system

Abstract

Provide a system. 【Solution means】 Means for acquiring conversation information, Means for analyzing acoustic data to identify the context of conversation, Means for detecting the end of conversation, Means for generating relevant topics using past conversation history and personal information, Means for providing the generated topics to the user, Perform context analysis of conversation using natural language processing technology, select the optimal topic based on profile information, Means for acquiring the current trend from an external information source to enhance the suitability of the topic, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] During a face-to-face conversation with a person, the problem that the conversation topic runs out may cause a sense of discomfort or unpleasantness. Also, in order to keep the conversation going smoothly, it is necessary to find an appropriate topic that follows the flow of the conversation and the other person's interests. Furthermore, even when trying to move on to a new topic during the conversation, it is difficult to obtain appropriate information on the spot. The present invention aims to address such problems, avoid interruptions in the conversation, and support the continuation of natural conversation.

Means for Solving the Problems

[0005] This invention provides a system that includes means for acquiring conversational information and analyzing acoustic data to identify the context of the conversation. This system includes means for detecting the end of a conversation and means for generating relevant topics using past conversation history and personal information. Furthermore, by providing the generated topics to the user, it is possible to maintain a natural flow of conversation. In addition, by including means for acquiring the latest information from an external knowledge database and using it for topic generation, it becomes possible to provide a wider variety of fresh topics. The system also includes means for automatically converting conversational information into text data using speech recognition, thereby realizing efficient data analysis and topic generation.

[0006] "Conversational information" refers to the content of dialogues captured as audio, as well as data related to the context of those conversations.

[0007] "Audio data" refers to information acquired as an audio signal, which is digital data obtained through methods such as audio recording.

[0008] "Context" refers to the flow of meaning and content within a conversation, or the situation in which it occurs.

[0009] "Means" refers to the apparatus, method, or technique used to achieve a particular purpose.

[0010] "Past conversation history" refers to a collection of data that includes records of conversations a user has had in the past.

[0011] "Personal information" refers to information related to a user, such as their attributes, interests, and hobbies.

[0012] "Topic generation" is the process of deriving new topics based on the flow and context of a conversation.

[0013] "Latest information" refers to current topics and news obtained from external knowledge databases, etc.

[0014] "Voice recognition" refers to the technology of converting voice into digital data and understanding its content.

[0015] "Character data" refers to data in text format converted from voice through voice recognition or the like.

Brief Explanation of Drawings

[0016] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units 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), and the like.

[0020] [[ID=XX]]

[0021] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor. <0000XX>

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the 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.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0030] 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 perceptible to the user 20 (e.g., audio 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 audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] The system of the present invention automatically acquires conversational information and provides appropriate topics to enable users to continue a smooth conversation. An embodiment thereof is shown below.

[0038] The user engages in conversation using an ear-worn IoT device. This device has the function of recording the user's conversation in real time and sending the audio data to a server. During or when the conversation is interrupted, the device sends the audio data to the server in a specific format.

[0039] The server uses speech recognition technology to transcribe the audio data received from the terminal. Furthermore, it uses natural language processing technology to analyze this text data and understand the context of the conversation. Based on this understanding of the context, the server compares it with the user's profile information and past conversation history to generate topics based on the user's interests and concerns.

[0040] The server retrieves the latest information from an external knowledge database to determine if the generated topic is appropriate for the current flow of conversation, and uses this information to facilitate topic generation. In this process, it uses an API to collect the necessary information in real time.

[0041] The device converts the topic sent from the server into speech and provides it to the user. This process allows the user to naturally resume the conversation.

[0042] As a concrete example, consider a scenario where a user is having a conversation with a friend at a cafe. If the conversation dies down, the device detects the pause and sends audio data to the server. Based on information about movies the user has recently been interested in, the server suggests, "Have you seen any of the recently talked-about movies?" This information is then provided to the user via audio through the device, allowing them to resume the conversation.

[0043] Thus, the present invention creates an environment in which users can continue a natural and smooth conversation by understanding the flow of the user's dialogue and providing new topics at the appropriate time.

[0044] The following describes the processing flow.

[0045] Step 1:

[0046] The user puts on an ear-worn IoT device and begins a conversation. The device continuously records this conversation in real time.

[0047] Step 2:

[0048] The device continuously monitors acoustic data during conversations and detects specific trigger conditions (e.g., duration of silence or speech patterns). When this trigger is detected, it sends the recorded audio data to the server.

[0049] Step 3:

[0050] The server receives audio data transmitted from the terminal and converts it into text data using speech recognition technology. Natural language processing is then applied to this text data to understand the context of the conversation.

[0051] Step 4:

[0052] Based on its understanding of the conversation context, the server refers to the user's profile information and past conversation history to identify relevant topics. Furthermore, it improves the relevance and freshness of topics by retrieving the latest relevant information in conjunction with external knowledge databases.

[0053] Step 5:

[0054] The server optimizes the identified topic, converts it to text format, and sends it to the terminal. During this process, it selects appropriate phrasing that takes the flow of the conversation into consideration.

[0055] Step 6:

[0056] The terminal transmits topics received from the server to the user using an audio output device. The user can then resume the conversation using the provided topics.

[0057] (Example 1)

[0058] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0059] In modern society, smooth and continuous conversation is a crucial element in building and maintaining relationships. However, a challenge arises when conversations break down due to a lack of topics, preventing intended communication from taking place. In particular, it is difficult to appropriately grasp the flow of the conversation while providing interesting topics, so there is a need to improve methods for generating topics to keep conversations going.

[0060] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0061] In this invention, the server includes means for acquiring voice information and analyzing acoustic information to identify the conversation situation, means for detecting the cessation of conversation, means for generating relevant topics using past dialogue history and user information, and means for providing the user with topics generated using a generative AI model. This enables the user to smoothly and naturally resume the conversation and continue the intended communication.

[0062] "Audio information" refers to acoustic signals, including the user's conversations and utterances.

[0063] "Acoustic information" refers to data such as frequency characteristics and volume used when analyzing audio information.

[0064] "Conversation context" refers to information that indicates the context, content, and progress of the current dialogue.

[0065] "Conversation interruption" refers to a state in which the dialogue between users is temporarily suspended.

[0066] "Dialogue history" refers to a record of conversations that a user has had in the past.

[0067] "User information" refers to personal information including data such as the user's interests, preferences, and past behavioral records.

[0068] "Related topics" refer to appropriate topics that are based on the user's interests and concerns, and that facilitate the smooth continuation of the conversation.

[0069] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate natural language text.

[0070] An "external information database" refers to an internet-based information source that provides the latest information and knowledge.

[0071] "Speech recognition" is a technology that analyzes speech information and converts it into text data.

[0072] This invention relates to a system that assists users in having smooth conversations. The system helps to continue a conversation by acquiring voice information and suggesting appropriate topics based on that information.

[0073] The user engages in conversation using an ear-worn device. This device records the user's conversation in real time using a built-in voice input device. This voice information is transmitted to a server via wireless communication technology.

[0074] The server uses speech recognition software (e.g., a general speech-to-text service) to convert the collected speech information into text. Next, natural language processing techniques are used to analyze this text information and identify the context and situation of the conversation. The user's dialogue history and personal profile are stored in a database, and based on this information, topics suitable for the user are generated.

[0075] The server uses a generative AI model to create new topics. During this process, it retrieves the latest information from an external database to verify the topic's relevance. This process selects topics that are likely to attract user interest.

[0076] The terminal takes topics provided by the server, converts them into speech using speech synthesis technology (e.g., a general speech conversion system), and presents them to the user. This series of operations allows the user to continue the conversation naturally and smoothly.

[0077] As a concrete example, consider a situation where a user is in a restaurant with a friend. If the conversation pauses, the device detects this and sends an audio message to the server. The server analyzes the content and suggests a movie topic related to the user's recent interests. This suggestion is provided to the user as an audio message such as, "Do you know any recent movies that are popular?", and the conversation resumes.

[0078] An example of a prompt for a generative AI model would be: "Please suggest a new conversation starter about a topic the user has recently been interested in. Please consider the context of the conversation the user is currently participating in." This allows the system to generate and suggest an appropriate topic to the user.

[0079] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0080] Step 1:

[0081] The user initiates a conversation using an ear-worn device. The terminal uses the voice input device within this device to record the user's conversation in real time. The input data is the user's voice, which is transmitted to the server via wireless communication technology. As output, the recorded voice data is transferred to the server.

[0082] Step 2:

[0083] The server receives audio data transmitted from the terminal. The input is the transmitted audio data. Speech recognition software is used to convert the audio data into text data. This data processing results in the content of the conversation being output as text.

[0084] Step 3:

[0085] The server analyzes the obtained text data using natural language processing techniques. The input is the converted text data. Through this data processing, the server identifies the situation and context of the conversation. The output is contextual information of the analyzed conversation, which makes the user's dialogue intent clearer.

[0086] Step 4:

[0087] The server references the user's user information and past conversation history based on the analysis results. The input is conversation context information, which is then processed by matching it with information in the database. This generates relevant topics that the user is likely to be interested in. The output is the generated topic ideas.

[0088] Step 5:

[0089] The server uses a generative AI model to give concrete form to topics. The input is a topic idea, and the AI ​​generates specific topics through data generation calculations. The output is a topic suggestion that can be presented to the user.

[0090] Step 6:

[0091] The server references an external information database to improve the timeliness and relevance of generated topics. Input consists of concretized topic proposals, which are augmented with the latest information obtained from the external database. Output consists of topics that are current and relevant.

[0092] Step 7:

[0093] The terminal receives a topic generated from the server and converts it into speech using speech synthesis technology. The input is the generated topic, and the output is the speech result. Specifically, it provides the topic as speech to the user's ears, thereby assisting in resuming the conversation.

[0094] (Application Example 1)

[0095] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0096] In modern society, providing appropriate topics of conversation is crucial for smooth communication within the family and socially. However, finding a suitable new topic when a conversation stalls is not easy, often leading to stagnation. Therefore, there is a need for support technologies to maintain a smooth flow of conversation.

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

[0098] In this invention, the server includes means for acquiring conversation information, means for analyzing acoustic data to identify the context of the conversation, means for detecting the end of a conversation, means for generating relevant topics using past conversation history and personal information, means for providing the generated topics to the user, and means for performing contextual analysis of the conversation using natural language processing technology, selecting the most appropriate topic based on profile information, and acquiring current trends from external information sources to improve the relevance of the topic. This enables the rapid detection of interruptions in conversation and the automatic provision of topics that match the user's interests and timing, thereby allowing for a natural and smooth continuation of conversation.

[0099] "Conversational information" refers to general data about human interaction obtained from audio and text data.

[0100] "Acoustic data" refers to a collection of signal data collected as speech, and is fundamental information for identifying the context of a conversation.

[0101] "Natural language processing technology" is a technology that uses computers to analyze, understand, and generate human language.

[0102] "Profile information" refers to attribute information about a user, such as their personal preferences, interests, and history, and is used when selecting appropriate topics.

[0103] "External information sources" refer to information providers accessible from outside the system, such as the internet and cloud services, and are the sources from which information on the latest trends and developments is obtained.

[0104] "Topic generation" is the process of proposing new topics to present to users based on acquired conversational information and the results of contextual analysis.

[0105] "Contextual analysis" is an analytical method that involves understanding the meaning of expressions and words used in a conversation and grasping the overall flow and situation of the conversation.

[0106] "Relevance" refers to the degree to which a generated topic fits the user's current conversation and interests.

[0107] This system will be implemented in a home-use conversational assistance robot. The robot will acquire voice data through an ear-worn IoT device and send that data to a server. The server will convert the voice data into text using speech recognition technology. The "speech_recognition" library will be used in this process. The converted text will be analyzed using natural language processing technology. This analysis will use a technology called "NLPProcessor" to understand the context of the conversation.

[0108] Based on the analysis results, the server generates appropriate topics using the user's profile information and past conversation history. During this process, it obtains current trends from external sources to verify topic suitability. The latest information is retrieved in real time from "external_database_api" and used. The generated topics are then sent back to the robot and provided to the user as speech using "text_to_speech" technology. In this speech generation process, the robot uses a speaker to allow the user to hear the speech, enabling smooth conversation continuation.

[0109] As a concrete example, consider a situation where a home robot encounters a lull in conversation between a visitor and a user. The robot detects this and suggests a new topic, such as, "Shall we talk about the latest trend in home gardening?" This topic is selected based on the user's interests and also takes into account the latest trends from external sources.

[0110] An example of a prompt for a generative AI model would be: "The current topic of conversation is 'home gardening'. Please suggest new related topics based on your user profile and current trends."

[0111] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0112] Step 1:

[0113] The terminal (robot) uses a voice input device to record conversations between the user and their surroundings in real time. The recorded voice data becomes input and is prepared to be sent to the server. The voice data is temporarily stored as a digital signal within the terminal.

[0114] Step 2:

[0115] The server converts the received audio data into text data using the "speech_recognition" library. The input for this step is audio data, and the output is text data. In the speech recognition process, an acoustic model and a grammatical model are used to ensure accurate text conversion.

[0116] Step 3:

[0117] The server analyzes the generated text data using a natural language processing technique called "NLPProcessor" to identify the context of the conversation. The input is text data, and contextual information is generated as output. Through this analysis, the server understands the keywords and sentence structure used, and grasps the purpose and content of the conversation.

[0118] Step 4:

[0119] The server matches user profile information and past conversation history to generate relevant topics based on context. The input is contextual information and the user profile, and the output is a new topic. An internal algorithm selects topics that match the user's interests.

[0120] Step 5:

[0121] The server uses "external_database_api" to retrieve the latest trends from external sources, improving the relevance of generated topics. The input is a hypothetical topic, and the output topic is generated incorporating external trend information. This operation enables the provision of topics that reflect social trends in real time.

[0122] Step 6:

[0123] The server finally sends the generated topic to the terminal. The input is the integrated topic, and the terminal receives and processes it as output. The terminal then waits, preparing for speech synthesis.

[0124] Step 7:

[0125] The device converts the received topic into speech using "text-to-speech" technology and delivers it to the user through the speaker. The input is the topic text, and the output is generated speech. The topic is conveyed with natural intonation using a speech synthesis engine.

[0126] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0127] The system of the present invention assists users in smoothly conducting natural conversations with others during a conversation. The system acquires conversational information through the user's ear-worn IoT device and includes an emotion engine that recognizes the user's emotions based on that data.

[0128] When a user engages in conversation, the device records audio data in real time and sends it to the server. The server uses speech recognition technology to convert this audio data into text data and analyzes the context of the conversation. Simultaneously, an emotion engine analyzes the user's emotional state based on factors such as tone, volume, and speed of voice.

[0129] The server generates relevant topics based on acquired sentiment and contextual information. These generated topics are optimized to reflect the user's current sentiment. It's also possible to enhance topic relevance by connecting to external knowledge databases and retrieving the latest topics.

[0130] The device notifies the user of these optimized topics via voice. This allows the user to resume the conversation without interruption and continue the dialogue in a natural flow. Furthermore, the topics offered are selected with the aim of maintaining the user's emotions in a positive state.

[0131] As a concrete example, consider a scenario where a user is chatting with a friend, but the conversation suddenly becomes a little subdued. The device detects this situation and sends audio data to the server. The server uses an emotion engine to analyze the user's slightly downcast mood and generates information about topics that the user might be interested in and that could brighten their mood, such as "interesting recent movies or hobbies." By providing this information to the user, they can steer the conversation back in a positive direction.

[0132] This system allows users to maintain the flow of conversation while also addressing their emotional needs.

[0133] The following describes the processing flow.

[0134] Step 1:

[0135] The user initiates a conversation using an ear-worn IoT device. The device has the capability to record this conversation in real time and begins acquiring acoustic data.

[0136] Step 2:

[0137] The device periodically sends the audio data being recorded to the server. This data includes features necessary for emotion recognition, such as the tone and speed of the voice.

[0138] Step 3:

[0139] The server converts the received audio data into text data using speech recognition. This makes the content of the conversation analyzable as text.

[0140] Step 4:

[0141] On the server, the emotion engine analyzes the acoustic data and evaluates the user's emotional state based on the characteristics of the voice. Based on the results of this analysis, it determines whether the current emotion is positive or negative.

[0142] Step 5:

[0143] The server performs a comprehensive analysis by combining contextual information from text data of conversations with sentiment analysis results. Based on past conversation history and new information from external knowledge databases, it generates topics optimized for the user.

[0144] Step 6:

[0145] The server sends the generated topic to the user's device at the time that best suits their mood. This topic is designed to facilitate the smooth continuation of the conversation.

[0146] Step 7:

[0147] The device provides the user with topics received from the server as audio output. The user can then introduce new topics based on the presented information and resume the conversation in a natural way.

[0148] (Example 2)

[0149] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0150] In modern communication, challenges exist such as interruptions in conversation and the inability to appropriately respond to the other person's emotions. In such situations, the flow of conversation stalls, making it difficult to maintain smooth interaction. Furthermore, the inability to offer topics relevant to the other person's feelings and situation can hinder deeper communication.

[0151] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0152] In this invention, the server includes means for acquiring conversational information and analyzing acoustic signals to identify the context of the conversation, means for analyzing emotional states based on the acoustic signals, and means for generating relevant topics using a generative artificial intelligence model. This enables the user to maintain the flow of the conversation and provide appropriate topics in accordance with the other person's emotions.

[0153] "Conversation information" refers to audio data related to user dialogue and the results of its analysis.

[0154] "Acoustic signals" refer to data collected electronically from the sounds emitted by the user.

[0155] "Context identification methods" refer to technologies that analyze the content and flow of a conversation based on acoustic signals to understand the current theme and topic.

[0156] "Methods for analyzing emotional states" refer to techniques that analyze tone, speed, and voice intensity from acoustic data to estimate the speaker's emotions and psychological state.

[0157] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information and ideas.

[0158] "Methods for generating topics" refer to the process of suggesting new conversation themes or topics based on the current context and emotional state.

[0159] "Means of providing information to users" refers to methods and technologies for transmitting generated information and topics to users and presenting them in various forms.

[0160] "External knowledge sources" refer to external databases and information sources that contain the latest information.

[0161] "Speech recognition" is a technology that analyzes speech and automatically converts it into text data.

[0162] This invention relates to a system that assists users in natural conversation using an ear-worn information processing device. The system aims to maintain the flow of the user's conversation and provide appropriate topics.

[0163] The server receives acoustic signals transmitted from the user. The terminal has the capability to record acoustic data in real time and send it to the server. This data is used to analyze the content and tone of the conversation. The server uses speech recognition software to convert this acoustic data into text data. A general-purpose speech processing platform is used as the specific software for "speech recognition."

[0164] The analyzed data is used for contextual analysis and sentiment analysis. Sentiment analysis employs techniques that estimate emotional states from the tone and intensity of acoustic signals. These techniques utilize speech analysis technology as a means of analyzing emotional states.

[0165] Next, the server uses a generative artificial intelligence model to generate relevant topics based on the user's current emotional state and the flow of the conversation. In this process, the algorithm uses the generative AI model to select appropriate information to improve the user experience. As an example of a specific prompt for the generative AI model, the instruction "Suggest topics that will evoke positive emotions in the user" can be used.

[0166] Finally, the terminal converts the information provided by the server into speech and notifies the user. Text-to-speech technology is used here. This allows the user to continue the conversation based on the provided topic, enabling smooth dialogue.

[0167] This system provides an easy-to-use solution for uninterrupted, real-time communication. To achieve this, it comprehensively utilizes a variety of software and technologies.

[0168] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0169] Step 1:

[0170] The user initiates a conversation using an ear-worn information processing device. The device acquires acoustic signals in real time and inputs them into the terminal as audio data. This audio data is recorded for subsequent processing.

[0171] Step 2:

[0172] The terminal packets the recorded audio data and sends it to the server. A secure communication channel is used during data transmission to ensure data reliability and security. As a result, the server can receive the audio data.

[0173] Step 3:

[0174] The server receives the audio data as input and converts it into text data using a speech recognition engine. This conversion utilizes a specific speech recognition algorithm to identify the context within the audio data. The output is text data.

[0175] Step 4:

[0176] The server performs sentiment analysis based on text data. Here, it estimates the user's emotional state from characteristics such as tone, voice intensity, and speed of the acoustic signal. The sentiment analysis engine uses a specific algorithm to quantify the emotional state. The output is data indicating the emotional state.

[0177] Step 5:

[0178] The server uses a generative artificial intelligence model to generate topics based on the obtained text data and emotional state. The generative AI model utilizes prompts to identify appropriate topics that match the user's current emotions. The output is a list of suggested topics.

[0179] Step 6:

[0180] Upon receiving a topic generated by the server, the terminal uses its text-to-speech function to send an audio notification to the user. This notification allows the user to receive a topic to continue the conversation at the appropriate time. The output is information provided to the user as an audio prompt.

[0181] (Application Example 2)

[0182] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0183] In modern society, individual conversations are often interrupted, and their smooth progression, which addresses emotional needs, is hindered. In such cases, there is a need for effective support to help users maintain a natural flow of conversation and continue positive dialogue. However, conventional technologies have struggled to accurately grasp the emotional state of users and generate relevant topics based on that understanding.

[0184] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0185] In this invention, the server includes means for analyzing acoustic data to identify the context of a conversation, means for analyzing the user's emotional state from the audio data, and means for generating relevant topics based on the user's emotional state and the context of the conversation. This allows the user to maintain a natural flow of conversation while being offered positive topics that correspond to their emotional state.

[0186] "Means for acquiring conversational information" refers to devices or technologies for collecting voice data emitted by users.

[0187] "Means of analyzing acoustic data to identify the context of a conversation" refers to a process or technology that analyzes acquired audio data to identify the content and theme of a conversation.

[0188] "Means for detecting the end of a conversation" refers to a function or algorithm for determining that a conversation has been interrupted or ended.

[0189] "Methods for analyzing a user's emotional state from voice data" refers to technologies that evaluate a user's emotions based on factors such as tone, speed, and volume of their voice.

[0190] "Means for generating relevant topics based on the user's emotional state and the context of the conversation" refers to methods for creating new and relevant topics or subjects based on the results of the user's sentiment analysis and the content of the conversation.

[0191] "Means of providing generated topics to users" refers to a mechanism for communicating new topics or suggestions presented to users.

[0192] An "external knowledge database" is an external source of information or a system that stores the latest information and data.

[0193] "A means of automatically converting conversational information into text data using speech recognition" refers to a technology that converts speech into text format, making it analyzable.

[0194] "Methods for generating prompt sentences using generative AI models" refers to methods that use artificial intelligence technology to generate prompts that output the most appropriate response or suggestion to the user.

[0195] The system for realizing this invention includes a process for collecting voice data, evaluating the user's emotional state, and generating relevant topics based on that evaluation. The system comprises, as its main components, a consumer robot equipped with a microphone, a server for processing the voice data, and a user interface for providing the generated topics to the user.

[0196] First, the consumer robot, acting as the terminal, collects the user's speech in real time through its built-in microphone. The collected speech data is sent to a server. The server uses the Google® Cloud Speech-to-Text API to convert the speech data into text data using speech recognition technology. This converted text data is then used to analyze the content of the speech.

[0197] Next, the server uses TENSORFLOW® to analyze the user's emotional state based on characteristics such as tone, speed, and volume of speech. Based on the emotional information obtained in this way and contextual information from speech recognition, it generates relevant topics using the OpenAI® GPT model. This generative AI model provides optimal prompts that match the user's current emotional state and maintain the flow of the conversation.

[0198] The generated topics are communicated to the user through the robot's speaker. The user can continue the conversation in a natural flow and deepen their emotional connection.

[0199] For example, if a user is having a quiet conversation with their family at the dinner table, the robot could suggest, "It seems a new restaurant recently opened in the neighborhood. Why don't you all go and check it out sometime?" This could help to revitalize the conversation.

[0200] An example of a prompt message would be: "Suggest topics that the user might be interested in and that will help them maintain a positive mood. Current topics include travel, hobbies, and music."

[0201] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0202] Step 1:

[0203] The device collects the user's speech via a built-in microphone. The input is the user's voice, and the output is digital audio data. This audio data is transmitted to the server in real time.

[0204] Step 2:

[0205] The server converts the received audio data into text data using the Google Cloud Speech-to-Text API. The input is audio data, and the output is speech-based text data. The server then uses the converted text data to prepare for identifying the context of the conversation.

[0206] Step 3:

[0207] The server analyzes text data and uses TensorFlow to evaluate the user's emotional state. The input is text data, and the output is the evaluation of the user's emotional state. The server analyzes elements such as tone, speed, and intensity of sound to represent the emotional state numerically.

[0208] Step 4:

[0209] The server uses the OpenAI GPT model to generate relevant topics, taking into account the user's emotional state and the context of the conversation. The input is emotion rating data and context data, and the output is the newly proposed topic. The server optimizes the generated prompt sentences and prepares them for the user.

[0210] Step 5:

[0211] The terminal notifies the user of generated topics via voice. The input is a topic suggested by the server, and the output is the voice output to the user. The robot uses a configured speaker to provide a new topic that prompts the user to resume the conversation.

[0212] 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 audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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 audio data.

[0213] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of 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">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0214] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0215] [Second Embodiment]

[0216] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0217] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0218] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0219] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0220] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0221] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0222] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0223] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0224] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0225] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0226] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0227] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0228] The system of the present invention automatically acquires conversational information and provides appropriate topics to enable users to continue a smooth conversation. An embodiment thereof is shown below.

[0229] The user engages in conversation using an ear-worn IoT device. This device has the function of recording the user's conversation in real time and sending the audio data to a server. During or when the conversation is interrupted, the device sends the audio data to the server in a specific format.

[0230] The server uses speech recognition technology to transcribe the audio data received from the terminal. Furthermore, it uses natural language processing technology to analyze this text data and understand the context of the conversation. Based on this understanding of the context, the server compares it with the user's profile information and past conversation history to generate topics based on the user's interests and concerns.

[0231] The server retrieves the latest information from an external knowledge database to determine if the generated topic is appropriate for the current flow of conversation, and uses this information to facilitate topic generation. In this process, it uses an API to collect the necessary information in real time.

[0232] The device converts the topic sent from the server into speech and provides it to the user. This process allows the user to naturally resume the conversation.

[0233] As a concrete example, consider a scenario where a user is having a conversation with a friend at a cafe. If the conversation dies down, the device detects the pause and sends audio data to the server. Based on information about movies the user has recently been interested in, the server suggests, "Have you seen any of the recently talked-about movies?" This information is then provided to the user via audio through the device, allowing them to resume the conversation.

[0234] Thus, the present invention creates an environment in which users can continue a natural and smooth conversation by understanding the flow of the user's dialogue and providing new topics at the appropriate time.

[0235] The following describes the processing flow.

[0236] Step 1:

[0237] The user puts on an ear-worn IoT device and begins a conversation. The device continuously records this conversation in real time.

[0238] Step 2:

[0239] The device continuously monitors acoustic data during conversations and detects specific trigger conditions (e.g., duration of silence or speech patterns). When this trigger is detected, it sends the recorded audio data to the server.

[0240] Step 3:

[0241] The server receives audio data transmitted from the terminal and converts it into text data using speech recognition technology. Natural language processing is then applied to this text data to understand the context of the conversation.

[0242] Step 4:

[0243] Based on its understanding of the conversation context, the server refers to the user's profile information and past conversation history to identify relevant topics. Furthermore, it improves the relevance and freshness of topics by retrieving the latest relevant information in conjunction with external knowledge databases.

[0244] Step 5:

[0245] The server optimizes the identified topic, converts it to text format, and sends it to the terminal. During this process, it selects appropriate phrasing that takes the flow of the conversation into consideration.

[0246] Step 6:

[0247] The terminal transmits topics received from the server to the user using an audio output device. The user can then resume the conversation using the provided topics.

[0248] (Example 1)

[0249] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0250] In modern society, smooth and continuous conversation is a crucial element in building and maintaining relationships. However, a challenge arises when conversations break down due to a lack of topics, preventing intended communication from taking place. In particular, it is difficult to appropriately grasp the flow of the conversation while providing interesting topics, so there is a need to improve methods for generating topics to keep conversations going.

[0251] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0252] In this invention, the server includes means for acquiring voice information and analyzing acoustic information to identify the conversation situation, means for detecting the cessation of conversation, means for generating relevant topics using past dialogue history and user information, and means for providing the user with topics generated using a generative AI model. This enables the user to smoothly and naturally resume the conversation and continue the intended communication.

[0253] "Audio information" refers to acoustic signals, including the user's conversations and utterances.

[0254] "Acoustic information" refers to data such as frequency characteristics and volume used when analyzing audio information.

[0255] "Conversation context" refers to information that indicates the context, content, and progress of the current dialogue.

[0256] "Conversation interruption" refers to a state in which the dialogue between users is temporarily suspended.

[0257] "Dialogue history" refers to a record of conversations that a user has had in the past.

[0258] "User information" refers to personal information including data such as the user's interests, preferences, and past behavioral records.

[0259] "Related topics" refer to appropriate topics that are based on the user's interests and concerns, and that facilitate the smooth continuation of the conversation.

[0260] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate natural language text.

[0261] An "external information database" refers to an internet-based information source that provides the latest information and knowledge.

[0262] "Speech recognition" is a technology that analyzes speech information and converts it into text data.

[0263] This invention relates to a system that assists users in having smooth conversations. The system helps to continue a conversation by acquiring voice information and suggesting appropriate topics based on that information.

[0264] The user engages in conversation using an ear-worn device. This device records the user's conversation in real time using a built-in voice input device. This voice information is transmitted to a server via wireless communication technology.

[0265] The server uses speech recognition software (e.g., a general speech-to-text service) to convert the collected speech information into text. Next, natural language processing techniques are used to analyze this text information and identify the context and situation of the conversation. The user's dialogue history and personal profile are stored in a database, and based on this information, topics suitable for the user are generated.

[0266] The server uses a generative AI model to create new topics. During this process, it retrieves the latest information from an external database to verify the topic's relevance. This process selects topics that are likely to attract user interest.

[0267] The terminal takes topics provided by the server, converts them into speech using speech synthesis technology (e.g., a general speech conversion system), and presents them to the user. This series of operations allows the user to continue the conversation naturally and smoothly.

[0268] As a concrete example, consider a situation where a user is in a restaurant with a friend. If the conversation pauses, the device detects this and sends an audio message to the server. The server analyzes the content and suggests a movie topic related to the user's recent interests. This suggestion is provided to the user as an audio message such as, "Do you know any recent movies that are popular?", and the conversation resumes.

[0269] An example of a prompt for a generative AI model would be: "Please suggest a new conversation starter about a topic the user has recently been interested in. Please consider the context of the conversation the user is currently participating in." This allows the system to generate and suggest an appropriate topic to the user.

[0270] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0271] Step 1:

[0272] The user initiates a conversation using an ear-worn device. The terminal uses the voice input device within this device to record the user's conversation in real time. The input data is the user's voice, which is transmitted to the server via wireless communication technology. As output, the recorded voice data is transferred to the server.

[0273] Step 2:

[0274] The server receives audio data transmitted from the terminal. The input is the transmitted audio data. Speech recognition software is used to convert the audio data into text data. This data processing results in the content of the conversation being output as text.

[0275] Step 3:

[0276] The server analyzes the obtained text data using natural language processing technology. The input is the converted text data. Through this data operation, the conversation situation and context are identified. The output is the analyzed conversation context information, which makes the user's dialogue intention clearer.

[0277] Step 4:

[0278] The server refers to the user's user information and past conversation history based on the analysis results. The input is the conversation context information, and data processing for collation with the information in the database is performed. Thereby, relevant topics that the user may be interested in are generated. The output is the generated topic ideas.

[0279] Step 5:

[0280] The server uses the generation AI model to concretize the topic. The input is the topic idea, and a specific topic is created through data generation operations by the AI. The output is a topic proposal that can be presented to the user.

[0281] Step 6:

[0282] The server refers to the external information database and improves the currency and compatibility of the generated topic. The input is the concretized topic proposal, and data reinforcement is performed using the latest information obtained from the external database. The output is a topic with up-to-date and relevant content.

[0283] Step 7:

[0284] The terminal receives the topic generated by the server and converts this topic into voice using voice synthesis technology. The input is the generated topic, and the result of voice conversion is output. As a specific operation, the topic is provided to the user's ear as voice, thereby assisting in the resumption of the conversation.

[0285] (Application Example 1)

[0286] Next, Application Example 1 will be described. In the following description, the data processing apparatus 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[0287] In modern society, in order for conversations within the family or in a social setting to proceed smoothly, it is important to provide appropriate topics. However, it is not easy to find an appropriate new topic when the conversation breaks off, and the flow of the conversation often stagnates. For this reason, support technologies for keeping the flow of conversation smooth are in demand.

[0288] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 1 is realized by the following means.

[0289] In this invention, the server includes means for acquiring conversation information, means for analyzing acoustic data to identify the context of the conversation, means for detecting the end of the conversation, means for generating relevant topics using past conversation history and personal information, means for providing the generated topics to the user, and means for performing context analysis of the conversation using natural language processing technology, selecting an optimal topic based on profile information, and obtaining current trends from an external information source to enhance the suitability of the topic. Thereby, it is possible to quickly detect breaks during a conversation and automatically provide topics according to the user's interests and timing, enabling a natural and smooth continuation of the conversation.

[0290] "Conversation information" is general data regarding human-to-human conversations obtained from acoustic data or text data.

[0291] "Acoustic data" is a set of signal data collected as sound, and is basic information for identifying the context of a conversation.

[0292] "Natural language processing technology" is a technology for analyzing, understanding, and generating human language using a computer.

[0293] "Profile information" refers to attribute information about a user, such as their personal preferences, interests, and history, and is used when selecting appropriate topics.

[0294] "External information sources" refer to information providers accessible from outside the system, such as the internet and cloud services, and are the sources from which information on the latest trends and developments is obtained.

[0295] "Topic generation" is the process of proposing new topics to present to users based on acquired conversational information and the results of contextual analysis.

[0296] "Contextual analysis" is an analytical method that involves understanding the meaning of expressions and words used in a conversation and grasping the overall flow and situation of the conversation.

[0297] "Relevance" refers to the degree to which a generated topic fits the user's current conversation and interests.

[0298] This system will be implemented in a home-use conversational assistance robot. The robot will acquire voice data through an ear-worn IoT device and send that data to a server. The server will convert the voice data into text using speech recognition technology. The "speech_recognition" library will be used in this process. The converted text will be analyzed using natural language processing technology. This analysis will use a technology called "NLPProcessor" to understand the context of the conversation.

[0299] Based on the analysis results, the server generates appropriate topics using the user's profile information and past conversation history. During this process, it obtains current trends from external sources to verify topic suitability. The latest information is retrieved in real time from "external_database_api" and used. The generated topics are then sent back to the robot and provided to the user as speech using "text_to_speech" technology. In this speech generation process, the robot uses a speaker to allow the user to hear the speech, enabling smooth conversation continuation.

[0300] As a concrete example, consider a situation where a home robot encounters a lull in conversation between a visitor and a user. The robot detects this and suggests a new topic, such as, "Shall we talk about the latest trend in home gardening?" This topic is selected based on the user's interests and also takes into account the latest trends from external sources.

[0301] An example of a prompt for a generative AI model would be: "The current topic of conversation is 'home gardening'. Please suggest new related topics based on your user profile and current trends."

[0302] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0303] Step 1:

[0304] The terminal (robot) uses a voice input device to record conversations between the user and their surroundings in real time. The recorded voice data becomes input and is prepared to be sent to the server. The voice data is temporarily stored as a digital signal within the terminal.

[0305] Step 2:

[0306] The server converts the received voice data into text data using the "speech_recognition" library. The input for this step is voice data, and text data is generated as the output. In the process of speech recognition, an acoustic model and a grammar model are used to perform accurate text conversion.

[0307] Step 3:

[0308] The server analyzes the generated text data using natural language processing technology called "NLPProcessor" to identify the context of the conversation. The input is text data, and context information is generated as the output. Through this analysis, the keywords and sentence structures used are understood to grasp the purpose and content of the conversation.

[0309] Step 4:

[0310] The server matches the user's profile information and past conversation history, and generates relevant topics based on the context. The input is context information and the user profile, and a new topic is obtained as the output. Through an internal algorithm, a topic that matches the user's interests is selected.

[0311] Step 5:

[0312] The server uses "external_database_api" to obtain the latest trends from an external information source to enhance the suitability of the generated topics. The input is a tentative topic, and an output topic that takes into account external trend information is generated. This operation enables the provision of topics that reflect the trends of society in real time.

[0313] Step 6:

[0314] The server finally sends the generated topic to the terminal. The input is the integrated topic, and reception processing on the terminal side is performed as the output. The terminal waits for this as preparation for speech synthesis.

[0315] Step 7:

[0316] The device converts the received topic into speech using "text-to-speech" technology and delivers it to the user through the speaker. The input is the topic text, and the output is generated speech. The topic is conveyed with natural intonation using a speech synthesis engine.

[0317] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0318] The system of the present invention assists users in smoothly conducting natural conversations with others during a conversation. The system acquires conversational information through the user's ear-worn IoT device and includes an emotion engine that recognizes the user's emotions based on that data.

[0319] When a user engages in conversation, the device records audio data in real time and sends it to the server. The server uses speech recognition technology to convert this audio data into text data and analyzes the context of the conversation. Simultaneously, an emotion engine analyzes the user's emotional state based on factors such as tone, volume, and speed of voice.

[0320] The server generates relevant topics based on acquired sentiment and contextual information. These generated topics are optimized to reflect the user's current sentiment. It's also possible to enhance topic relevance by connecting to external knowledge databases and retrieving the latest topics.

[0321] The device notifies the user of these optimized topics via voice. This allows the user to resume the conversation without interruption and continue the dialogue in a natural flow. Furthermore, the topics offered are selected with the aim of maintaining the user's emotions in a positive state.

[0322] As a concrete example, consider a scenario where a user is chatting with a friend, but the conversation suddenly becomes a little subdued. The device detects this situation and sends audio data to the server. The server uses an emotion engine to analyze the user's slightly downcast mood and generates information about topics that the user might be interested in and that could brighten their mood, such as "interesting recent movies or hobbies." By providing this information to the user, they can steer the conversation back in a positive direction.

[0323] This system allows users to maintain the flow of conversation while also addressing their emotional needs.

[0324] The following describes the processing flow.

[0325] Step 1:

[0326] The user initiates a conversation using an ear-worn IoT device. The device has the capability to record this conversation in real time and begins acquiring acoustic data.

[0327] Step 2:

[0328] The device periodically sends the audio data being recorded to the server. This data includes features necessary for emotion recognition, such as the tone and speed of the voice.

[0329] Step 3:

[0330] The server converts the received audio data into text data using speech recognition. This makes the content of the conversation analyzable as text.

[0331] Step 4:

[0332] On the server, the emotion engine analyzes the acoustic data and evaluates the user's emotional state based on the characteristics of the voice. Based on the results of this analysis, it determines whether the current emotion is positive or negative.

[0333] Step 5:

[0334] The server performs a comprehensive analysis by combining contextual information from text data of conversations with sentiment analysis results. Based on past conversation history and new information from external knowledge databases, it generates topics optimized for the user.

[0335] Step 6:

[0336] The server sends the generated topic to the user's device at the time that best suits their mood. This topic is designed to facilitate the smooth continuation of the conversation.

[0337] Step 7:

[0338] The device provides the user with topics received from the server as audio output. The user can then introduce new topics based on the presented information and resume the conversation in a natural way.

[0339] (Example 2)

[0340] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0341] In modern communication, challenges exist such as interruptions in conversation and the inability to appropriately respond to the other person's emotions. In such situations, the flow of conversation stalls, making it difficult to maintain smooth interaction. Furthermore, the inability to offer topics relevant to the other person's feelings and situation can hinder deeper communication.

[0342] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0343] In this invention, the server includes means for acquiring conversational information and analyzing acoustic signals to identify the context of the conversation, means for analyzing emotional states based on the acoustic signals, and means for generating relevant topics using a generative artificial intelligence model. This enables the user to maintain the flow of the conversation and provide appropriate topics in accordance with the other person's emotions.

[0344] "Conversation information" refers to audio data related to user dialogue and the results of its analysis.

[0345] "Acoustic signals" refer to data collected electronically from the sounds emitted by the user.

[0346] "Context identification methods" refer to technologies that analyze the content and flow of a conversation based on acoustic signals to understand the current theme and topic.

[0347] "Methods for analyzing emotional states" refer to techniques that analyze tone, speed, and voice intensity from acoustic data to estimate the speaker's emotions and psychological state.

[0348] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information and ideas.

[0349] "Methods for generating topics" refer to the process of suggesting new conversation themes or topics based on the current context and emotional state.

[0350] "Means of providing information to users" refers to methods and technologies for communicating generated information and topics to users and presenting them in various forms.

[0351] "External knowledge sources" refer to external databases and information sources that contain the latest information.

[0352] "Speech recognition" is a technology that analyzes speech and automatically converts it into text data.

[0353] This invention relates to a system that assists users in natural conversation using an ear-worn information processing device. The system aims to maintain the flow of the user's conversation and provide appropriate topics.

[0354] The server receives acoustic signals transmitted from the user. The terminal has the capability to record acoustic data in real time and send it to the server. This data is used to analyze the content and tone of the conversation. The server uses speech recognition software to convert this acoustic data into text data. A general-purpose speech processing platform is used as the specific software for "speech recognition."

[0355] The analyzed data is used for contextual analysis and sentiment analysis. Sentiment analysis employs techniques that estimate emotional states from the tone and intensity of acoustic signals. These techniques utilize speech analysis technology as a means of analyzing emotional states.

[0356] Next, the server uses a generative artificial intelligence model to generate relevant topics based on the user's current emotional state and the flow of the conversation. In this process, the algorithm uses the generative AI model to select appropriate information to improve the user experience. As an example of a specific prompt for the generative AI model, the instruction "Suggest topics that will evoke positive emotions in the user" can be used.

[0357] Finally, the terminal converts the information provided by the server into speech and notifies the user. Text-to-speech technology is used here. This allows the user to continue the conversation based on the provided topic, enabling smooth dialogue.

[0358] This system provides an easy-to-use solution for uninterrupted, real-time communication. To achieve this, it comprehensively utilizes a variety of software and technologies.

[0359] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0360] Step 1:

[0361] The user initiates a conversation using an ear-worn information processing device. The device acquires acoustic signals in real time and inputs them into the terminal as audio data. This audio data is recorded for subsequent processing.

[0362] Step 2:

[0363] The terminal packets the recorded audio data and sends it to the server. A secure communication channel is used during data transmission to ensure data reliability and security. As a result, the server can receive the audio data.

[0364] Step 3:

[0365] The server receives the audio data as input and converts it into text data using a speech recognition engine. This conversion utilizes a specific speech recognition algorithm to identify the context within the audio data. The output is text data.

[0366] Step 4:

[0367] The server performs sentiment analysis based on text data. Here, it estimates the user's emotional state from characteristics such as tone, voice intensity, and speed of the acoustic signal. The sentiment analysis engine uses a specific algorithm to quantify the emotional state. The output is data indicating the emotional state.

[0368] Step 5:

[0369] The server uses a generative artificial intelligence model to generate topics based on the obtained text data and emotional state. The generative AI model utilizes prompts to identify appropriate topics that match the user's current emotions. The output is a list of suggested topics.

[0370] Step 6:

[0371] Upon receiving a topic generated by the server, the terminal uses its text-to-speech function to send an audio notification to the user. This notification allows the user to receive a topic to continue the conversation at the appropriate time. The output is information provided to the user as an audio prompt.

[0372] (Application Example 2)

[0373] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0374] In modern society, individual conversations are often interrupted, and their smooth progression, which addresses emotional needs, is hindered. In such cases, there is a need for effective support to help users maintain a natural flow of conversation and continue positive dialogue. However, conventional technologies have struggled to accurately grasp the emotional state of users and generate relevant topics based on that understanding.

[0375] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0376] In this invention, the server includes means for analyzing acoustic data to identify the context of a conversation, means for analyzing the user's emotional state from the audio data, and means for generating relevant topics based on the user's emotional state and the context of the conversation. This allows the user to maintain a natural flow of conversation while being offered positive topics that correspond to their emotional state.

[0377] "Means for acquiring conversational information" refers to devices or technologies for collecting voice data emitted by users.

[0378] "Means of analyzing acoustic data to identify the context of a conversation" refers to a process or technology that analyzes acquired audio data to identify the content and theme of a conversation.

[0379] "Means for detecting the end of a conversation" refers to a function or algorithm for determining that a conversation has been interrupted or ended.

[0380] "Methods for analyzing a user's emotional state from voice data" refers to technologies that evaluate a user's emotions based on factors such as tone, speed, and volume of their voice.

[0381] "Means for generating relevant topics based on the user's emotional state and the context of the conversation" refers to methods for creating new and relevant topics or subjects based on the results of the user's sentiment analysis and the content of the conversation.

[0382] "Means of providing generated topics to users" refers to a mechanism for communicating new topics or suggestions presented to users.

[0383] An "external knowledge database" is an external source of information or a system that stores the latest information and data.

[0384] "Methods for automatically converting conversational information into text data using speech recognition" refers to technologies that convert speech into text format, making it analyzable.

[0385] "Methods for generating prompt text using generative AI models" refer to methods that use artificial intelligence technology to generate prompts that output the most appropriate response or suggestion to the user.

[0386] The system for realizing this invention includes a process for collecting voice data, evaluating the user's emotional state, and generating relevant topics based on that evaluation. The system comprises, as its main components, a consumer robot equipped with a microphone, a server for processing the voice data, and a user interface for providing the generated topics to the user.

[0387] First, the consumer robot, acting as the terminal, collects the user's speech in real time through its built-in microphone. The collected speech data is sent to a server. The server uses the Google Cloud Speech-to-Text API to convert the speech data into text data using speech recognition technology. This converted text data is then used to analyze the content of the speech.

[0388] Next, the server uses TensorFlow to analyze the user's emotional state based on features such as tone, speed, and volume of speech. Based on the emotional information obtained in this way and the contextual information from speech recognition, it uses the OpenAI GPT model to generate relevant topics. This generative AI model provides optimal prompts that match the user's current emotional state and maintain the flow of the conversation.

[0389] The generated topics are communicated to the user through the robot's speaker. The user can continue the conversation in a natural flow and deepen their emotional connection.

[0390] For example, if a user is having a quiet conversation with their family at the dinner table, the robot could suggest, "It seems a new restaurant recently opened in the neighborhood. Why don't you all go and check it out sometime?" This could help to revitalize the conversation.

[0391] An example of a prompt message would be: "Suggest topics that the user might be interested in and that will help them maintain a positive mood. Current topics include travel, hobbies, and music."

[0392] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0393] Step 1:

[0394] The device collects the user's speech via a built-in microphone. The input is the user's voice, and the output is digital audio data. This audio data is transmitted to the server in real time.

[0395] Step 2:

[0396] The server converts the received audio data into text data using the Google Cloud Speech-to-Text API. The input is audio data, and the output is speech-based text data. The server then uses the converted text data to prepare for identifying the context of the conversation.

[0397] Step 3:

[0398] The server analyzes text data and uses TensorFlow to evaluate the user's emotional state. The input is text data, and the output is the evaluation of the user's emotional state. The server analyzes elements such as tone, speed, and intensity of sound to represent the emotional state numerically.

[0399] Step 4:

[0400] The server uses the OpenAI GPT model to generate relevant topics, taking into account the user's emotional state and the context of the conversation. The input is emotion rating data and context data, and the output is the newly proposed topic. The server optimizes the generated prompt sentences and prepares them for the user.

[0401] Step 5:

[0402] The terminal notifies the user of generated topics via voice. The input is a topic suggested by the server, and the output is the voice output to the user. The robot uses a configured speaker to provide a new topic that prompts the user to resume the conversation.

[0403] 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 audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0404] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0405] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0406] [Third Embodiment]

[0407] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0408] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0409] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0410] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0411] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0412] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0413] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0414] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0415] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0416] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0417] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0418] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0419] The system of the present invention automatically acquires conversational information and provides appropriate topics to enable users to continue a smooth conversation. An embodiment thereof is shown below.

[0420] The user engages in conversation using an ear-worn IoT device. This device has the function of recording the user's conversation in real time and sending the audio data to a server. During or when the conversation is interrupted, the device sends the audio data to the server in a specific format.

[0421] The server uses speech recognition technology to transcribe the audio data received from the terminal. Furthermore, it uses natural language processing technology to analyze this text data and understand the context of the conversation. Based on this understanding of the context, the server compares it with the user's profile information and past conversation history to generate topics based on the user's interests and concerns.

[0422] The server retrieves the latest information from an external knowledge database to determine if the generated topic is appropriate for the current flow of conversation, and uses this information to facilitate topic generation. In this process, it uses an API to collect the necessary information in real time.

[0423] The device converts the topic sent from the server into speech and provides it to the user. This process allows the user to naturally resume the conversation.

[0424] As a concrete example, consider a scenario where a user is having a conversation with a friend at a cafe. If the conversation dies down, the device detects the pause and sends audio data to the server. Based on information about movies the user has recently been interested in, the server suggests, "Have you seen any of the recently talked-about movies?" This information is then provided to the user via audio through the device, allowing them to resume the conversation.

[0425] Thus, the present invention creates an environment in which users can continue a natural and smooth conversation by understanding the flow of the user's dialogue and providing new topics at the appropriate time.

[0426] The following describes the processing flow.

[0427] Step 1:

[0428] The user puts on an ear-worn IoT device and begins a conversation. The device continuously records this conversation in real time.

[0429] Step 2:

[0430] The device continuously monitors acoustic data during conversations and detects specific trigger conditions (e.g., duration of silence or speech patterns). When this trigger is detected, it sends the recorded audio data to the server.

[0431] Step 3:

[0432] The server receives audio data transmitted from the terminal and converts it into text data using speech recognition technology. Natural language processing is then applied to this text data to understand the context of the conversation.

[0433] Step 4:

[0434] Based on its understanding of the conversation context, the server refers to the user's profile information and past conversation history to identify relevant topics. Furthermore, it improves the relevance and freshness of topics by retrieving the latest relevant information in conjunction with external knowledge databases.

[0435] Step 5:

[0436] The server optimizes the identified topic, converts it to text format, and sends it to the terminal. During this process, it selects appropriate phrasing that takes the flow of the conversation into consideration.

[0437] Step 6:

[0438] The terminal transmits topics received from the server to the user using an audio output device. The user can then resume the conversation using the provided topics.

[0439] (Example 1)

[0440] Next, we will describe Example 1. 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."

[0441] In modern society, smooth and continuous conversation is a crucial element in building and maintaining relationships. However, a challenge arises when conversations break down due to a lack of topics, preventing intended communication from taking place. In particular, it is difficult to appropriately grasp the flow of the conversation while providing interesting topics, so there is a need to improve methods for generating topics to keep conversations going.

[0442] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0443] In this invention, the server includes means for acquiring voice information and analyzing acoustic information to identify the conversation situation, means for detecting the cessation of conversation, means for generating relevant topics using past dialogue history and user information, and means for providing the user with topics generated using a generative AI model. This enables the user to smoothly and naturally resume the conversation and continue the intended communication.

[0444] "Audio information" refers to acoustic signals, including the user's conversations and utterances.

[0445] "Acoustic information" refers to data such as frequency characteristics and volume used when analyzing audio information.

[0446] "Conversation context" refers to information that indicates the context, content, and progress of the current dialogue.

[0447] "Conversation interruption" refers to a state in which the dialogue between users is temporarily suspended.

[0448] "Dialogue history" refers to a record of conversations that a user has had in the past.

[0449] "User information" refers to personal information including data such as the user's interests, preferences, and past behavioral records.

[0450] "Related topics" refer to appropriate topics that are based on the user's interests and concerns, and that facilitate the smooth continuation of the conversation.

[0451] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate natural language text.

[0452] An "external information database" refers to an internet-based information source that provides the latest information and knowledge.

[0453] "Speech recognition" is a technology that analyzes speech information and converts it into text data.

[0454] This invention relates to a system that assists users in having smooth conversations. The system helps to continue a conversation by acquiring voice information and suggesting appropriate topics based on that information.

[0455] The user engages in conversation using an ear-worn device. This device records the user's conversation in real time using a built-in voice input device. This voice information is transmitted to a server via wireless communication technology.

[0456] The server uses speech recognition software (e.g., a general speech-to-text service) to convert the collected speech information into text. Next, natural language processing techniques are used to analyze this text information and identify the context and situation of the conversation. The user's dialogue history and personal profile are stored in a database, and based on this information, topics suitable for the user are generated.

[0457] The server uses a generative AI model to create new topics. During this process, it retrieves the latest information from an external database to verify the topic's relevance. This process selects topics that are likely to attract user interest.

[0458] The terminal takes topics provided by the server, converts them into speech using speech synthesis technology (e.g., a general speech conversion system), and presents them to the user. This series of operations allows the user to continue the conversation naturally and smoothly.

[0459] As a concrete example, consider a situation where a user is in a restaurant with a friend. If the conversation pauses, the device detects this and sends an audio message to the server. The server analyzes the content and suggests a movie topic related to the user's recent interests. This suggestion is provided to the user as an audio message such as, "Do you know any recent movies that are popular?", and the conversation resumes.

[0460] An example of a prompt for a generative AI model would be: "Please suggest a new conversation starter about a topic the user has recently been interested in. Please consider the context of the conversation the user is currently participating in." This allows the system to generate and suggest an appropriate topic to the user.

[0461] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0462] Step 1:

[0463] The user initiates a conversation using an ear-worn device. The terminal uses the voice input device within this device to record the user's conversation in real time. The input data is the user's voice, which is transmitted to the server via wireless communication technology. As output, the recorded voice data is transferred to the server.

[0464] Step 2:

[0465] The server receives audio data transmitted from the terminal. The input is the transmitted audio data. Speech recognition software is used to convert the audio data into text data. This data processing results in the content of the conversation being output as text.

[0466] Step 3:

[0467] The server analyzes the obtained text data using natural language processing techniques. The input is the converted text data. Through this data processing, the server identifies the situation and context of the conversation. The output is contextual information of the analyzed conversation, which makes the user's dialogue intent clearer.

[0468] Step 4:

[0469] The server references the user's user information and past conversation history based on the analysis results. The input is conversation context information, which is then processed by matching it with information in the database. This generates relevant topics that the user is likely to be interested in. The output is the generated topic ideas.

[0470] Step 5:

[0471] The server uses a generative AI model to give concrete form to topics. The input is a topic idea, and the AI ​​generates specific topics through data generation calculations. The output is a topic suggestion that can be presented to the user.

[0472] Step 6:

[0473] The server references an external information database to improve the timeliness and relevance of generated topics. Input consists of concretized topic proposals, which are augmented with the latest information obtained from the external database. Output consists of topics that are current and relevant.

[0474] Step 7:

[0475] The terminal receives a topic generated from the server and converts it into speech using speech synthesis technology. The input is the generated topic, and the output is the speech result. Specifically, it provides the topic as speech to the user's ears, thereby assisting in resuming the conversation.

[0476] (Application Example 1)

[0477] Next, we will explain Application Example 1. In the following explanation, 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."

[0478] In modern society, providing appropriate topics of conversation is crucial for smooth communication within the family and socially. However, finding a suitable new topic when a conversation stalls is not easy, often leading to stagnation. Therefore, there is a need for support technologies to maintain a smooth flow of conversation.

[0479] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0480] In this invention, the server includes means for acquiring conversation information, means for analyzing acoustic data to identify the context of the conversation, means for detecting the end of a conversation, means for generating relevant topics using past conversation history and personal information, means for providing the generated topics to the user, and means for performing contextual analysis of the conversation using natural language processing technology, selecting the most appropriate topic based on profile information, and acquiring current trends from external information sources to improve the relevance of the topic. This enables the rapid detection of interruptions in conversation and the automatic provision of topics that match the user's interests and timing, thereby allowing for a natural and smooth continuation of conversation.

[0481] "Conversational information" refers to general data about human interaction obtained from audio and text data.

[0482] "Acoustic data" refers to a collection of signal data collected as speech, and is fundamental information for identifying the context of a conversation.

[0483] "Natural language processing technology" is a technology that uses computers to analyze, understand, and generate human language.

[0484] "Profile information" refers to attribute information about a user, such as their personal preferences, interests, and history, and is used when selecting appropriate topics.

[0485] "External information sources" refer to information providers accessible from outside the system, such as the internet and cloud services, and are the sources from which information on the latest trends and developments is obtained.

[0486] "Topic generation" is the process of proposing new topics to present to users based on acquired conversational information and the results of contextual analysis.

[0487] "Contextual analysis" is an analytical method that involves understanding the meaning of expressions and words used in a conversation and grasping the overall flow and situation of the conversation.

[0488] "Relevance" refers to the degree to which a generated topic fits the user's current conversation and interests.

[0489] This system will be implemented in a home-use conversational assistance robot. The robot will acquire voice data through an ear-worn IoT device and send that data to a server. The server will convert the voice data into text using speech recognition technology. The "speech_recognition" library will be used in this process. The converted text will be analyzed using natural language processing technology. This analysis will use a technology called "NLPProcessor" to understand the context of the conversation.

[0490] Based on the analysis results, the server generates appropriate topics using the user's profile information and past conversation history. During this process, it obtains current trends from external sources to verify topic suitability. The latest information is retrieved in real time from "external_database_api" and used. The generated topics are then sent back to the robot and provided to the user as speech using "text_to_speech" technology. In this speech generation process, the robot uses a speaker to allow the user to hear the speech, enabling smooth conversation continuation.

[0491] As a concrete example, consider a situation where a home robot encounters a lull in conversation between a visitor and a user. The robot detects this and suggests a new topic, such as, "Shall we talk about the latest trend in home gardening?" This topic is selected based on the user's interests and also takes into account the latest trends from external sources.

[0492] An example of a prompt for a generative AI model would be: "The current topic of conversation is 'home gardening'. Please suggest new related topics based on your user profile and current trends."

[0493] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0494] Step 1:

[0495] The terminal (robot) uses a voice input device to record conversations between the user and their surroundings in real time. The recorded voice data becomes input and is prepared to be sent to the server. The voice data is temporarily stored as a digital signal within the terminal.

[0496] Step 2:

[0497] The server converts the received audio data into text data using the "speech_recognition" library. The input for this step is audio data, and the output is text data. In the speech recognition process, an acoustic model and a grammatical model are used to ensure accurate text conversion.

[0498] Step 3:

[0499] The server analyzes the generated text data using a natural language processing technique called "NLPProcessor" to identify the context of the conversation. The input is text data, and contextual information is generated as output. Through this analysis, the server understands the keywords and sentence structure used, and grasps the purpose and content of the conversation.

[0500] Step 4:

[0501] The server matches user profile information and past conversation history to generate relevant topics based on context. The input is contextual information and the user profile, and the output is a new topic. An internal algorithm selects topics that match the user's interests.

[0502] Step 5:

[0503] The server uses "external_database_api" to retrieve the latest trends from external sources, improving the relevance of generated topics. The input is a hypothetical topic, and the output topic is generated incorporating external trend information. This operation enables the provision of topics that reflect social trends in real time.

[0504] Step 6:

[0505] The server finally sends the generated topic to the terminal. The input is the integrated topic, and the terminal receives and processes it as output. The terminal then waits, preparing for speech synthesis.

[0506] Step 7:

[0507] The device converts the received topic into speech using "text-to-speech" technology and delivers it to the user through the speaker. The input is the topic text, and the output is generated speech. The topic is conveyed with natural intonation using a speech synthesis engine.

[0508] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0509] The system of the present invention assists users in smoothly conducting natural conversations with others during a conversation. The system acquires conversational information through the user's ear-worn IoT device and includes an emotion engine that recognizes the user's emotions based on that data.

[0510] When a user engages in conversation, the device records audio data in real time and sends it to the server. The server uses speech recognition technology to convert this audio data into text data and analyzes the context of the conversation. Simultaneously, an emotion engine analyzes the user's emotional state based on factors such as tone, volume, and speed of voice.

[0511] The server generates relevant topics based on acquired sentiment and contextual information. These generated topics are optimized to reflect the user's current sentiment. It's also possible to enhance topic relevance by connecting to external knowledge databases and retrieving the latest topics.

[0512] The device notifies the user of these optimized topics via voice. This allows the user to resume the conversation without interruption and continue the dialogue in a natural flow. Furthermore, the topics offered are selected with the aim of maintaining the user's emotions in a positive state.

[0513] As a concrete example, consider a scenario where a user is chatting with a friend, but the conversation suddenly becomes a little subdued. The device detects this situation and sends audio data to the server. The server uses an emotion engine to analyze the user's slightly downcast mood and generates information about topics that the user might be interested in and that could brighten their mood, such as "interesting recent movies or hobbies." By providing this information to the user, they can steer the conversation back in a positive direction.

[0514] This system allows users to maintain the flow of conversation while also addressing their emotional needs.

[0515] The following describes the processing flow.

[0516] Step 1:

[0517] The user initiates a conversation using an ear-worn IoT device. The device has the capability to record this conversation in real time and begins acquiring acoustic data.

[0518] Step 2:

[0519] The device periodically sends the audio data being recorded to the server. This data includes features necessary for emotion recognition, such as the tone and speed of the voice.

[0520] Step 3:

[0521] The server converts the received audio data into text data using speech recognition. This makes the content of the conversation analyzable as text.

[0522] Step 4:

[0523] On the server, the emotion engine analyzes the acoustic data and evaluates the user's emotional state based on the characteristics of the voice. Based on the results of this analysis, it determines whether the current emotion is positive or negative.

[0524] Step 5:

[0525] The server performs a comprehensive analysis by combining contextual information from text data of conversations with sentiment analysis results. Based on past conversation history and new information from external knowledge databases, it generates topics optimized for the user.

[0526] Step 6:

[0527] The server sends the generated topic to the user's device at the time that best suits their mood. This topic is designed to facilitate the smooth continuation of the conversation.

[0528] Step 7:

[0529] The device provides the user with topics received from the server as audio output. The user can then introduce new topics based on the presented information and resume the conversation in a natural way.

[0530] (Example 2)

[0531] Next, we will describe Example 2. 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."

[0532] In modern communication, challenges exist such as interruptions in conversation and the inability to appropriately respond to the other person's emotions. In such situations, the flow of conversation stalls, making it difficult to maintain smooth interaction. Furthermore, the inability to offer topics relevant to the other person's feelings and situation can hinder deeper communication.

[0533] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0534] In this invention, the server includes means for acquiring conversational information and analyzing acoustic signals to identify the context of the conversation, means for analyzing emotional states based on the acoustic signals, and means for generating relevant topics using a generative artificial intelligence model. This enables the user to maintain the flow of the conversation and provide appropriate topics in accordance with the other person's emotions.

[0535] "Conversation information" refers to audio data related to user dialogue and the results of its analysis.

[0536] "Acoustic signals" refer to data collected electronically from the sounds emitted by the user.

[0537] "Context identification methods" refer to technologies that analyze the content and flow of a conversation based on acoustic signals to understand the current theme and topic.

[0538] "Methods for analyzing emotional states" refer to techniques that analyze tone, speed, and voice intensity from acoustic data to estimate the speaker's emotions and psychological state.

[0539] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information and ideas.

[0540] "Methods for generating topics" refer to the process of suggesting new conversation themes or topics based on the current context and emotional state.

[0541] "Means of providing information to users" refers to methods and technologies for communicating generated information and topics to users and presenting them in various forms.

[0542] "External knowledge sources" refer to external databases and information sources that contain the latest information.

[0543] "Speech recognition" is a technology that analyzes speech and automatically converts it into text data.

[0544] This invention relates to a system that assists users in natural conversation using an ear-worn information processing device. The system aims to maintain the flow of the user's conversation and provide appropriate topics.

[0545] The server receives acoustic signals transmitted from the user. The terminal has the capability to record acoustic data in real time and send it to the server. This data is used to analyze the content and tone of the conversation. The server uses speech recognition software to convert this acoustic data into text data. A general-purpose speech processing platform is used as the specific software for "speech recognition."

[0546] The analyzed data is used for contextual analysis and sentiment analysis. Sentiment analysis employs techniques that estimate emotional states from the tone and intensity of acoustic signals. These techniques utilize speech analysis technology as a means of analyzing emotional states.

[0547] Next, the server uses a generative artificial intelligence model to generate relevant topics based on the user's current emotional state and the flow of the conversation. In this process, the algorithm uses the generative AI model to select appropriate information to improve the user experience. As an example of a specific prompt for the generative AI model, the instruction "Suggest topics that will evoke positive emotions in the user" can be used.

[0548] Finally, the terminal converts the information provided by the server into speech and notifies the user. Text-to-speech technology is used here. This allows the user to continue the conversation based on the provided topic, enabling smooth dialogue.

[0549] This system provides an easy-to-use solution for uninterrupted, real-time communication. To achieve this, it comprehensively utilizes a variety of software and technologies.

[0550] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0551] Step 1:

[0552] The user initiates a conversation using an ear-worn information processing device. The device acquires acoustic signals in real time and inputs them into the terminal as audio data. This audio data is recorded for subsequent processing.

[0553] Step 2:

[0554] The terminal packets the recorded audio data and sends it to the server. A secure communication channel is used during data transmission to ensure data reliability and security. As a result, the server can receive the audio data.

[0555] Step 3:

[0556] The server receives the audio data as input and converts it into text data using a speech recognition engine. This conversion utilizes a specific speech recognition algorithm to identify the context within the audio data. The output is text data.

[0557] Step 4:

[0558] The server performs sentiment analysis based on text data. Here, it estimates the user's emotional state from characteristics such as tone, voice intensity, and speed of the acoustic signal. The sentiment analysis engine uses a specific algorithm to quantify the emotional state. The output is data indicating the emotional state.

[0559] Step 5:

[0560] The server uses a generative artificial intelligence model to generate topics based on the obtained text data and emotional state. The generative AI model utilizes prompts to identify appropriate topics that match the user's current emotions. The output is a list of suggested topics.

[0561] Step 6:

[0562] Upon receiving a topic generated by the server, the terminal uses its text-to-speech function to send an audio notification to the user. This notification allows the user to receive a topic to continue the conversation at the appropriate time. The output is information provided to the user as an audio prompt.

[0563] (Application Example 2)

[0564] Next, we will explain application example 2. In the following explanation, 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."

[0565] In modern society, individual conversations are often interrupted, and their smooth progression, which addresses emotional needs, is hindered. In such cases, there is a need for effective support to help users maintain a natural flow of conversation and continue positive dialogue. However, conventional technologies have struggled to accurately grasp the emotional state of users and generate relevant topics based on that understanding.

[0566] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0567] In this invention, the server includes means for analyzing acoustic data to identify the context of a conversation, means for analyzing the user's emotional state from the audio data, and means for generating relevant topics based on the user's emotional state and the context of the conversation. This allows the user to maintain a natural flow of conversation while being offered positive topics that correspond to their emotional state.

[0568] "Means for acquiring conversational information" refers to devices or technologies for collecting voice data emitted by users.

[0569] "Means of analyzing acoustic data to identify the context of a conversation" refers to a process or technology that analyzes acquired audio data to identify the content and theme of a conversation.

[0570] "Means for detecting the end of a conversation" refers to a function or algorithm for determining that a conversation has been interrupted or ended.

[0571] "Methods for analyzing a user's emotional state from voice data" refers to technologies that evaluate a user's emotions based on factors such as tone, speed, and volume of their voice.

[0572] "Means for generating relevant topics based on the user's emotional state and the context of the conversation" refers to methods for creating new and relevant topics or subjects based on the results of the user's sentiment analysis and the content of the conversation.

[0573] "Means of providing generated topics to users" refers to a mechanism for communicating new topics or suggestions presented to users.

[0574] An "external knowledge database" is an external source of information or a system that stores the latest information and data.

[0575] "Methods for automatically converting conversational information into text data using speech recognition" refers to technologies that convert speech into text format, making it analyzable.

[0576] "Methods for generating prompt text using generative AI models" refer to methods that use artificial intelligence technology to generate prompts that output the most appropriate response or suggestion to the user.

[0577] The system for realizing this invention includes a process for collecting voice data, evaluating the user's emotional state, and generating relevant topics based on that evaluation. The system comprises, as its main components, a consumer robot equipped with a microphone, a server for processing the voice data, and a user interface for providing the generated topics to the user.

[0578] First, the consumer robot, acting as the terminal, collects the user's speech in real time through its built-in microphone. The collected speech data is sent to a server. The server uses the Google Cloud Speech-to-Text API to convert the speech data into text data using speech recognition technology. This converted text data is then used to analyze the content of the speech.

[0579] Next, the server uses TensorFlow to analyze the user's emotional state based on features such as tone, speed, and volume of speech. Based on the emotional information obtained in this way and the contextual information from speech recognition, it uses the OpenAI GPT model to generate relevant topics. This generative AI model provides optimal prompts that match the user's current emotional state and maintain the flow of the conversation.

[0580] The generated topics are communicated to the user through the robot's speaker. The user can continue the conversation in a natural flow and deepen their emotional connection.

[0581] For example, if a user is having a quiet conversation with their family at the dinner table, the robot could suggest, "It seems a new restaurant recently opened in the neighborhood. Why don't you all go and check it out sometime?" This could help to revitalize the conversation.

[0582] An example of a prompt message would be: "Suggest topics that the user might be interested in and that will help them maintain a positive mood. Current topics include travel, hobbies, and music."

[0583] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0584] Step 1:

[0585] The device collects the user's speech via a built-in microphone. The input is the user's voice, and the output is digital audio data. This audio data is transmitted to the server in real time.

[0586] Step 2:

[0587] The server converts the received audio data into text data using the Google Cloud Speech-to-Text API. The input is audio data, and the output is speech-based text data. The server then uses the converted text data to prepare for identifying the context of the conversation.

[0588] Step 3:

[0589] The server analyzes text data and uses TensorFlow to evaluate the user's emotional state. The input is text data, and the output is the evaluation of the user's emotional state. The server analyzes elements such as tone, speed, and intensity of sound to represent the emotional state numerically.

[0590] Step 4:

[0591] The server uses the OpenAI GPT model to generate relevant topics, taking into account the user's emotional state and the context of the conversation. The input is emotion rating data and context data, and the output is the newly proposed topic. The server optimizes the generated prompt sentences and prepares them for the user.

[0592] Step 5:

[0593] The terminal notifies the user of generated topics via voice. The input is a topic suggested by the server, and the output is the voice output to the user. The robot uses a configured speaker to provide a new topic that prompts the user to resume the conversation.

[0594] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0595] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0596] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0597] [Fourth Embodiment]

[0598] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0599] As shown in Figure 7, the 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.

[0600] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0601] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0602] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0603] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0604] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0605] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0606] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0607] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0608] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0609] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0610] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0611] The system of the present invention automatically acquires conversational information and provides appropriate topics to enable users to continue a smooth conversation. An embodiment thereof is shown below.

[0612] The user engages in conversation using an ear-worn IoT device. This device has the function of recording the user's conversation in real time and sending the audio data to a server. During or when the conversation is interrupted, the device sends the audio data to the server in a specific format.

[0613] The server uses speech recognition technology to transcribe the audio data received from the terminal. Furthermore, it uses natural language processing technology to analyze this text data and understand the context of the conversation. Based on this understanding of the context, the server compares it with the user's profile information and past conversation history to generate topics based on the user's interests and concerns.

[0614] The server retrieves the latest information from an external knowledge database to determine if the generated topic is appropriate for the current flow of conversation, and uses this information to facilitate topic generation. In this process, it uses an API to collect the necessary information in real time.

[0615] The device converts the topic sent from the server into speech and provides it to the user. This process allows the user to naturally resume the conversation.

[0616] As a concrete example, consider a scenario where a user is having a conversation with a friend at a cafe. If the conversation dies down, the device detects the pause and sends audio data to the server. Based on information about movies the user has recently been interested in, the server suggests, "Have you seen any of the recently talked-about movies?" This information is then provided to the user via audio through the device, allowing them to resume the conversation.

[0617] Thus, the present invention creates an environment in which users can continue a natural and smooth conversation by understanding the flow of the user's dialogue and providing new topics at the appropriate time.

[0618] The following describes the processing flow.

[0619] Step 1:

[0620] The user puts on an ear-worn IoT device and begins a conversation. The device continuously records this conversation in real time.

[0621] Step 2:

[0622] The device continuously monitors acoustic data during conversations and detects specific trigger conditions (e.g., duration of silence or speech patterns). When this trigger is detected, it sends the recorded audio data to the server.

[0623] Step 3:

[0624] The server receives audio data transmitted from the terminal and converts it into text data using speech recognition technology. Natural language processing is then applied to this text data to understand the context of the conversation.

[0625] Step 4:

[0626] Based on its understanding of the conversation context, the server refers to the user's profile information and past conversation history to identify relevant topics. Furthermore, it improves the relevance and freshness of topics by retrieving the latest relevant information in conjunction with external knowledge databases.

[0627] Step 5:

[0628] The server optimizes the identified topic, converts it to text format, and sends it to the terminal. During this process, it selects appropriate phrasing that takes the flow of the conversation into consideration.

[0629] Step 6:

[0630] The terminal transmits topics received from the server to the user using an audio output device. The user can then resume the conversation using the provided topics.

[0631] (Example 1)

[0632] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0633] In modern society, smooth and continuous conversation is a crucial element in building and maintaining relationships. However, a challenge arises when conversations break down due to a lack of topics, preventing intended communication from taking place. In particular, it is difficult to appropriately grasp the flow of the conversation while providing interesting topics, so there is a need to improve methods for generating topics to keep conversations going.

[0634] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0635] In this invention, the server includes means for acquiring voice information and analyzing acoustic information to identify the conversation situation, means for detecting the cessation of conversation, means for generating relevant topics using past dialogue history and user information, and means for providing the user with topics generated using a generative AI model. This enables the user to smoothly and naturally resume the conversation and continue the intended communication.

[0636] "Audio information" refers to acoustic signals, including the user's conversations and utterances.

[0637] "Acoustic information" refers to data such as frequency characteristics and volume used when analyzing audio information.

[0638] "Conversation context" refers to information that indicates the context, content, and progress of the current dialogue.

[0639] "Conversation interruption" refers to a state in which the dialogue between users is temporarily suspended.

[0640] "Dialogue history" refers to a record of conversations that a user has had in the past.

[0641] "User information" refers to personal information including data such as the user's interests, preferences, and past behavioral records.

[0642] "Related topics" refer to appropriate topics that are based on the user's interests and concerns, and that facilitate the smooth continuation of the conversation.

[0643] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate natural language text.

[0644] An "external information database" refers to an internet-based information source that provides the latest information and knowledge.

[0645] "Speech recognition" is a technology that analyzes speech information and converts it into text data.

[0646] This invention relates to a system that assists users in having smooth conversations. The system helps to continue a conversation by acquiring voice information and suggesting appropriate topics based on that information.

[0647] The user engages in conversation using an ear-worn device. This device records the user's conversation in real time using a built-in voice input device. This voice information is transmitted to a server via wireless communication technology.

[0648] The server uses speech recognition software (e.g., a general speech-to-text service) to convert the collected speech information into text. Next, natural language processing techniques are used to analyze this text information and identify the context and situation of the conversation. The user's dialogue history and personal profile are stored in a database, and based on this information, topics suitable for the user are generated.

[0649] The server uses a generative AI model to create new topics. During this process, it retrieves the latest information from an external database to verify the topic's relevance. This process selects topics that are likely to attract user interest.

[0650] The terminal takes topics provided by the server, converts them into speech using speech synthesis technology (e.g., a general speech conversion system), and presents them to the user. This series of operations allows the user to continue the conversation naturally and smoothly.

[0651] As a concrete example, consider a situation where a user is in a restaurant with a friend. If the conversation pauses, the device detects this and sends an audio message to the server. The server analyzes the content and suggests a movie topic related to the user's recent interests. This suggestion is provided to the user as an audio message such as, "Do you know any recent movies that are popular?", and the conversation resumes.

[0652] An example of a prompt for a generative AI model would be: "Please suggest a new conversation starter about a topic the user has recently been interested in. Please consider the context of the conversation the user is currently participating in." This allows the system to generate and suggest an appropriate topic to the user.

[0653] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0654] Step 1:

[0655] The user initiates a conversation using an ear-worn device. The terminal uses the voice input device within this device to record the user's conversation in real time. The input data is the user's voice, which is transmitted to the server via wireless communication technology. As output, the recorded voice data is transferred to the server.

[0656] Step 2:

[0657] The server receives audio data transmitted from the terminal. The input is the transmitted audio data. Speech recognition software is used to convert the audio data into text data. This data processing results in the content of the conversation being output as text.

[0658] Step 3:

[0659] The server analyzes the obtained text data using natural language processing techniques. The input is the converted text data. Through this data processing, the server identifies the situation and context of the conversation. The output is contextual information of the analyzed conversation, which makes the user's dialogue intent clearer.

[0660] Step 4:

[0661] The server references the user's user information and past conversation history based on the analysis results. The input is conversation context information, which is then processed by matching it with information in the database. This generates relevant topics that the user is likely to be interested in. The output is the generated topic ideas.

[0662] Step 5:

[0663] The server uses a generative AI model to give concrete form to topics. The input is a topic idea, and the AI ​​generates specific topics through data generation calculations. The output is a topic suggestion that can be presented to the user.

[0664] Step 6:

[0665] The server references an external information database to improve the timeliness and relevance of generated topics. Input consists of concretized topic proposals, which are augmented with the latest information obtained from the external database. Output consists of topics that are current and relevant.

[0666] Step 7:

[0667] The terminal receives a topic generated from the server and converts it into speech using speech synthesis technology. The input is the generated topic, and the output is the speech result. Specifically, it provides the topic as speech to the user's ears, thereby assisting in resuming the conversation.

[0668] (Application Example 1)

[0669] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0670] In modern society, providing appropriate topics of conversation is crucial for smooth communication within the family and socially. However, finding a suitable new topic when a conversation stalls is not easy, often leading to stagnation. Therefore, there is a need for support technologies to maintain a smooth flow of conversation.

[0671] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0672] In this invention, the server includes means for acquiring conversation information, means for analyzing acoustic data to identify the context of the conversation, means for detecting the end of a conversation, means for generating relevant topics using past conversation history and personal information, means for providing the generated topics to the user, and means for performing contextual analysis of the conversation using natural language processing technology, selecting the most appropriate topic based on profile information, and acquiring current trends from external information sources to improve the relevance of the topic. This enables the rapid detection of interruptions in conversation and the automatic provision of topics that match the user's interests and timing, thereby allowing for a natural and smooth continuation of conversation.

[0673] "Conversational information" refers to general data about human interaction obtained from audio and text data.

[0674] "Acoustic data" refers to a collection of signal data collected as speech, and is fundamental information for identifying the context of a conversation.

[0675] "Natural language processing technology" is a technology that uses computers to analyze, understand, and generate human language.

[0676] "Profile information" refers to attribute information about a user, such as their personal preferences, interests, and history, and is used when selecting appropriate topics.

[0677] "External information sources" refer to information providers accessible from outside the system, such as the internet and cloud services, and are the sources from which information on the latest trends and developments is obtained.

[0678] "Topic generation" is the process of proposing new topics to present to users based on acquired conversational information and the results of contextual analysis.

[0679] "Contextual analysis" is an analytical method that involves understanding the meaning of expressions and words used in a conversation and grasping the overall flow and situation of the conversation.

[0680] "Relevance" refers to the degree to which a generated topic fits the user's current conversation and interests.

[0681] This system will be implemented in a home-use conversational assistance robot. The robot will acquire voice data through an ear-worn IoT device and send that data to a server. The server will convert the voice data into text using speech recognition technology. The "speech_recognition" library will be used in this process. The converted text will be analyzed using natural language processing technology. This analysis will use a technology called "NLPProcessor" to understand the context of the conversation.

[0682] Based on the analysis results, the server generates appropriate topics using the user's profile information and past conversation history. During this process, it obtains current trends from external sources to verify topic suitability. The latest information is retrieved in real time from "external_database_api" and used. The generated topics are then sent back to the robot and provided to the user as speech using "text_to_speech" technology. In this speech generation process, the robot uses a speaker to allow the user to hear the speech, enabling smooth conversation continuation.

[0683] As a concrete example, consider a situation where a home robot encounters a lull in conversation between a visitor and a user. The robot detects this and suggests a new topic, such as, "Shall we talk about the latest trend in home gardening?" This topic is selected based on the user's interests and also takes into account the latest trends from external sources.

[0684] An example of a prompt for a generative AI model would be: "The current topic of conversation is 'home gardening'. Please suggest new related topics based on your user profile and current trends."

[0685] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0686] Step 1:

[0687] The terminal (robot) uses a voice input device to record conversations between the user and their surroundings in real time. The recorded voice data becomes input and is prepared to be sent to the server. The voice data is temporarily stored as a digital signal within the terminal.

[0688] Step 2:

[0689] The server converts the received audio data into text data using the "speech_recognition" library. The input for this step is audio data, and the output is text data. In the speech recognition process, an acoustic model and a grammatical model are used to ensure accurate text conversion.

[0690] Step 3:

[0691] The server analyzes the generated text data using a natural language processing technique called "NLPProcessor" to identify the context of the conversation. The input is text data, and contextual information is generated as output. Through this analysis, the server understands the keywords and sentence structure used, and grasps the purpose and content of the conversation.

[0692] Step 4:

[0693] The server matches user profile information and past conversation history to generate relevant topics based on context. The input is contextual information and the user profile, and the output is a new topic. An internal algorithm selects topics that match the user's interests.

[0694] Step 5:

[0695] The server uses "external_database_api" to retrieve the latest trends from external sources, improving the relevance of generated topics. The input is a hypothetical topic, and the output topic is generated incorporating external trend information. This operation enables the provision of topics that reflect social trends in real time.

[0696] Step 6:

[0697] The server finally sends the generated topic to the terminal. The input is the integrated topic, and the terminal receives and processes it as output. The terminal then waits, preparing for speech synthesis.

[0698] Step 7:

[0699] The device converts the received topic into speech using "text-to-speech" technology and delivers it to the user through the speaker. The input is the topic text, and the output is generated speech. The topic is conveyed with natural intonation using a speech synthesis engine.

[0700] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0701] The system of the present invention assists users in smoothly conducting natural conversations with others during a conversation. The system acquires conversational information through the user's ear-worn IoT device and includes an emotion engine that recognizes the user's emotions based on that data.

[0702] When a user engages in conversation, the device records audio data in real time and sends it to the server. The server uses speech recognition technology to convert this audio data into text data and analyzes the context of the conversation. Simultaneously, an emotion engine analyzes the user's emotional state based on factors such as tone, volume, and speed of voice.

[0703] The server generates relevant topics based on acquired sentiment and contextual information. These generated topics are optimized to reflect the user's current sentiment. It's also possible to enhance topic relevance by connecting to external knowledge databases and retrieving the latest topics.

[0704] The device notifies the user of these optimized topics via voice. This allows the user to resume the conversation without interruption and continue the dialogue in a natural flow. Furthermore, the topics offered are selected with the aim of maintaining the user's emotions in a positive state.

[0705] As a concrete example, consider a scenario where a user is chatting with a friend, but the conversation suddenly becomes a little subdued. The device detects this situation and sends audio data to the server. The server uses an emotion engine to analyze the user's slightly downcast mood and generates information about topics that the user might be interested in and that could brighten their mood, such as "interesting recent movies or hobbies." By providing this information to the user, they can steer the conversation back in a positive direction.

[0706] This system allows users to maintain the flow of conversation while also addressing their emotional needs.

[0707] The following describes the processing flow.

[0708] Step 1:

[0709] The user initiates a conversation using an ear-worn IoT device. The device has the capability to record this conversation in real time and begins acquiring acoustic data.

[0710] Step 2:

[0711] The device periodically sends the audio data being recorded to the server. This data includes features necessary for emotion recognition, such as the tone and speed of the voice.

[0712] Step 3:

[0713] The server converts the received audio data into text data using speech recognition. This makes the content of the conversation analyzable as text.

[0714] Step 4:

[0715] On the server, the emotion engine analyzes the acoustic data and evaluates the user's emotional state based on the characteristics of the voice. Based on the results of this analysis, it determines whether the current emotion is positive or negative.

[0716] Step 5:

[0717] The server performs a comprehensive analysis by combining contextual information from text data of conversations with sentiment analysis results. Based on past conversation history and new information from external knowledge databases, it generates topics optimized for the user.

[0718] Step 6:

[0719] The server sends the generated topic to the user's device at the time that best suits their mood. This topic is designed to facilitate the smooth continuation of the conversation.

[0720] Step 7:

[0721] The device provides the user with topics received from the server as audio output. The user can then introduce new topics based on the presented information and resume the conversation in a natural way.

[0722] (Example 2)

[0723] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0724] In modern communication, challenges exist such as interruptions in conversation and the inability to appropriately respond to the other person's emotions. In such situations, the flow of conversation stalls, making it difficult to maintain smooth interaction. Furthermore, the inability to offer topics relevant to the other person's feelings and situation can hinder deeper communication.

[0725] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0726] In this invention, the server includes means for acquiring conversational information and analyzing acoustic signals to identify the context of the conversation, means for analyzing emotional states based on the acoustic signals, and means for generating relevant topics using a generative artificial intelligence model. This enables the user to maintain the flow of the conversation and provide appropriate topics in accordance with the other person's emotions.

[0727] "Conversation information" refers to audio data related to user dialogue and the results of its analysis.

[0728] "Acoustic signals" refer to data collected electronically from the sounds emitted by the user.

[0729] "Context identification methods" refer to technologies that analyze the content and flow of a conversation based on acoustic signals to understand the current theme and topic.

[0730] "Methods for analyzing emotional states" refer to techniques that analyze tone, speed, and voice intensity from acoustic data to estimate the speaker's emotions and psychological state.

[0731] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information and ideas.

[0732] "Methods for generating topics" refer to the process of suggesting new conversation themes or topics based on the current context and emotional state.

[0733] "Means of providing information to users" refers to methods and technologies for communicating generated information and topics to users and presenting them in various forms.

[0734] "External knowledge sources" refer to external databases and information sources that contain the latest information.

[0735] "Speech recognition" is a technology that analyzes speech and automatically converts it into text data.

[0736] This invention relates to a system that assists users in natural conversation using an ear-worn information processing device. The system aims to maintain the flow of the user's conversation and provide appropriate topics.

[0737] The server receives acoustic signals transmitted from the user. The terminal has the capability to record acoustic data in real time and send it to the server. This data is used to analyze the content and tone of the conversation. The server uses speech recognition software to convert this acoustic data into text data. A general-purpose speech processing platform is used as the specific software for "speech recognition."

[0738] The analyzed data is used for contextual analysis and sentiment analysis. Sentiment analysis employs techniques that estimate emotional states from the tone and intensity of acoustic signals. These techniques utilize speech analysis technology as a means of analyzing emotional states.

[0739] Next, the server uses a generative artificial intelligence model to generate relevant topics based on the user's current emotional state and the flow of the conversation. In this process, the algorithm uses the generative AI model to select appropriate information to improve the user experience. As an example of a specific prompt for the generative AI model, the instruction "Suggest topics that will evoke positive emotions in the user" can be used.

[0740] Finally, the terminal converts the information provided by the server into speech and notifies the user. Text-to-speech technology is used here. This allows the user to continue the conversation based on the provided topic, enabling smooth dialogue.

[0741] This system provides an easy-to-use solution for uninterrupted, real-time communication. To achieve this, it comprehensively utilizes a variety of software and technologies.

[0742] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0743] Step 1:

[0744] The user initiates a conversation using an ear-worn information processing device. The device acquires acoustic signals in real time and inputs them into the terminal as audio data. This audio data is recorded for subsequent processing.

[0745] Step 2:

[0746] The terminal packets the recorded audio data and sends it to the server. A secure communication channel is used during data transmission to ensure data reliability and security. As a result, the server can receive the audio data.

[0747] Step 3:

[0748] The server receives the audio data as input and converts it into text data using a speech recognition engine. This conversion utilizes a specific speech recognition algorithm to identify the context within the audio data. The output is text data.

[0749] Step 4:

[0750] The server performs sentiment analysis based on text data. Here, it estimates the user's emotional state from characteristics such as tone, voice intensity, and speed of the acoustic signal. The sentiment analysis engine uses a specific algorithm to quantify the emotional state. The output is data indicating the emotional state.

[0751] Step 5:

[0752] The server uses a generative artificial intelligence model to generate topics based on the obtained text data and emotional state. The generative AI model utilizes prompts to identify appropriate topics that match the user's current emotions. The output is a list of suggested topics.

[0753] Step 6:

[0754] Upon receiving a topic generated by the server, the terminal uses its text-to-speech function to send an audio notification to the user. This notification allows the user to receive a topic to continue the conversation at the appropriate time. The output is information provided to the user as an audio prompt.

[0755] (Application Example 2)

[0756] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0757] In modern society, individual conversations are often interrupted, and their smooth progression, which addresses emotional needs, is hindered. In such cases, there is a need for effective support to help users maintain a natural flow of conversation and continue positive dialogue. However, conventional technologies have struggled to accurately grasp the emotional state of users and generate relevant topics based on that understanding.

[0758] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0759] In this invention, the server includes means for analyzing acoustic data to identify the context of a conversation, means for analyzing the user's emotional state from the audio data, and means for generating relevant topics based on the user's emotional state and the context of the conversation. This allows the user to maintain a natural flow of conversation while being offered positive topics that correspond to their emotional state.

[0760] "Means for acquiring conversational information" refers to devices or technologies for collecting voice data emitted by users.

[0761] "Means of analyzing acoustic data to identify the context of a conversation" refers to a process or technology that analyzes acquired audio data to identify the content and theme of a conversation.

[0762] "Means for detecting the end of a conversation" refers to a function or algorithm for determining that a conversation has been interrupted or ended.

[0763] "Methods for analyzing a user's emotional state from voice data" refers to technologies that evaluate a user's emotions based on factors such as tone, speed, and volume of their voice.

[0764] "Means for generating relevant topics based on the user's emotional state and the context of the conversation" refers to methods for creating new and relevant topics or subjects based on the results of the user's sentiment analysis and the content of the conversation.

[0765] "Means of providing generated topics to users" refers to a mechanism for communicating new topics or suggestions presented to users.

[0766] An "external knowledge database" is an external source of information or a system that stores the latest information and data.

[0767] "Methods for automatically converting conversational information into text data using speech recognition" refers to technologies that convert speech into text format, making it analyzable.

[0768] "Methods for generating prompt text using generative AI models" refer to methods that use artificial intelligence technology to generate prompts that output the most appropriate response or suggestion to the user.

[0769] The system for realizing this invention includes a process for collecting voice data, evaluating the user's emotional state, and generating relevant topics based on that evaluation. The system comprises, as its main components, a consumer robot equipped with a microphone, a server for processing the voice data, and a user interface for providing the generated topics to the user.

[0770] First, the consumer robot, acting as the terminal, collects the user's speech in real time through its built-in microphone. The collected speech data is sent to a server. The server uses the Google Cloud Speech-to-Text API to convert the speech data into text data using speech recognition technology. This converted text data is then used to analyze the content of the speech.

[0771] Next, the server uses TensorFlow to analyze the user's emotional state based on features such as tone, speed, and volume of speech. Based on the emotional information obtained in this way and the contextual information from speech recognition, it uses the OpenAI GPT model to generate relevant topics. This generative AI model provides optimal prompts that match the user's current emotional state and maintain the flow of the conversation.

[0772] The generated topics are communicated to the user through the robot's speaker. The user can continue the conversation in a natural flow and deepen their emotional connection.

[0773] For example, if a user is having a quiet conversation with their family at the dinner table, the robot could suggest, "It seems a new restaurant recently opened in the neighborhood. Why don't you all go and check it out sometime?" This could help to revitalize the conversation.

[0774] An example of a prompt message would be: "Suggest topics that the user might be interested in and that will help them maintain a positive mood. Current topics include travel, hobbies, and music."

[0775] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0776] Step 1:

[0777] The device collects the user's speech via a built-in microphone. The input is the user's voice, and the output is digital audio data. This audio data is transmitted to the server in real time.

[0778] Step 2:

[0779] The server converts the received audio data into text data using the Google Cloud Speech-to-Text API. The input is audio data, and the output is speech-based text data. The server then uses the converted text data to prepare for identifying the context of the conversation.

[0780] Step 3:

[0781] The server analyzes text data and uses TensorFlow to evaluate the user's emotional state. The input is text data, and the output is the evaluation of the user's emotional state. The server analyzes elements such as tone, speed, and intensity of sound to represent the emotional state numerically.

[0782] Step 4:

[0783] The server uses the OpenAI GPT model to generate relevant topics, taking into account the user's emotional state and the context of the conversation. The input is emotion rating data and context data, and the output is the newly proposed topic. The server optimizes the generated prompt sentences and prepares them for the user.

[0784] Step 5:

[0785] The terminal notifies the user of generated topics via voice. The input is a topic suggested by the server, and the output is the voice output to the user. The robot uses a configured speaker to provide a new topic that prompts the user to resume the conversation.

[0786] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0787] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0788] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0789] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0790] Figure 9 shows an emotion map 400 in 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 the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0791] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0792] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0793] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0794] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it 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 emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0795] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0796] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0797] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a 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 the input data.

[0798] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0799] Alternatively, 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 in response to a request from the data processing device 12.

[0800] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0801] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0802] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0803] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0804] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0805] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0806] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0807] The following is further disclosed regarding the embodiments described above.

[0808] (Claim 1)

[0809] Obtain conversation information,

[0810] A means of analyzing acoustic data to identify the context of a conversation,

[0811] A means for detecting the end of a conversation,

[0812] A means of generating relevant topics using past conversation history and personal information,

[0813] A means of providing the generated topics to users,

[0814] A system that includes this.

[0815] (Claim 2)

[0816] The system according to claim 1, comprising means for obtaining the latest information from an external knowledge database and using it for topic generation.

[0817] (Claim 3)

[0818] The system according to claim 1, comprising means for automatically converting conversational information into text data by speech recognition.

[0819] "Example 1"

[0820] (Claim 1)

[0821] Acquire audio information,

[0822] A means of analyzing acoustic information to identify the context of a conversation,

[0823] A means of detecting the cessation of conversation,

[0824] A means of generating relevant topics using past dialogue history and user information,

[0825] A means of providing users with topics generated using a generative AI model,

[0826] A system that includes this.

[0827] (Claim 2)

[0828] The system according to claim 1, further comprising means for obtaining the latest information from an external information database and using it for topic generation.

[0829] (Claim 3)

[0830] The system according to claim 1, comprising means for automatically converting speech information into text information by speech recognition.

[0831] "Application Example 1"

[0832] (Claim 1)

[0833] Means of obtaining conversational information,

[0834] A means of analyzing acoustic data to identify the context of a conversation,

[0835] A means for detecting the end of a conversation,

[0836] A means of generating relevant topics using past conversation history and personal information,

[0837] A means of providing the generated topics to users,

[0838] Using natural language processing techniques, the context of the conversation is analyzed, and the most appropriate topic is selected based on the profile information.

[0839] A means of obtaining current trends from external information sources to improve the relevance of the topic,

[0840] A system that includes this.

[0841] (Claim 2)

[0842] The system according to claim 1, comprising means for obtaining the latest information from an external knowledge database and using it for topic generation.

[0843] (Claim 3)

[0844] The system according to claim 1, comprising means for automatically converting conversational information into text data by speech recognition.

[0845] "Example 2 of combining an emotion engine"

[0846] (Claim 1)

[0847] Obtain conversation information,

[0848] A means of analyzing acoustic signals to identify the context of a conversation,

[0849] A method for analyzing emotional states based on acoustic signals,

[0850] A means of generating relevant topics using a generative artificial intelligence model,

[0851] A means of providing the generated topics to users,

[0852] A system that includes this.

[0853] (Claim 2)

[0854] The system according to claim 1, comprising means for obtaining information from an external knowledge source and using it for topic generation.

[0855] (Claim 3)

[0856] The system according to claim 1, comprising means for automatically converting conversational information into text data by speech recognition.

[0857] "Application example 2 when combining with an emotional engine"

[0858] (Claim 1)

[0859] Obtain conversation information,

[0860] A means of analyzing acoustic data to identify the context of a conversation,

[0861] A means for detecting the end of a conversation,

[0862] A method for analyzing the emotional state of a user from voice data,

[0863] A means for generating relevant topics based on the user's emotional state and the context of the conversation,

[0864] A means of providing the generated topics to users,

[0865] A system that includes this.

[0866] (Claim 2)

[0867] The system according to claim 1, comprising means for obtaining the latest information from an external knowledge database and using it for topic generation.

[0868] (Claim 3)

[0869] The system according to claim 1, comprising means for automatically converting conversational information into text data by speech recognition, and means for generating prompt sentences using a generation AI model. [Explanation of Symbols]

[0870] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means of obtaining conversational information, A means of analyzing acoustic data to identify the context of a conversation, A means for detecting the end of a conversation, A means of generating relevant topics using past conversation history and personal information, A means of providing the generated topics to users, Using natural language processing techniques, the context of the conversation is analyzed, and the most appropriate topic is selected based on the profile information. A means of obtaining current trends from external information sources to improve the relevance of the topic, A system that includes this.

2. The system according to claim 1, comprising means for obtaining the latest information from an external knowledge database and using it for topic generation.

3. The system according to claim 1, comprising means for automatically converting conversational information into text data by speech recognition.

Images