information. The input is the emotion analysis results and viewing history information. The output is a list of recommended content. Using the ContentRecommender engine, a recommendation algorithm based on emotion and past viewing history is applied. Specifically, if the emotional state is "nostalgic," hit dramas and movies from the past are selected, and if the emotional state is "boring," exciting new content is selected.

[0980] Step 6:

[0981] The server provides the selected content to the user's smartphone device. The input is a list of recommended content. The output is the content to be displayed on the device. Specifically, the server sends the URL and metadata of the selected content to the device so that the user can view it.

[0982] Step 7:

[0983] The terminal visually displays the provided content to the user. The input is the content information provided by the server. The output is the content displayed on the terminal screen. Specifically, the application UI is updated to allow the user to easily view the recommended content.

[0984] Through the above processing steps, content that is optimized for the user's emotional state is provided.

[0985] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0986] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0987] In the above embodiment, an example was given in which the specific process was performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the robot 414.

[0988] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0989] FIG. 9 is a diagram showing an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0990] These emotions are distributed in the 3 o'clock direction of emotion map 400 and usually fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0991] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out you go on emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0992] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0993] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0994] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in Fig. 10. Fig. 10 shows an example in which multiple emotions, "relief," "calm," and "encouraging," have similar emotion values.

[0995] Although the system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, the system according to the present disclosure is not necessarily implemented in a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program that runs on a personal computer, or an application that runs on a smartphone or the like. The method according to the present disclosure may be provided to a user in the form of SaaS (Software as a Service).

[0996] In the above embodiment, an example is given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to input data.

[0997] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable computer-readable non-transitory storage medium such as a Universal Serial Bus (USB) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0998] In addition, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 upon request from the data processing device 12.

[0999] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1000] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing the specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[1001] The hardware resource that executes the specific process may be one of these various processors, or may be a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[1002] As an example of a configuration using one processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a configuration using a processor that realizes the functions of the entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1003] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[1004] The above description and illustrations are detailed descriptions of the parts related to the technology of the present disclosure, and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, function, action, and effect is an example of the configuration, function, action, and effect of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above description and illustrations, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the above description and illustrations omit explanations of technical common sense that do not require explanation in order to enable the implementation of the technology of the present disclosure.

[1005] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or standard was specifically and individually indicated to be incorporated by reference.

[1006] The following is further disclosed regarding the above embodiment.

[1007] (Appendix 1) An extraction means for analyzing user comments and extracting information about movies or television programs that the user has previously watched; a selection means for selecting content related to the movie or television program based on the information; providing means for providing the content to the user; A system comprising:

[1008] (Appendix 2) the selection means selects the content having the same or similar at least one of the title, genre, cast, and theme song of the movie or TV program; 2. The system of claim 1.

[1009] (Appendix 3) Further comprising a means for identifying the era in which the movie or TV program was screened or broadcast based on the information; the selection means selects the content indicating at least one of a background and an event of the era; 10. The system of claim 1 or 2.

[1010] (Appendix 4) An emotion engine for recognizing an emotion of the user, The selection means customizes the content to be selected based on the emotion of the user. 4. A system according to any one of claims 1 to 3.

[1011] "Example 1" (Claim 1) An extraction means for analyzing user comments and extracting information about video works that the user has previously viewed; A means for identifying the era in which the video work was screened or broadcast based on the extracted information; a selection means for selecting content related to the video work based on the extracted information and the identified era; providing means for providing the selected content to the user; A system including:

[1012] (Claim 2) the selection means selects the content having the same or similar at least one of the title, genre, cast, and theme song of the video work; 2. The system of claim 1.

[1013] (Claim 3) the selection means selects the content indicating at least one of a background and an event of the identified era. 2. The system of claim 1.

[1014] "Application example 1" (Claim 1) An extraction means for analyzing a user's comments and extracting information about viewing content that the user has viewed in the past; A selection means for selecting related information related to the viewing content based on the information; providing means for providing the related information to the user; A generative AI model that generates new content based on the related information; A system including:

[1015] (Claim 2) the selection means selects the related information having at least one of the title, category, performers, and musical piece of the audiovisual content that is the same as or similar to the audiovisual content; 2. The system of claim 1.

[1016] (Claim 3) A determination means for determining a time period in which the viewing content was released or broadcast based on the information, the selection means selects the related information indicating at least one of a background and an event of the era; A generative AI model that generates new prompts related to the background and events of the period; 2. The system of claim 1.

[1017] "Example 2 of combining emotion engines" (Claim 1) A means for acquiring facial expression and voice data of a user in real time; an analysis means for analyzing the facial expression and voice data to estimate an emotional state of a user; a selection means for selecting content suited to the emotional state of the user based on the analysis means; providing means for providing the content selected by the selection means to a user; A system including:

[1018] (Claim 2) 2. The system according to claim 1, wherein the selection means is means for preferentially selecting content that was popular in the past or content that the user has previously viewed when the user's emotional state indicates "nostalgia."

[1019] (Claim 3) 2. The system according to claim 1, wherein the selection means is means for preferentially selecting content of a different genre or format when the user's emotional state indicates "boring."

[1020] "Application example 2 when combining emotion engines" New Claims (Claim 1) An extraction means for analyzing a user's utterances and facial expressions to extract information about movies or television programs that the user has previously watched; a selection means for selecting content related to the movie or TV program based on the information and a result of a user sentiment analysis; providing means for providing the content to the user; An adjustment means for customizing the content provided based on the user's emotions; A display means for displaying content that matches the emotion of the user; A system including:

[1021] (Claim 2) 2. The system according to claim 1, wherein the selection means selects the content that is the same or similar in at least one of the title, genre, cast, and theme song of the movie or TV program.

[1022] (Claim 3) Further comprising an identification means for identifying the era in which the movie or TV program was screened or broadcast based on the information; 2. The system according to claim 1, wherein the selection means selects the content that indicates at least one of a background and an event of the era. [Explanation of symbols]

[1023] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. An extraction means for analyzing user comments and extracting information about video works that the user has previously viewed; a selection means for selecting content related to the video work based on the extracted information; providing means for providing the selected content to the user; system.

2. The method further includes a means for identifying a time period in which the video work was screened or broadcast based on the extracted information, The selection means selects the content based on the identified era. The system of claim 1 .

3. Further comprising a generative AI model that generates a document that recommends or explains the selected content based on the selected content; the providing means provides the document together with the content to the user; The system of claim 1 .

4. The device further includes an analysis means for analyzing facial expression or voice data of the user to estimate an emotional state of the user, The selection means selects the content appropriate to the emotional state of the user. The system of claim 1 .

Images