system

Abstract

We provide the system. [Solution] A means of acquiring voice for recording the voice of the parent, A voice learning method that analyzes the parent's voice and learns its characteristics as digital data, A voice generation means for generating arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature data, An output means for outputting the generated audio to a child, An input processing means for receiving input from a child as voice or text and generating a response, A data management system for saving and analyzing conversation logs, A system that includes this.

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, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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] With the increase in dual-income families, parents have concerns when leaving their children home alone, while children experience loneliness and anxiety due to being alone. There are also many families dissatisfied with the costs and environment of after-school care, and a cost-effective solution is required.

Means for Solving the Problems

[0005] The present invention provides a system that records the voice of a parent, learns its characteristics, and uses a generative AI that reproduces the parent's voice in conversations with children. By providing an audio acquisition means, an audio learning means, an audio generation means, an input processing means, an output means, and a data management means, an experience as if the child is talking to the parent is provided, reducing the anxiety of the parent and the loneliness of the child.

[0006] "Voice acquisition means" refers to a device or function for recording or documenting the voice of a parent.

[0007] A "speech learning method" is a device or function that analyzes acquired speech and learns its characteristics as digital data.

[0008] "Voice generation means" refers to a device or function that generates arbitrary words by imitating the voice of a parent based on learned voice data.

[0009] "Output means" refers to a device or function that plays the generated audio towards a child.

[0010] "Input processing means" refers to a device or function for receiving voice or text input from a child and generating a response based on it.

[0011] "Data management means" refers to a device or function for saving and analyzing conversation logs with children.

[0012] "Voice recognition means" refers to a technology or function for converting a child's voice input into text data.

[0013] "Emotional analysis means" refers to a technology or function that analyzes emotions in order for the voice generated during a response to mimic the emotions of the parent. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This 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 Example 2 when an 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 an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described according to the accompanying drawings.

[0016] First, the language used in the following description will be explained.

[0017] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0018] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0019] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0020] 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).

[0021] 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."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, 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.

[0032] 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.

[0033] 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.

[0034] 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".

[0035] This invention is a system aimed at reducing feelings of isolation and alleviating parental anxiety in dual-income households where children are home alone. This system uses a generative AI to reproduce the parent's voice, enabling natural conversation with the child. Specific embodiments of this invention are described below.

[0036] The user (parent) uses a dedicated application to record their voice and upload it to the server. This voice data is analyzed on the server to learn the characteristics of the parent's voice. The learning process includes parameters such as the parent's voice pitch, speed, and intonation, and these characteristics are integrated into the voice model.

[0037] The device receives voice or text input from the child. The received voice is converted into text data on the server using speech recognition technology. Based on this text data, the server generates an appropriate response and creates that response as a voice that resembles the parent's voice.

[0038] The generated audio is played from the device to the child, enabling a real-time conversational experience. The child can speak further through the device, and each time a new response is generated, sustaining a natural conversation.

[0039] As a concrete example of this system, imagine a scenario where a child returns home from school and the device speaks to them in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, saying, "That sounds interesting, what did you make?" and plays it back from the device.

[0040] Furthermore, the server records all conversations and performs periodic analysis. Parents can later review these conversation logs and use the information to improve future communication. In this way, the present invention is a system that technically complements parent-child conversations even in the absence of parents, providing a sense of security and connection.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The user (parent) launches the app and records their voice. The recorded audio data is uploaded directly from the device to the server.

[0044] Step 2:

[0045] The server analyzes the received audio data and extracts characteristics of the parent's voice. This includes information such as voice pitch, speed, intonation, and word patterns, and uses this data to train an AI model.

[0046] Step 3:

[0047] The device receives information entered by the child via voice or text. In the case of voice input, it uses speech recognition technology to convert it into text data.

[0048] Step 4:

[0049] The server performs natural language processing on the received text data to generate appropriate responses to questions and conversations. These responses are designed to mimic the tone and phrasing of the parent.

[0050] Step 5:

[0051] Based on the generated text response, the server uses a pre-trained speech model to generate speech data that reproduces the parent's voice.

[0052] Step 6:

[0053] The device plays the generated audio data to the child in real time. This allows the child to have an experience similar to interacting with their parent.

[0054] Step 7:

[0055] If the conversation continues, return to step 3 and receive new input from the child. This maintains a continuous and natural flow of dialogue.

[0056] Step 8:

[0057] The server logs all conversations and analyzes them later. This helps identify frequently occurring phrases and emotional states in conversations, and the results are fed back to parents for review.

[0058] (Example 1)

[0059] 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."

[0060] In dual-income households, children often feel isolated when their parents are absent, and parents themselves feel anxious because they are unaware of their children's situation. There was a need for technological solutions to compensate for parental absence, alleviate children's feelings of isolation, and provide parents with a sense of security.

[0061] 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.

[0062] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing and learning the characteristics of the parent's voice, voice generation means for generating arbitrary responses that resemble the parent's voice using a generative AI model, and means for outputting the voice to the child. This makes it possible to have natural conversations with the child using the parent's voice even when the parent is absent, thereby alleviating the child's sense of isolation and providing the parent with a sense of security.

[0063] "Voice acquisition means" refers to a device or function for recording the voice of a parent.

[0064] "Voice learning methods" refer to a function that analyzes the characteristics of a parent's voice, such as pitch, speed, and intonation, and learns them as digital information.

[0065] "Voice generation means" refers to a function that uses a generative AI model to generate arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature information.

[0066] "Means of reproducing sound" refers to functions including speakers and audio playback devices for outputting generated sound to children.

[0067] The "input processing means" is a function that receives input from a child, either as voice or text, and generates an appropriate response.

[0068] "Voice recognition means" refers to a function that converts a child's voice input into text data and generates a response.

[0069] A "conversation management system" is a data management function that records all conversations to enable future analysis.

[0070] "Emotional analysis means" refers to a function that performs emotional analysis on the generated voice to mimic the parent's emotional state and provide a more natural response.

[0071] A "conversation recording device" is a function that records the process by which generated responses are output to the child in real time, and uses this information for future analysis.

[0072] This invention is a system designed to alleviate feelings of isolation in children and reduce parental anxiety during the absence of parents in dual-income households. The system aims to enable natural conversations with children by faithfully reproducing the parent's voice using a generative AI model. Specific embodiments of this invention are described below.

[0073] First, the user (parent) records their own voice using a dedicated application. A device with a microphone is used for this recording. The recorded audio data is sent to a server. The server uses voice analysis software to extract features such as pitch, speed, and intonation from the audio data. Based on this data, a generative AI model is trained to create a voice generation model that reproduces the parent's voice.

[0074] The device receives voice and text input from the child. The received voice data is sent to a server. On the server, speech recognition technology is used to convert the voice into text. Based on this text, an appropriate response is generated by a generation AI. This response is synthesized to sound like the parent's voice.

[0075] This generated audio data is sent to the device and played back to the child. This allows the child to have a conversational experience as if their parent were present.

[0076] As a concrete example, imagine a scenario where, upon a child returning home from school, the device automatically speaks in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, such as, "That sounds interesting, what did you make?", and plays it back from the device. This enables natural, real-time conversation.

[0077] As a concrete example of a prompt, an instruction such as "Ask the child about their thoughts today in a parent's voice, and generate a response that mimics the parent's voice" can be input into a generative AI model, thereby facilitating natural dialogue between parent and child.

[0078] In this way, the present invention is a system that uses technical means to compensate for the absence of parents and maintain the bond between parents and children.

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

[0080] Step 1:

[0081] The user (parent) records their voice using a dedicated application. Here, a device with a microphone is used to record the parent's speech as a digital audio file. This audio data serves as input data for later analysis of the parent's voice characteristics. The completed audio recording is temporarily stored for use in the next step.

[0082] Step 2:

[0083] The user uploads recorded audio data to the server. Based on this uploaded audio data, the server uses speech analysis software to extract features such as pitch, speed, and intonation. These extracted features are used as training data for a speech generation AI model. This allows the server to create a speech generation model that reproduces the parent's voice. The output is digital data that retains the features of the parent's voice.

[0084] Step 3:

[0085] The device receives voice or text input from the child. Here, the child inputs voice or text through the device's microphone or keyboard. The input is in real time, and the device temporarily stores this data before sending it to the server. In the case of voice input, the input data is directly processed for speech recognition in the next step.

[0086] Step 4:

[0087] The server processes the audio data sent by the child using speech recognition software and converts it into text data. This text data is then used as new input for response generation. Preprocessing, such as noise reduction and speech intensity normalization, is performed. The output is the audio converted to text.

[0088] Step 5:

[0089] The server generates an appropriate response based on the text data. It uses a generative AI model to analyze the input text and create a response based on the prompt. This response is synthesized as speech using parental voice features. The output of this step is the generated response as speech data.

[0090] Step 6:

[0091] The server sends the generated audio data to the device. The device plays the received audio through its speaker and delivers it to the child. The played audio is very similar to the parent's voice, providing the child with a natural conversational experience.

[0092] Step 7:

[0093] The server records the entire conversation and saves it to a database for analysis. The data saved as a conversation log can be reviewed later by parents to help improve parent-child communication. The output is data recording the content of 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 dual-income households, a key challenge is how to enhance the safety and emotional security of children when they are home alone. It is necessary to alleviate children's feelings of isolation while their parents are away, while also providing emergency preparedness information and safety tips to reduce parental anxiety.

[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 voice acquisition means for recording the parent's voice, voice learning means for analyzing the parent's voice and learning its characteristics as digital data, voice generation means for generating arbitrary responses that resemble the parent's voice based on the recorded voice and learned characteristic data, and information provision means for providing safety information and emergency response methods. This makes it possible to provide safety information and enhance a sense of security by engaging in natural conversation with the child on behalf of the parent.

[0099] "Voice acquisition means" refers to devices or methods for recording the voices of parents.

[0100] "Voice learning methods" refer to devices or methods that analyze the acquired voice of a parent and learn its characteristics as digital data.

[0101] "Voice generation means" refers to a device or method that generates an arbitrary response in a manner resembling the voice of a parent, based on recorded voice and learned feature data.

[0102] "Output means" refers to a device or method for outputting the generated sound to a child.

[0103] "Input processing means" refers to a device or method that receives input from a child as voice or text and generates a response.

[0104] "Data management means" refers to devices or methods for saving and analyzing conversation logs.

[0105] "Information provision means" refers to devices or methods for providing children with safety information and emergency response procedures.

[0106] This system provides a sense of security to children even when the parents are absent by acquiring the parents' voices, generating responses based on them, and providing them to the children. Specifically, voice acquisition is performed using the user's (parent's) smart device. Using this device, the parent records their voice and uploads it to the server. The server uses a cloud service as a voice learning tool, analyzing the voice data to learn the parents' voice characteristics. Google's Cloud Speech-to-Text API is used for this learning process.

[0107] The speech generation method uses OpenAI's (registered trademark) speech synthesis API to generate responses that resemble the parent's voice based on learned feature data. The server is always running and quickly generates and outputs responses to requests from the terminal. This generated voice is delivered to the child through a dedicated application running on the terminal.

[0108] The terminal transmits received audio and text to the server as input processing. Furthermore, it can provide safety information and emergency response measures through information provision mechanisms. This is not only a simulation of everyday communication, but also a crucial function for protecting children's safety.

[0109] In this way, the device speaks to the child on behalf of the parent, and the conversation logs are saved to a cloud service using data management tools for analysis. Parents can later review these logs to improve family communication and security. For example, when the child returns home, the device asks in the parent's voice, "What did you do at school today?", and the generated voice responds with the child's answer, "It was fun, let's do our best tomorrow too." An example of a prompt to support such a conversation is, "Please respond in a way that shows interest as a parent in what your child has said."

[0110] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0111] Step 1:

[0112] Users use a dedicated application to record their voice via their smart device and upload this data to a server. The input is the user's voice, and the output is an audio file stored on cloud storage. The recorded audio is sent to the server by an audio acquisition method and saved as an audio file.

[0113] Step 2:

[0114] The server analyzes uploaded audio data using speech learning methods. The input is an audio file, and internally, the Google Cloud Speech-to-Text API is used to extract digital feature data such as pitch, speed, and intonation. This digital feature data is integrated into a generative AI model and stored as training data. The output is a data model that has learned the parent's speech features.

[0115] Step 3:

[0116] The device receives voice or text input from the child. The input is either the child's voice or text; in the case of voice input, it is converted to text using speech recognition technology. The converted text data is output and sent to the server.

[0117] Step 4:

[0118] The server generates a response based on the received text data. It uses prompts to instruct the AI ​​model, which then plans an appropriate response. The input is text data based on the child's speech, and the output is the text data to be used as the response. An example of a prompt is, "As a parent, please respond to what your child has said with interest."

[0119] Step 5:

[0120] The server uses speech generation technology to convert the planned text data into speech that resembles the parent's voice. Using the OpenAI speech synthesis API, it applies speech feature data to the input text to generate a response in the parent's voice. The output is the generated speech data.

[0121] Step 6:

[0122] The generated audio data is sent to the device and played back to the child through the device's speaker as the output. The input is audio data, and the output is the child's response via the audio interface.

[0123] Step 7:

[0124] The server stores all conversation logs using data management tools and performs periodic analysis. The input is conversation data recorded via communication, which is then analyzed to extract key points. This allows parents to review the conversation logs later and use the information to improve safety and communication. The output is the data resulting from the analysis.

[0125] 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.

[0126] This invention is a system that can recognize the emotions of the user (parent) and generate an optimal response accordingly. By considering the parent's emotions, this system improves the quality of responses, making conversations with children more natural and humane.

[0127] The user (parent) records voice using a dedicated app and uploads the voice data to the server from their device. The server receives this data, analyzes it, and extracts voice characteristics. Furthermore, it uses an emotion engine to recognize the parent's emotional state from this voice. Emotion recognition includes a process of classifying emotions into categories such as happiness, anger, and surprise.

[0128] When a child inputs information via voice or text through the device, the device sends it to the server. The server generates an appropriate response based on the input, taking into account the parent's emotional state. The emotion engine detects emotional fluctuations in real time and adjusts the response accordingly. For example, if the parent is relaxed, the response will be in a calm tone; if the parent is excited, it will be in an energetic tone.

[0129] The generated responses are provided to the child as speech synthesized to mimic the characteristics of the parent's voice. This allows the child to enjoy conversations that reflect the parent's emotions. As the interaction with the child progresses, the server manages all conversations and associated emotional data as logs, which the user can review later.

[0130] For example, suppose a child inputs, "I had a bad day at school today." If the system recognizes that the parent's current emotional state is calm and that encouragement is needed, it will generate an encouraging response such as, "That must have been tough. But it's okay, tomorrow will surely be a better day," and play it back in the parent's voice.

[0131] Thus, the present invention functions as a system that enables empathetic dialogue with children by generating responses that include the parent's emotions.

[0132] The following describes the processing flow.

[0133] Step 1:

[0134] The user (parent) launches a dedicated app and records audio. The recorded audio data is then sent directly to the server by the device.

[0135] Step 2:

[0136] The server analyzes the received audio data and extracts voice characteristics. In this process, it uses an emotion engine to recognize the emotions contained in the audio. The emotion engine infers emotions from the tone, speed, and intonation of the voice and records them as data.

[0137] Step 3:

[0138] When a child inputs voice or text into the device, the device forwards this input to a server. The server converts this input into text format and performs natural language processing.

[0139] Step 4:

[0140] The server generates an appropriate response to the child's input. When generating the response, it takes into account the parent's emotional data, which has been analyzed in advance, and sets the message tone to match the parent's current emotional state.

[0141] Step 5:

[0142] The server performs speech synthesis based on the generated response and creates speech data that resembles the parent's voice using pre-learned characteristics of the parent's voice.

[0143] Step 6:

[0144] The device plays this generated audio back to the child in real time. This allows the child to experience a conversation that reflects the parent's emotions.

[0145] Step 7:

[0146] If the conversation continues, return to step 3 and repeat the same process with the new input.

[0147] Step 8:

[0148] The server logs all conversation data and emotional states. This saved data can be reviewed later by parents to help improve communication with their children.

[0149] (Example 2)

[0150] 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".

[0151] In conversations with children, mechanical responses that disregard the parent's emotional state are often produced, leading to a problem where children do not feel adequately cared for. In particular, when a parent's emotions do not translate into appropriate responses to the child's needs or situation, parent-child communication may become weak. Therefore, the challenge is to achieve natural dialogue that takes the parent's emotions into consideration.

[0152] 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.

[0153] In this invention, the server includes emotion recognition means for analyzing the parent's emotional state and reflecting it in the response, speech synthesis means for generating arbitrary responses that resemble the parent's speech characteristics based on recorded voice data and learned feature data, and data management means for storing conversation history and related data for later analysis. This makes it possible to provide children with personalized and interactive responses that take the parent's emotions into consideration.

[0154] "Data acquisition means" refers to a device or algorithm for acquiring and recording the voice of a parent.

[0155] "Analysis means" refers to a device or program for extracting and learning speech features in digital format from acquired speech data.

[0156] A "speech synthesis means" is a device or algorithm for generating arbitrary responses that resemble the speech features of the parent, based on learned feature data.

[0157] "Output control means" refers to a device or program for outputting generated audio to a child under selected conditions.

[0158] "Input processing means" refers to a device or program for receiving inquiries from children in voice or text format and generating appropriate responses.

[0159] An "emotion recognition tool" is a device or algorithm that analyzes a parent's emotional state and uses the results to inform the response.

[0160] "Data management means" refers to a device or program for properly storing conversation history and related data so that it can be analyzed later.

[0161] "Speech recognition means" refers to a device or software for converting speech input into text data.

[0162] "Emotional analysis means" refers to a device or program that analyzes emotional states in real time and adjusts responses so that the generated voice can mimic the emotional state of the parent.

[0163] This system takes into account the parent's emotional state and is configured to generate natural and humane responses in interactions with children. Its embodiments are described in detail below.

[0164] The user (parent) records their voice using a dedicated software application. This recorded voice is processed as hardware by a mobile device such as a smartphone or tablet and uploaded to a server via a data acquisition method. Secure communication protocols such as HTTPS are used to ensure the security of the communication.

[0165] The server uses audio signal processing libraries such as "OpenSMILE" to extract audio features such as pitch, energy, and periodicity from the acquired audio data. Next, it uses emotion recognition algorithms such as "DeepMoji" to analyze the emotional state contained in the audio data. This information allows the system to understand the parent's current emotions in real time.

[0166] The child inputs their situation and emotions into the device via voice or text. The device receives this input and sends it to a server running a natural language generation model such as "GPT-3(registered trademark)" to generate an appropriate response tailored to the parent's emotional state. In this process, the prompt is constructed for the generating AI model. For example, a prompt might be set to, "How would you respond if the parent is feeling happy and the child says they have made a new friend?"

[0167] The generated responses are converted into speech using speech synthesis technology such as "Google Text-to-Speech." By mimicking the parent's voice characteristics, the child can experience a conversation that feels as if they are actually talking to their parent. The device then plays the synthesized speech, providing the child with responses that reflect the parent's emotions.

[0168] Furthermore, the server saves all conversation data and associated sentiment data as logs. This allows users to review this data later and use it to reflect on and improve their communication with their children.

[0169] This system aims to facilitate smooth communication between parents and children by generating personalized and interactive responses that take into account the parent's feelings.

[0170] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0171] Step 1:

[0172] The user (parent) records their voice using a dedicated application. At this stage, the input is the parent's voice, and the output is a digital audio file. The recorded audio file is stored on the device and sent to the server by a data acquisition means. Specifically, the device uploads the file to the server using a secure protocol (e.g., HTTPS) via Wi-Fi or mobile data communication.

[0173] Step 2:

[0174] The server uses an audio signal processing library such as "OpenSMILE" to analyze the received audio data. The input is the previously uploaded audio file, and the output is extracted audio feature data (pitch, energy, periodicity, etc.). Specifically, the server passes the audio data to an analysis algorithm, generates feature vectors, captures changes in each time frame, and records them as features.

[0175] Step 3:

[0176] The server uses emotion recognition algorithms such as "DeepMoji" to determine the parent's emotional state based on voice feature data. The input is voice feature data, and the output is a specific emotion category (e.g., happiness, anger, surprise). In this process, the server inputs the feature data into a neural network model, calculates the probability of each emotion category, and determines the most likely emotional state.

[0177] Step 4:

[0178] Children input voice or text through a dedicated terminal application. The input is the text or voice entered by the child, and the output is text data sent to the server. Specifically, the child's voice is converted to text by the speech recognition function on the terminal and sent to the server as input.

[0179] Step 5:

[0180] The server generates a response using a generative AI model such as "GPT-3" based on the child's text input it receives. In this case, the input is the child's text and the parent's emotional state, and the output is the generated response sentence. Specifically, the server provides the input sentence as a prompt sentence to the generative AI model and obtains a text response that takes the parent's emotional state into account. For example, a prompt such as "How would you respond if the parent's emotional state is calm and the child says they have made a new friend?" might be used.

[0181] Step 6:

[0182] The server converts the generated response into speech using a text-to-speech engine (e.g., "Google Text-to-Speech"). The input is the generated response sentence, and the output is audio data. Specifically, the synthesized speech is parameterized to mimic the parent's voice and encoded as an audio file.

[0183] Step 7:

[0184] The device receives the synthesized speech and plays it back to the child through the speaker. At this stage, the input is the audio data received from the server, and the output is the speech as physical sound. Specifically, the device loads the audio file into a buffer and plays it back through the speaker at a normal volume using playback software.

[0185] Step 8:

[0186] The server records all parent-child dialogue data and associated emotional information in a database, retaining it for later user access. The input is all data generated during the conversation, and the output is a structured log dataset. Specifically, it stores timestamps, text, voice features, emotional states, etc., for each dialogue session as database records, providing a basis for future analysis.

[0187] (Application Example 2)

[0188] 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".

[0189] It is difficult for parents and store staff to automatically generate responses that accurately reflect their emotional state when interacting with children or customers, making it challenging to maintain natural and human-centered communication. Therefore, there is a need for improved communication in close relationships and enhanced customer satisfaction in store services. Furthermore, there is a demand for systems that can support appropriate responses based on the staff's emotional state during customer service.

[0190] 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.

[0191] In this invention, the server includes voice acquisition means for recording the voice of a parent or staff member; voice learning means for analyzing the voice of a parent or staff member and learning its characteristics as digital data; and voice generation means for generating appropriate responses that resemble the voice of a parent or staff member based on the recorded voice and learned characteristic data. This makes it possible to provide a system that enables appropriate responses according to the emotional state of the parent or staff member, and facilitates natural and humane communication with children and customers.

[0192] "Voice acquisition means" refers to devices or technologies for recording the voices of parents or staff.

[0193] "Speech learning methods" refer to technologies that analyze acquired speech and learn its characteristics as digital data.

[0194] "Voice generation means" refers to a technology that generates appropriate responses in a manner similar to the voice of a parent or staff member, based on recorded voice and learned feature data.

[0195] "Output means" refers to a device or method for providing the generated audio to a child or customer.

[0196] "Input processing means" refers to technology that receives input from a child or customer as voice or text and generates a response.

[0197] "Data management methods" refer to technologies for saving and analyzing conversation logs.

[0198] "Emotional adjustment techniques" refer to methods that suggest appropriate responses during customer service based on the emotional state of the staff member.

[0199] "Speech recognition means" refers to technology for converting speech into text data.

[0200] "Emotional analysis methods" are methods of analyzing emotions so that the generated voice can mimic the emotional state of a parent or staff member.

[0201] The system for implementing this invention is initiated when a parent or staff member records speech using a speech acquisition means and uploads the speech data to a server via a terminal. The server analyzes the speech using a speech learning means and extracts its features as digital data. In this process, the pitch, tone, and speed of the speech are taken into consideration.

[0202] The server uses emotion adjustment mechanisms to recognize the emotional state of staff and generate real-time, adjustable responses. At this time, a generative AI model is utilized to process the audio or text received by the input processing mechanism from the child or customer. The voice generation mechanism then mimics the voices of parents or staff to create the optimal response. The output mechanism provides the generated audio to the child or customer, enabling natural and human-like dialogue.

[0203] For example, if a user is wearing smart glasses, the system uses voice recognition to transcribe customer requests into text and suggests products based on that content. The server uses data management to save conversation logs and organizes the analytical data for later access by the user.

[0204] A concrete example of a prompt would be, "Think of the best way to interact with a customer in a relaxed manner. For example, how can you make a friendly suggestion when recommending a new product?" This prompt is used to provide guidance for the generative AI model to generate responses that improve the quality of customer service.

[0205] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0206] Step 1:

[0207] The user records the voice of a parent or staff member using an audio acquisition device and uploads the audio data to the server via the terminal. The input is audio data, and the output is digital audio data stored on the server. The server performs format conversion to prepare this data.

[0208] Step 2:

[0209] The server uses speech learning methods to analyze uploaded audio data and extract speech features. Here, it generates speech feature vectors using Mel-frequency cepstrum coefficients (MFCCs) and speech pitch, among other things. The input is digital audio data, and the output is a speech feature vector.

[0210] Step 3:

[0211] The server uses emotion adjustment mechanisms to analyze emotional states from speech feature vectors. In this process, it determines an emotional category based on each feature, classifying them into categories such as relaxation or stress. The input is a speech feature vector, and the output is emotional state information.

[0212] Step 4:

[0213] The terminal receives input from a child or customer. The input processing device converts the speech into text data. The input is customer speech data, and the output is text data. The terminal uses speech recognition software to perform this conversion.

[0214] Step 5:

[0215] The server uses a generative AI model to generate the optimal response from emotional state information and text data. The generated response is output as text containing expressions appropriate to the emotion. The AI ​​model used here is a pre-trained natural language generation model.

[0216] Step 6:

[0217] The server uses speech generation means to convert the generated text responses into speech. In this process, speech synthesis technology is used to mimic the voices of parents or staff, and to generate speech with emotionally appropriate tones. The input is text responses, and the output is synthesized speech data.

[0218] Step 7:

[0219] The device uses an output mechanism to provide the generated synthesized voice to the child or customer. Through this voice, the user can experience natural and human-like dialogue. The output is delivered via a speaker or headphones.

[0220] This series of steps enables the system to engage in natural conversations that respond to the emotions of parents and staff, thereby improving efficiency in real-world customer service situations.

[0221] 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.

[0222] Data generation model 58 is a 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.

[0223] 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.

[0224] [Second Embodiment]

[0225] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0226] 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.

[0227] 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).

[0228] 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.

[0229] 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.

[0230] 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).

[0231] 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.

[0232] 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.

[0233] 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.

[0234] 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.

[0235] 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.

[0236] 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".

[0237] This invention is a system aimed at reducing feelings of isolation and alleviating parental anxiety in dual-income households where children are home alone. This system uses a generative AI to reproduce the parent's voice, enabling natural conversation with the child. Specific embodiments of this invention are described below.

[0238] The user (parent) uses a dedicated application to record their voice and upload it to the server. This voice data is analyzed on the server to learn the characteristics of the parent's voice. The learning process includes parameters such as the parent's voice pitch, speed, and intonation, and these characteristics are integrated into the voice model.

[0239] The device receives voice or text input from the child. The received voice is converted into text data on the server using speech recognition technology. Based on this text data, the server generates an appropriate response and creates that response as a voice that resembles the parent's voice.

[0240] The generated audio is played from the device to the child, enabling a real-time conversational experience. The child can speak further through the device, and each time a new response is generated, sustaining a natural conversation.

[0241] As a concrete example of this system, imagine a scenario where a child returns home from school and the device speaks to them in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, saying, "That sounds interesting, what did you make?" and plays it back from the device.

[0242] Furthermore, the server records all conversations and performs periodic analysis. Parents can later review these conversation logs and use the information to improve future communication. In this way, the present invention is a system that technically complements parent-child conversations even in the absence of parents, providing a sense of security and connection.

[0243] The following describes the processing flow.

[0244] Step 1:

[0245] The user (parent) launches the app and records their voice. The recorded audio data is uploaded directly from the device to the server.

[0246] Step 2:

[0247] The server analyzes the received audio data and extracts characteristics of the parent's voice. This includes information such as voice pitch, speed, intonation, and word patterns, and uses this data to train an AI model.

[0248] Step 3:

[0249] The device receives information entered by the child via voice or text. In the case of voice input, it uses speech recognition technology to convert it into text data.

[0250] Step 4:

[0251] The server performs natural language processing on the received text data to generate appropriate responses to questions and conversations. These responses are designed to mimic the tone and phrasing of the parent.

[0252] Step 5:

[0253] Based on the generated text response, the server uses a pre-trained speech model to generate speech data that reproduces the parent's voice.

[0254] Step 6:

[0255] The device plays the generated audio data to the child in real time. This allows the child to have an experience similar to interacting with their parent.

[0256] Step 7:

[0257] If the conversation continues, return to step 3 and receive new input from the child. This maintains a continuous and natural flow of dialogue.

[0258] Step 8:

[0259] The server logs all conversations and analyzes them later. This helps identify frequently occurring phrases and emotional states in conversations, and the results are fed back to parents for review.

[0260] (Example 1)

[0261] 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."

[0262] In dual-income households, children often feel isolated when their parents are absent, and parents themselves feel anxious because they are unaware of their children's situation. There was a need for technological solutions to compensate for parental absence, alleviate children's feelings of isolation, and provide parents with a sense of security.

[0263] 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.

[0264] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing and learning the characteristics of the parent's voice, voice generation means for generating arbitrary responses that resemble the parent's voice using a generative AI model, and means for outputting the voice to the child. This makes it possible to have natural conversations with the child using the parent's voice even when the parent is absent, thereby alleviating the child's sense of isolation and providing the parent with a sense of security.

[0265] "Voice acquisition means" refers to a device or function for recording the voice of a parent.

[0266] "Voice learning methods" refer to a function that analyzes the characteristics of a parent's voice, such as pitch, speed, and intonation, and learns them as digital information.

[0267] "Voice generation means" refers to a function that uses a generative AI model to generate arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature information.

[0268] "Means of reproducing sound" refers to functions including speakers and audio playback devices for outputting generated sound to children.

[0269] The "input processing means" is a function that receives input from a child, either as voice or text, and generates an appropriate response.

[0270] "Voice recognition means" refers to a function that converts a child's voice input into text data and generates a response.

[0271] A "conversation management system" is a data management function that records all conversations to enable future analysis.

[0272] "Emotional analysis means" refers to a function that performs emotional analysis on the generated voice to mimic the parent's emotional state and provide a more natural response.

[0273] A "conversation recording device" is a function that records the process by which generated responses are output to the child in real time, and uses this information for future analysis.

[0274] This invention is a system designed to alleviate feelings of isolation in children and reduce parental anxiety during the absence of parents in dual-income households. The system aims to enable natural conversations with children by faithfully reproducing the parent's voice using a generative AI model. Specific embodiments of this invention are described below.

[0275] First, the user (parent) records their own voice using a dedicated application. A device with a microphone is used for this recording. The recorded audio data is sent to a server. The server uses voice analysis software to extract features such as pitch, speed, and intonation from the audio data. Based on this data, a generative AI model is trained to create a voice generation model that reproduces the parent's voice.

[0276] The device receives voice and text input from the child. The received voice data is sent to a server. On the server, speech recognition technology is used to convert the voice into text. Based on this text, an appropriate response is generated by a generation AI. This response is synthesized to sound like the parent's voice.

[0277] This generated audio data is sent to the device and played back to the child. This allows the child to have a conversational experience as if their parent were present.

[0278] As a specific example, there is a scenario where when a child returns home from school, the terminal automatically speaks to the child in the voice of the parent, asking "What did you do today?" When the child answers, "I did art today," the server generates a response in the voice of the parent, such as "That sounds interesting. What did you make?" and plays it back from the terminal. This enables a natural real-time conversation to be realized.

[0279] As a specific example of the prompt sentence, an instruction such as "Ask the child about today's feelings in the voice of the parent and generate a response similar to the voice of the parent" is input into the generative AI model, which promotes natural conversation between parents and children.

[0280] In this way, the present invention is a system that uses technical means to complement the absence of the parent and maintain the bond between parents and children.

[0281] The flow of the specific process in Example 1 will be described using FIG. 11.

[0282] Step 1:

[0283] The user (parent) uses a dedicated application to record their voice. Here, a device with a microphone is utilized to record the voice of the parent speaking as a digital audio file. This voice data serves as input data for later analysis of the characteristics of the parent's voice. The recorded voice data is temporarily stored for use in the next step.

[0284] Step 2:

[0285] The user uploads the recorded voice data to the server. Based on this uploaded voice data, the server uses voice analysis software to extract features such as the pitch, speed, and intonation of the voice. The extracted features are used as training data for the generative AI model. Thereby, the server creates a voice generation model that reproduces the voice of the parent. The output is digital data that retains the characteristics of the parent's voice.

[0286] Step 3:

[0287] The terminal receives voice or text input from the child. Here, the child inputs voice or text through the microphone or keyboard of the terminal. The input is performed in real time, and the terminal temporarily stores this data to send it to the server. In the case of voice, the input data is directly subjected to speech recognition in the next step.

[0288] Step 4:

[0289] The server processes the voice data sent from the child with speech recognition software and converts it into text data. The text-converted data is used as new input for response generation. Here, preprocessing such as noise removal and normalization of voice intensity is performed. The output is data that has converted voice into text.

[0290] Step 5:

[0291] The server generates an appropriate response based on the text data. It analyzes the input text using a generation AI model and creates a response based on the prompt sentence. This response is synthesized as voice using the characteristics of the parent's voice. The output of this step is the response generated as voice data.

[0292] Step 6:

[0293] The server sends the generated voice data to the terminal. The terminal plays the received voice through the speaker and delivers it to the child. The played voice is very similar to the parent's voice and provides a natural conversation experience for the child.

[0294] Step 7:

[0295] The server records the entire conversation and saves it to a database for analysis. The data saved as a conversation log can be reviewed later by parents to help improve parent-child communication. The output is data recording the content of the conversation.

[0296] (Application Example 1)

[0297] 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 glasses 214 will be referred to as the "terminal."

[0298] In dual-income households, a key challenge is how to enhance the safety and emotional security of children when they are home alone. It is necessary to alleviate children's feelings of isolation while their parents are away, while also providing emergency preparedness information and safety tips to reduce parental anxiety.

[0299] 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.

[0300] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing the parent's voice and learning its characteristics as digital data, voice generation means for generating arbitrary responses that resemble the parent's voice based on the recorded voice and learned characteristic data, and information provision means for providing safety information and emergency response methods. This makes it possible to provide safety information and enhance a sense of security by engaging in natural conversation with the child on behalf of the parent.

[0301] "Voice acquisition means" refers to devices or methods for recording the voices of parents.

[0302] "Voice learning methods" refer to devices or methods that analyze the acquired voice of a parent and learn its characteristics as digital data.

[0303] "Voice generation means" refers to a device or method that generates an arbitrary response in a manner resembling the voice of a parent, based on recorded voice and learned feature data.

[0304] The "output means" is a device or method for outputting the generated voice to the child.

[0305] The "input processing means" is a device or method for receiving an input from a child as voice or text and generating a response.

[0306] The "data management means" is a device or method for storing and analyzing the conversation log.

[0307] The "information providing means" is a device or method for providing safety information and emergency response methods to the child.

[0308] This system obtains the parent's voice, generates a response based on it, and provides it to the child, giving a sense of security even when the parent is absent. Specifically, the voice acquisition means is performed by the user (parent)'s smart device. Using this device, the parent records their own voice and uploads it to the server. The server uses cloud services as the speech learning means, analyzes the voice data, and learns the parent's voice characteristics. Google Cloud Speech-to-Text API is used for this learning.

[0309] As the voice generation means, using OpenAI's text-to-speech API, a response similar to the parent's voice is generated based on the learned feature data. The server is always running and quickly generates and outputs a response to a request from the terminal. This generated voice is delivered to the child through a dedicated application running on the terminal.

[0310] The terminal transmits the received voice or text to the server as the input processing means. Furthermore, it is also possible to provide safety information and emergency response measures through the information providing means. This not only simulates daily communication but is also an important function for protecting the child's safety.

[0311] In this way, the device speaks to the child on behalf of the parent, and the conversation logs are saved to a cloud service using data management tools for analysis. Parents can later review these logs to improve family communication and security. For example, when the child returns home, the device asks in the parent's voice, "What did you do at school today?", and the generated voice responds with the child's answer, "It was fun, let's do our best tomorrow too." An example of a prompt to support such a conversation is, "Please respond in a way that shows interest as a parent in what your child has said."

[0312] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0313] Step 1:

[0314] Users use a dedicated application to record their voice via their smart device and upload this data to a server. The input is the user's voice, and the output is an audio file stored on cloud storage. The recorded audio is sent to the server by an audio acquisition method and saved as an audio file.

[0315] Step 2:

[0316] The server analyzes uploaded audio data using speech learning methods. The input is an audio file, and internally, the Google Cloud Speech-to-Text API is used to extract digital feature data such as pitch, speed, and intonation. This digital feature data is integrated into a generative AI model and stored as training data. The output is a data model that has learned the parent's speech features.

[0317] Step 3:

[0318] The device receives voice or text input from the child. The input is either the child's voice or text; in the case of voice input, it is converted to text using speech recognition technology. The converted text data is output and sent to the server.

[0319] Step 4:

[0320] The server generates a response based on the received text data. It uses prompts to instruct the AI ​​model, which then plans an appropriate response. The input is text data based on the child's speech, and the output is the text data to be used as the response. An example of a prompt is, "As a parent, please respond to what your child has said with interest."

[0321] Step 5:

[0322] The server uses speech generation technology to convert the planned text data into speech that resembles the parent's voice. Using the OpenAI speech synthesis API, it applies speech feature data to the input text to generate a response in the parent's voice. The output is the generated speech data.

[0323] Step 6:

[0324] The generated audio data is sent to the device and played back to the child through the device's speaker as the output. The input is audio data, and the output is the child's response via the audio interface.

[0325] Step 7:

[0326] The server stores all conversation logs using data management tools and performs periodic analysis. The input is conversation data recorded via communication, which is then analyzed to extract key points. This allows parents to review the conversation logs later and use the information to improve safety and communication. The output is the data resulting from the analysis.

[0327] 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.

[0328] This invention is a system that can recognize the emotions of the user (parent) and generate an optimal response accordingly. By considering the parent's emotions, this system improves the quality of responses, making conversations with children more natural and humane.

[0329] The user (parent) records voice using a dedicated app and uploads the voice data to the server from their device. The server receives this data, analyzes it, and extracts voice characteristics. Furthermore, it uses an emotion engine to recognize the parent's emotional state from this voice. Emotion recognition includes a process of classifying emotions into categories such as happiness, anger, and surprise.

[0330] When a child inputs information via voice or text through the device, the device sends it to the server. The server generates an appropriate response based on the input, taking into account the parent's emotional state. The emotion engine detects emotional fluctuations in real time and adjusts the response accordingly. For example, if the parent is relaxed, the response will be in a calm tone; if the parent is excited, it will be in an energetic tone.

[0331] The generated responses are provided to the child as speech synthesized to mimic the characteristics of the parent's voice. This allows the child to enjoy conversations that reflect the parent's emotions. As the interaction with the child progresses, the server manages all conversations and associated emotional data as logs, which the user can review later.

[0332] For example, suppose a child inputs, "I had a bad day at school today." If the system recognizes that the parent's current emotional state is calm and that encouragement is needed, it will generate an encouraging response such as, "That must have been tough. But it's okay, tomorrow will surely be a better day," and play it back in the parent's voice.

[0333] Thus, the present invention functions as a system that enables empathetic dialogue with children by generating responses that include the parent's emotions.

[0334] The following describes the processing flow.

[0335] Step 1:

[0336] The user (parent) launches a dedicated app and records audio. The recorded audio data is then sent directly to the server by the device.

[0337] Step 2:

[0338] The server analyzes the received audio data and extracts voice characteristics. In this process, it uses an emotion engine to recognize the emotions contained in the audio. The emotion engine infers emotions from the tone, speed, and intonation of the voice and records them as data.

[0339] Step 3:

[0340] When a child inputs voice or text into the device, the device forwards this input to a server. The server converts this input into text format and performs natural language processing.

[0341] Step 4:

[0342] The server generates an appropriate response to the child's input. When generating the response, it takes into account the parent's emotional data, which has been analyzed in advance, and sets the message tone to match the parent's current emotional state.

[0343] Step 5:

[0344] The server performs speech synthesis based on the generated response and creates speech data that resembles the parent's voice using pre-learned characteristics of the parent's voice.

[0345] Step 6:

[0346] The device plays this generated audio back to the child in real time. This allows the child to experience a conversation that reflects the parent's emotions.

[0347] Step 7:

[0348] If the conversation continues, return to step 3 and repeat the same process with the new input.

[0349] Step 8:

[0350] The server logs all conversation data and emotional states. This saved data can be reviewed later by parents to help improve communication with their children.

[0351] (Example 2)

[0352] 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".

[0353] In conversations with children, mechanical responses that disregard the parent's emotional state are often produced, leading to a problem where children do not feel adequately cared for. In particular, when a parent's emotions do not translate into appropriate responses to the child's needs or situation, parent-child communication may become weak. Therefore, the challenge is to achieve natural dialogue that takes the parent's emotions into consideration.

[0354] 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.

[0355] In this invention, the server includes emotion recognition means for analyzing the parent's emotional state and reflecting it in the response, speech synthesis means for generating arbitrary responses that resemble the parent's speech characteristics based on recorded voice data and learned feature data, and data management means for storing conversation history and related data for later analysis. This makes it possible to provide children with personalized and interactive responses that take the parent's emotions into consideration.

[0356] "Data acquisition means" refers to a device or algorithm for acquiring and recording the voice of a parent.

[0357] "Analysis means" refers to a device or program for extracting and learning speech features in digital format from acquired speech data.

[0358] A "speech synthesis means" is a device or algorithm for generating arbitrary responses that resemble the speech features of the parent, based on learned feature data.

[0359] "Output control means" refers to a device or program for outputting generated audio to a child under selected conditions.

[0360] "Input processing means" refers to a device or program for receiving inquiries from children in voice or text format and generating appropriate responses.

[0361] An "emotion recognition tool" is a device or algorithm that analyzes a parent's emotional state and uses the results to inform the response.

[0362] "Data management means" refers to a device or program for properly storing conversation history and related data so that it can be analyzed later.

[0363] "Speech recognition means" refers to a device or software for converting speech input into text data.

[0364] "Emotional analysis means" refers to a device or program that analyzes emotional states in real time and adjusts responses so that the generated voice can mimic the emotional state of the parent.

[0365] This system takes into account the parent's emotional state and is configured to generate natural and humane responses in interactions with children. Its embodiments are described in detail below.

[0366] The user (parent) records their voice using a dedicated software application. This recorded voice is processed as hardware by a mobile device such as a smartphone or tablet and uploaded to a server via a data acquisition method. Secure communication protocols such as HTTPS are used to ensure the security of the communication.

[0367] The server uses audio signal processing libraries such as "OpenSMILE" to extract audio features such as pitch, energy, and periodicity from the acquired audio data. Next, it uses emotion recognition algorithms such as "DeepMoji" to analyze the emotional state contained in the audio data. This information allows the system to understand the parent's current emotions in real time.

[0368] The child inputs their situation and feelings into the device via voice or text. The device receives this input and sends it to a server running a natural language generation model such as "GPT-3," which generates an appropriate response tailored to the parent's emotional state. In this process, the prompt sentence is constructed for the generation AI model. For example, a prompt sentence might be set to, "How would you respond if the parent's mood is happy and the child says they have made a new friend?"

[0369] The generated responses are converted into speech using speech synthesis technology such as "Google Text-to-Speech." By mimicking the parent's voice characteristics, the child can experience a conversation that feels as if they are actually talking to their parent. The device then plays the synthesized speech, providing the child with responses that reflect the parent's emotions.

[0370] Furthermore, the server saves all conversation data and associated sentiment data as logs. This allows users to review this data later and use it to reflect on and improve their communication with their children.

[0371] This system aims to facilitate smooth communication between parents and children by generating personalized and interactive responses that take into account the parent's feelings.

[0372] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0373] Step 1:

[0374] The user (parent) records their voice using a dedicated application. At this stage, the input is the parent's voice, and the output is a digital audio file. The recorded audio file is stored on the device and sent to the server by a data acquisition means. Specifically, the device uploads the file to the server using a secure protocol (e.g., HTTPS) via Wi-Fi or mobile data communication.

[0375] Step 2:

[0376] The server uses an audio signal processing library such as "OpenSMILE" to analyze the received audio data. The input is the previously uploaded audio file, and the output is extracted audio feature data (pitch, energy, periodicity, etc.). Specifically, the server passes the audio data to an analysis algorithm, generates feature vectors, captures changes in each time frame, and records them as features.

[0377] Step 3:

[0378] The server uses emotion recognition algorithms such as "DeepMoji" to determine the parent's emotional state based on voice feature data. The input is voice feature data, and the output is a specific emotion category (e.g., happiness, anger, surprise). In this process, the server inputs the feature data into a neural network model, calculates the probability of each emotion category, and determines the most likely emotional state.

[0379] Step 4:

[0380] Children input voice or text through a dedicated terminal application. The input is the text or voice entered by the child, and the output is text data sent to the server. Specifically, the child's voice is converted to text by the speech recognition function on the terminal and sent to the server as input.

[0381] Step 5:

[0382] The server generates a response using a generative AI model such as "GPT-3" based on the child's text input it receives. In this case, the input is the child's text and the parent's emotional state, and the output is the generated response sentence. Specifically, the server provides the input sentence as a prompt sentence to the generative AI model and obtains a text response that takes the parent's emotional state into account. For example, a prompt such as "How would you respond if the parent's emotional state is calm and the child says they have made a new friend?" might be used.

[0383] Step 6:

[0384] The server converts the generated response into speech using a text-to-speech engine (e.g., "Google Text-to-Speech"). The input is the generated response sentence, and the output is audio data. Specifically, the synthesized speech is parameterized to mimic the parent's voice and encoded as an audio file.

[0385] Step 7:

[0386] The device receives the synthesized speech and plays it back to the child through the speaker. At this stage, the input is the audio data received from the server, and the output is the speech as physical sound. Specifically, the device loads the audio file into a buffer and plays it back through the speaker at a normal volume using playback software.

[0387] Step 8:

[0388] The server records all parent-child dialogue data and associated emotional information in a database, retaining it for later user access. The input is all data generated during the conversation, and the output is a structured log dataset. Specifically, it stores timestamps, text, voice features, emotional states, etc., for each dialogue session as database records, providing a basis for future analysis.

[0389] (Application Example 2)

[0390] 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."

[0391] It is difficult for parents and store staff to automatically generate responses that accurately reflect their emotional state when interacting with children or customers, making it challenging to maintain natural and human-centered communication. Therefore, there is a need for improved communication in close relationships and enhanced customer satisfaction in store services. Furthermore, there is a demand for systems that can support appropriate responses based on the staff's emotional state during customer service.

[0392] 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.

[0393] In this invention, the server includes voice acquisition means for recording the voice of a parent or staff member; voice learning means for analyzing the voice of a parent or staff member and learning its characteristics as digital data; and voice generation means for generating appropriate responses that resemble the voice of a parent or staff member based on the recorded voice and learned characteristic data. This makes it possible to provide a system that enables appropriate responses according to the emotional state of the parent or staff member, and facilitates natural and humane communication with children and customers.

[0394] "Voice acquisition means" refers to devices or technologies for recording the voices of parents or staff.

[0395] "Speech learning methods" refer to technologies that analyze acquired speech and learn its characteristics as digital data.

[0396] "Voice generation means" refers to a technology that generates appropriate responses in a manner similar to the voice of a parent or staff member, based on recorded voice and learned feature data.

[0397] "Output means" refers to a device or method for providing the generated audio to a child or customer.

[0398] "Input processing means" refers to technology that receives input from a child or customer as voice or text and generates a response.

[0399] "Data management methods" refer to technologies for saving and analyzing conversation logs.

[0400] "Emotional adjustment techniques" refer to methods that suggest appropriate responses during customer service based on the emotional state of the staff member.

[0401] "Speech recognition means" refers to technology for converting speech into text data.

[0402] "Emotional analysis methods" are methods of analyzing emotions so that the generated voice can mimic the emotional state of a parent or staff member.

[0403] The system for implementing this invention is initiated when a parent or staff member records speech using a speech acquisition means and uploads the speech data to a server via a terminal. The server analyzes the speech using a speech learning means and extracts its features as digital data. In this process, the pitch, tone, and speed of the speech are taken into consideration.

[0404] The server uses emotion adjustment mechanisms to recognize the emotional state of staff and generate real-time, adjustable responses. At this time, a generative AI model is utilized to process the audio or text received by the input processing mechanism from the child or customer. The voice generation mechanism then mimics the voices of parents or staff to create the optimal response. The output mechanism provides the generated audio to the child or customer, enabling natural and human-like dialogue.

[0405] For example, if a user is wearing smart glasses, the system uses voice recognition to transcribe customer requests into text and suggests products based on that content. The server uses data management to save conversation logs and organizes the analytical data for later access by the user.

[0406] A concrete example of a prompt would be, "Think of the best way to interact with a customer in a relaxed manner. For example, how can you make a friendly suggestion when recommending a new product?" This prompt is used to provide guidance for the generative AI model to generate responses that improve the quality of customer service.

[0407] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0408] Step 1:

[0409] The user records the voice of a parent or staff member using an audio acquisition device and uploads the audio data to the server via the terminal. The input is audio data, and the output is digital audio data stored on the server. The server performs format conversion to prepare this data.

[0410] Step 2:

[0411] The server uses speech learning methods to analyze uploaded audio data and extract speech features. Here, it generates speech feature vectors using Mel-frequency cepstrum coefficients (MFCCs) and speech pitch, among other things. The input is digital audio data, and the output is a speech feature vector.

[0412] Step 3:

[0413] The server uses emotion adjustment mechanisms to analyze emotional states from speech feature vectors. In this process, it determines an emotional category based on each feature, classifying them into categories such as relaxation or stress. The input is a speech feature vector, and the output is emotional state information.

[0414] Step 4:

[0415] The terminal receives input from a child or customer. The input processing device converts the speech into text data. The input is customer speech data, and the output is text data. The terminal uses speech recognition software to perform this conversion.

[0416] Step 5:

[0417] The server uses a generative AI model to generate the optimal response from emotional state information and text data. The generated response is output as text containing expressions appropriate to the emotion. The AI ​​model used here is a pre-trained natural language generation model.

[0418] Step 6:

[0419] The server uses speech generation means to convert the generated text responses into speech. In this process, speech synthesis technology is used to mimic the voices of parents or staff, and to generate speech with emotionally appropriate tones. The input is text responses, and the output is synthesized speech data.

[0420] Step 7:

[0421] The device uses an output mechanism to provide the generated synthesized voice to the child or customer. Through this voice, the user can experience natural and human-like dialogue. The output is delivered via a speaker or headphones.

[0422] This series of steps enables the system to engage in natural conversations that respond to the emotions of parents and staff, thereby improving efficiency in real-world customer service situations.

[0423] 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.

[0424] 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.

[0425] 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.

[0426] [Third Embodiment]

[0427] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0428] 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.

[0429] 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).

[0430] 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.

[0431] 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.

[0432] 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).

[0433] 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.

[0434] 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.

[0435] 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.

[0436] 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.

[0437] 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.

[0438] 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".

[0439] This invention is a system aimed at reducing feelings of isolation and alleviating parental anxiety in dual-income households where children are home alone. This system uses a generative AI to reproduce the parent's voice, enabling natural conversation with the child. Specific embodiments of this invention are described below.

[0440] The user (parent) uses a dedicated application to record their voice and upload it to the server. This voice data is analyzed on the server to learn the characteristics of the parent's voice. The learning process includes parameters such as the parent's voice pitch, speed, and intonation, and these characteristics are integrated into the voice model.

[0441] The device receives voice or text input from the child. The received voice is converted into text data on the server using speech recognition technology. Based on this text data, the server generates an appropriate response and creates that response as a voice that resembles the parent's voice.

[0442] The generated audio is played from the device to the child, enabling a real-time conversational experience. The child can speak further through the device, and each time a new response is generated, sustaining a natural conversation.

[0443] As a concrete example of this system, imagine a scenario where a child returns home from school and the device speaks to them in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, saying, "That sounds interesting, what did you make?" and plays it back from the device.

[0444] Furthermore, the server records all conversations and performs periodic analysis. Parents can later review these conversation logs and use the information to improve future communication. In this way, the present invention is a system that technically complements parent-child conversations even in the absence of parents, providing a sense of security and connection.

[0445] The following describes the processing flow.

[0446] Step 1:

[0447] The user (parent) launches the app and records their voice. The recorded audio data is uploaded directly from the device to the server.

[0448] Step 2:

[0449] The server analyzes the received audio data and extracts characteristics of the parent's voice. This includes information such as voice pitch, speed, intonation, and word patterns, and uses this data to train an AI model.

[0450] Step 3:

[0451] The device receives information entered by the child via voice or text. In the case of voice input, it uses speech recognition technology to convert it into text data.

[0452] Step 4:

[0453] The server performs natural language processing on the received text data to generate appropriate responses to questions and conversations. These responses are designed to mimic the tone and phrasing of the parent.

[0454] Step 5:

[0455] Based on the generated text response, the server uses a pre-trained speech model to generate speech data that reproduces the parent's voice.

[0456] Step 6:

[0457] The device plays the generated audio data to the child in real time. This allows the child to have an experience similar to interacting with their parent.

[0458] Step 7:

[0459] If the conversation continues, return to step 3 and receive new input from the child. This maintains a continuous and natural flow of dialogue.

[0460] Step 8:

[0461] The server logs all conversations and analyzes them later. This helps identify frequently occurring phrases and emotional states in conversations, and the results are fed back to parents for review.

[0462] (Example 1)

[0463] 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."

[0464] In dual-income households, children often feel isolated when their parents are absent, and parents themselves feel anxious because they are unaware of their children's situation. There was a need for technological solutions to compensate for parental absence, alleviate children's feelings of isolation, and provide parents with a sense of security.

[0465] 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.

[0466] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing and learning the characteristics of the parent's voice, voice generation means for generating arbitrary responses that resemble the parent's voice using a generative AI model, and means for outputting the voice to the child. This makes it possible to have natural conversations with the child using the parent's voice even when the parent is absent, thereby alleviating the child's sense of isolation and providing the parent with a sense of security.

[0467] "Voice acquisition means" refers to a device or function for recording the voice of a parent.

[0468] "Voice learning methods" refer to a function that analyzes the characteristics of a parent's voice, such as pitch, speed, and intonation, and learns them as digital information.

[0469] "Voice generation means" refers to a function that uses a generative AI model to generate arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature information.

[0470] "Means of reproducing sound" refers to functions including speakers and audio playback devices for outputting generated sound to children.

[0471] The "input processing means" is a function that receives input from a child, either as voice or text, and generates an appropriate response.

[0472] "Voice recognition means" refers to a function that converts a child's voice input into text data and generates a response.

[0473] A "conversation management system" is a data management function that records all conversations to enable future analysis.

[0474] "Emotional analysis means" refers to a function that performs emotional analysis on the generated voice to mimic the parent's emotional state and provide a more natural response.

[0475] A "conversation recording device" is a function that records the process by which generated responses are output to the child in real time, and uses this information for future analysis.

[0476] This invention is a system designed to alleviate feelings of isolation in children and reduce parental anxiety during the absence of parents in dual-income households. The system aims to enable natural conversations with children by faithfully reproducing the parent's voice using a generative AI model. Specific embodiments of this invention are described below.

[0477] First, the user (parent) records their own voice using a dedicated application. A device with a microphone is used for this recording. The recorded audio data is sent to a server. The server uses voice analysis software to extract features such as pitch, speed, and intonation from the audio data. Based on this data, a generative AI model is trained to create a voice generation model that reproduces the parent's voice.

[0478] The device receives voice and text input from the child. The received voice data is sent to a server. On the server, speech recognition technology is used to convert the voice into text. Based on this text, an appropriate response is generated by a generation AI. This response is synthesized to sound like the parent's voice.

[0479] This generated audio data is sent to the device and played back to the child. This allows the child to have a conversational experience as if their parent were present.

[0480] As a concrete example, imagine a scenario where, upon a child returning home from school, the device automatically speaks in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, such as, "That sounds interesting, what did you make?", and plays it back from the device. This enables natural, real-time conversation.

[0481] As a concrete example of a prompt, an instruction such as "Ask the child about their thoughts today in a parent's voice, and generate a response that mimics the parent's voice" can be input into a generative AI model, thereby facilitating natural dialogue between parent and child.

[0482] In this way, the present invention is a system that uses technical means to compensate for the absence of parents and maintain the bond between parents and children.

[0483] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0484] Step 1:

[0485] The user (parent) records their voice using a dedicated application. Here, a device with a microphone is used to record the parent's speech as a digital audio file. This audio data serves as input data for later analysis of the parent's voice characteristics. The completed audio recording is temporarily stored for use in the next step.

[0486] Step 2:

[0487] The user uploads recorded audio data to the server. Based on this uploaded audio data, the server uses speech analysis software to extract features such as pitch, speed, and intonation. These extracted features are used as training data for a speech generation AI model. This allows the server to create a speech generation model that reproduces the parent's voice. The output is digital data that retains the features of the parent's voice.

[0488] Step 3:

[0489] The device receives voice or text input from the child. Here, the child inputs voice or text through the device's microphone or keyboard. The input is in real time, and the device temporarily stores this data before sending it to the server. In the case of voice input, the input data is directly processed for speech recognition in the next step.

[0490] Step 4:

[0491] The server processes the audio data sent by the child using speech recognition software and converts it into text data. This text data is then used as new input for response generation. Preprocessing, such as noise reduction and speech intensity normalization, is performed. The output is the audio converted to text.

[0492] Step 5:

[0493] The server generates an appropriate response based on the text data. It uses a generative AI model to analyze the input text and create a response based on the prompt. This response is synthesized as speech using parental voice features. The output of this step is the generated response as speech data.

[0494] Step 6:

[0495] The server sends the generated audio data to the device. The device plays the received audio through its speaker and delivers it to the child. The played audio is very similar to the parent's voice, providing the child with a natural conversational experience.

[0496] Step 7:

[0497] The server records the entire conversation and saves it to a database for analysis. The data saved as a conversation log can be reviewed later by parents to help improve parent-child communication. The output is data recording the content of the conversation.

[0498] (Application Example 1)

[0499] 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."

[0500] In dual-income households, a key challenge is how to enhance the safety and emotional security of children when they are home alone. It is necessary to alleviate children's feelings of isolation while their parents are away, while also providing emergency preparedness information and safety tips to reduce parental anxiety.

[0501] 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.

[0502] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing the parent's voice and learning its characteristics as digital data, voice generation means for generating arbitrary responses that resemble the parent's voice based on the recorded voice and learned characteristic data, and information provision means for providing safety information and emergency response methods. This makes it possible to provide safety information and enhance a sense of security by engaging in natural conversation with the child on behalf of the parent.

[0503] "Voice acquisition means" refers to devices or methods for recording the voices of parents.

[0504] "Voice learning methods" refer to devices or methods that analyze the acquired voice of a parent and learn its characteristics as digital data.

[0505] "Voice generation means" refers to a device or method that generates an arbitrary response in a manner resembling the voice of a parent, based on recorded voice and learned feature data.

[0506] "Output means" refers to a device or method for outputting the generated sound to a child.

[0507] "Input processing means" refers to a device or method that receives input from a child as voice or text and generates a response.

[0508] "Data management means" refers to devices or methods for saving and analyzing conversation logs.

[0509] "Information provision means" refers to devices or methods for providing children with safety information and emergency response procedures.

[0510] This system provides a sense of security to children even when the parents are absent by acquiring the parents' voices and generating responses based on them. Specifically, voice acquisition is performed using the user's (parent's) smart device. Using this device, the parent records their voice and uploads it to the server. The server uses a cloud service as a voice learning tool, analyzing the voice data to learn the parents' voice characteristics. The Google Cloud Speech-to-Text API is used for this learning process.

[0511] The system uses OpenAI's speech synthesis API to generate responses that mimic the parent's voice based on learned feature data. The server is always running, quickly generating and outputting responses to requests from the device. This generated voice is delivered to the child through a dedicated application running on the device.

[0512] The terminal transmits received audio and text to the server as input processing. Furthermore, it can provide safety information and emergency response measures through information provision mechanisms. This is not only a simulation of everyday communication, but also a crucial function for protecting children's safety.

[0513] In this way, the device speaks to the child on behalf of the parent, and the conversation logs are saved to a cloud service using data management tools for analysis. Parents can later review these logs to improve family communication and security. For example, when the child returns home, the device asks in the parent's voice, "What did you do at school today?", and the generated voice responds with the child's answer, "It was fun, let's do our best tomorrow too." An example of a prompt to support such a conversation is, "Please respond in a way that shows interest as a parent in what your child has said."

[0514] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0515] Step 1:

[0516] Users use a dedicated application to record their voice via their smart device and upload this data to a server. The input is the user's voice, and the output is an audio file stored on cloud storage. The recorded audio is sent to the server by an audio acquisition method and saved as an audio file.

[0517] Step 2:

[0518] The server analyzes uploaded audio data using speech learning methods. The input is an audio file, and internally, the Google Cloud Speech-to-Text API is used to extract digital feature data such as pitch, speed, and intonation. This digital feature data is integrated into a generative AI model and stored as training data. The output is a data model that has learned the parent's speech features.

[0519] Step 3:

[0520] The device receives voice or text input from the child. The input is either the child's voice or text; in the case of voice input, it is converted to text using speech recognition technology. The converted text data is output and sent to the server.

[0521] Step 4:

[0522] The server generates a response based on the received text data. It uses prompts to instruct the AI ​​model, which then plans an appropriate response. The input is text data based on the child's speech, and the output is the text data to be used as the response. An example of a prompt is, "As a parent, please respond to what your child has said with interest."

[0523] Step 5:

[0524] The server uses speech generation technology to convert the planned text data into speech that resembles the parent's voice. Using the OpenAI speech synthesis API, it applies speech feature data to the input text to generate a response in the parent's voice. The output is the generated speech data.

[0525] Step 6:

[0526] The generated audio data is sent to the device and played back to the child through the device's speaker as the output. The input is audio data, and the output is the child's response via the audio interface.

[0527] Step 7:

[0528] The server stores all conversation logs using data management tools and performs periodic analysis. The input is conversation data recorded via communication, which is then analyzed to extract key points. This allows parents to review the conversation logs later and use the information to improve safety and communication. The output is the data resulting from the analysis.

[0529] 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.

[0530] This invention is a system that can recognize the emotions of the user (parent) and generate an optimal response accordingly. By considering the parent's emotions, this system improves the quality of responses, making conversations with children more natural and humane.

[0531] The user (parent) records voice using a dedicated app and uploads the voice data to the server from their device. The server receives this data, analyzes it, and extracts voice characteristics. Furthermore, it uses an emotion engine to recognize the parent's emotional state from this voice. Emotion recognition includes a process of classifying emotions into categories such as happiness, anger, and surprise.

[0532] When a child inputs information via voice or text through the device, the device sends it to the server. The server generates an appropriate response based on the input, taking into account the parent's emotional state. The emotion engine detects emotional fluctuations in real time and adjusts the response accordingly. For example, if the parent is relaxed, the response will be in a calm tone; if the parent is excited, it will be in an energetic tone.

[0533] The generated responses are provided to the child as speech synthesized to mimic the characteristics of the parent's voice. This allows the child to enjoy conversations that reflect the parent's emotions. As the interaction with the child progresses, the server manages all conversations and associated emotional data as logs, which the user can review later.

[0534] For example, suppose a child inputs, "I had a bad day at school today." If the system recognizes that the parent's current emotional state is calm and that encouragement is needed, it will generate an encouraging response such as, "That must have been tough. But it's okay, tomorrow will surely be a better day," and play it back in the parent's voice.

[0535] Thus, the present invention functions as a system that enables empathetic dialogue with children by generating responses that include the parent's emotions.

[0536] The following describes the processing flow.

[0537] Step 1:

[0538] The user (parent) launches a dedicated app and records audio. The recorded audio data is then sent directly to the server by the device.

[0539] Step 2:

[0540] The server analyzes the received audio data and extracts voice characteristics. In this process, it uses an emotion engine to recognize the emotions contained in the audio. The emotion engine infers emotions from the tone, speed, and intonation of the voice and records them as data.

[0541] Step 3:

[0542] When a child inputs voice or text into the device, the device forwards this input to a server. The server converts this input into text format and performs natural language processing.

[0543] Step 4:

[0544] The server generates an appropriate response to the child's input. When generating the response, it takes into account the parent's emotional data, which has been analyzed in advance, and sets the message tone to match the parent's current emotional state.

[0545] Step 5:

[0546] The server performs speech synthesis based on the generated response and creates speech data that resembles the parent's voice using pre-learned characteristics of the parent's voice.

[0547] Step 6:

[0548] The device plays this generated audio back to the child in real time. This allows the child to experience a conversation that reflects the parent's emotions.

[0549] Step 7:

[0550] If the conversation continues, return to step 3 and repeat the same process with the new input.

[0551] Step 8:

[0552] The server logs all conversation data and emotional states. This saved data can be reviewed later by parents to help improve communication with their children.

[0553] (Example 2)

[0554] 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."

[0555] In conversations with children, mechanical responses that disregard the parent's emotional state are often produced, leading to a problem where children do not feel adequately cared for. In particular, when a parent's emotions do not translate into appropriate responses to the child's needs or situation, parent-child communication may become weak. Therefore, the challenge is to achieve natural dialogue that takes the parent's emotions into consideration.

[0556] 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.

[0557] In this invention, the server includes emotion recognition means for analyzing the parent's emotional state and reflecting it in the response, speech synthesis means for generating arbitrary responses that resemble the parent's speech characteristics based on recorded voice data and learned feature data, and data management means for storing conversation history and related data for later analysis. This makes it possible to provide children with personalized and interactive responses that take the parent's emotions into consideration.

[0558] "Data acquisition means" refers to a device or algorithm for acquiring and recording the voice of a parent.

[0559] "Analysis means" refers to a device or program for extracting and learning speech features in digital format from acquired speech data.

[0560] A "speech synthesis means" is a device or algorithm for generating arbitrary responses that resemble the speech features of the parent, based on learned feature data.

[0561] "Output control means" refers to a device or program for outputting generated audio to a child under selected conditions.

[0562] "Input processing means" refers to a device or program for receiving inquiries from children in voice or text format and generating appropriate responses.

[0563] An "emotion recognition tool" is a device or algorithm that analyzes a parent's emotional state and uses the results to inform the response.

[0564] "Data management means" refers to a device or program for properly storing conversation history and related data so that it can be analyzed later.

[0565] "Speech recognition means" refers to a device or software for converting speech input into text data.

[0566] "Emotional analysis means" refers to a device or program that analyzes emotional states in real time and adjusts responses so that the generated voice can mimic the emotional state of the parent.

[0567] This system takes into account the parent's emotional state and is configured to generate natural and humane responses in interactions with children. Its embodiments are described in detail below.

[0568] The user (parent) records their voice using a dedicated software application. This recorded voice is processed as hardware by a mobile device such as a smartphone or tablet and uploaded to a server via a data acquisition method. Secure communication protocols such as HTTPS are used to ensure the security of the communication.

[0569] The server uses audio signal processing libraries such as "OpenSMILE" to extract audio features such as pitch, energy, and periodicity from the acquired audio data. Next, it uses emotion recognition algorithms such as "DeepMoji" to analyze the emotional state contained in the audio data. This information allows the system to understand the parent's current emotions in real time.

[0570] The child inputs their situation and feelings into the device via voice or text. The device receives this input and sends it to a server running a natural language generation model such as "GPT-3," which generates an appropriate response tailored to the parent's emotional state. In this process, the prompt sentence is constructed for the generation AI model. For example, a prompt sentence might be set to, "How would you respond if the parent's mood is happy and the child says they have made a new friend?"

[0571] The generated responses are converted into speech using speech synthesis technology such as "Google Text-to-Speech." By mimicking the parent's voice characteristics, the child can experience a conversation that feels as if they are actually talking to their parent. The device then plays the synthesized speech, providing the child with responses that reflect the parent's emotions.

[0572] Furthermore, the server saves all conversation data and associated sentiment data as logs. This allows users to review this data later and use it to reflect on and improve their communication with their children.

[0573] This system aims to facilitate smooth communication between parents and children by generating personalized and interactive responses that take into account the parent's feelings.

[0574] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0575] Step 1:

[0576] The user (parent) records their voice using a dedicated application. At this stage, the input is the parent's voice, and the output is a digital audio file. The recorded audio file is stored on the device and sent to the server by a data acquisition means. Specifically, the device uploads the file to the server using a secure protocol (e.g., HTTPS) via Wi-Fi or mobile data communication.

[0577] Step 2:

[0578] The server uses an audio signal processing library such as "OpenSMILE" to analyze the received audio data. The input is the previously uploaded audio file, and the output is extracted audio feature data (pitch, energy, periodicity, etc.). Specifically, the server passes the audio data to an analysis algorithm, generates feature vectors, captures changes in each time frame, and records them as features.

[0579] Step 3:

[0580] The server uses emotion recognition algorithms such as "DeepMoji" to determine the parent's emotional state based on voice feature data. The input is voice feature data, and the output is a specific emotion category (e.g., happiness, anger, surprise). In this process, the server inputs the feature data into a neural network model, calculates the probability of each emotion category, and determines the most likely emotional state.

[0581] Step 4:

[0582] Children input voice or text through a dedicated terminal application. The input is the text or voice entered by the child, and the output is text data sent to the server. Specifically, the child's voice is converted to text by the speech recognition function on the terminal and sent to the server as input.

[0583] Step 5:

[0584] The server generates a response using a generative AI model such as "GPT-3" based on the child's text input it receives. In this case, the input is the child's text and the parent's emotional state, and the output is the generated response sentence. Specifically, the server provides the input sentence as a prompt sentence to the generative AI model and obtains a text response that takes the parent's emotional state into account. For example, a prompt such as "How would you respond if the parent's emotional state is calm and the child says they have made a new friend?" might be used.

[0585] Step 6:

[0586] The server converts the generated response into speech using a text-to-speech engine (e.g., "Google Text-to-Speech"). The input is the generated response sentence, and the output is the audio data. Specifically, the synthesized speech is parameterized to mimic the parent's voice and encoded as an audio file.

[0587] Step 7:

[0588] The device receives the synthesized speech and plays it back to the child through the speaker. At this stage, the input is the audio data received from the server, and the output is the speech as physical sound. Specifically, the device loads the audio file into a buffer and plays it back through the speaker at a normal volume using playback software.

[0589] Step 8:

[0590] The server records all parent-child dialogue data and associated emotional information in a database, retaining it for later user access. The input is all data generated during the conversation, and the output is a structured log dataset. Specifically, it stores timestamps, text, voice features, emotional states, etc., for each dialogue session as database records, providing a basis for future analysis.

[0591] (Application Example 2)

[0592] 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."

[0593] It is difficult for parents and store staff to automatically generate responses that accurately reflect their emotional state when interacting with children or customers, making it challenging to maintain natural and human-centered communication. Therefore, there is a need for improved communication in close relationships and enhanced customer satisfaction in store services. Furthermore, there is a demand for systems that can support appropriate responses based on the staff's emotional state during customer service.

[0594] 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.

[0595] In this invention, the server includes voice acquisition means for recording the voice of a parent or staff member; voice learning means for analyzing the voice of a parent or staff member and learning its characteristics as digital data; and voice generation means for generating appropriate responses that resemble the voice of a parent or staff member based on the recorded voice and learned characteristic data. This makes it possible to provide a system that enables appropriate responses according to the emotional state of the parent or staff member, and facilitates natural and humane communication with children and customers.

[0596] "Voice acquisition means" refers to devices or technologies for recording the voices of parents or staff.

[0597] "Speech learning methods" refer to technologies that analyze acquired speech and learn its characteristics as digital data.

[0598] "Voice generation means" refers to a technology that generates appropriate responses in a manner similar to the voice of a parent or staff member, based on recorded voice and learned feature data.

[0599] "Output means" refers to a device or method for providing the generated audio to a child or customer.

[0600] "Input processing means" refers to technology that receives input from a child or customer as voice or text and generates a response.

[0601] "Data management methods" refer to technologies for saving and analyzing conversation logs.

[0602] "Emotional adjustment techniques" refer to methods that suggest appropriate responses during customer service based on the emotional state of the staff member.

[0603] "Speech recognition means" refers to technology for converting speech into text data.

[0604] "Emotional analysis methods" are methods of analyzing emotions so that the generated voice can mimic the emotional state of a parent or staff member.

[0605] The system for implementing this invention is initiated when a parent or staff member records speech using a speech acquisition means and uploads the speech data to a server via a terminal. The server analyzes the speech using a speech learning means and extracts its features as digital data. In this process, the pitch, tone, and speed of the speech are taken into consideration.

[0606] The server uses emotion adjustment mechanisms to recognize the emotional state of staff and generate real-time, adjustable responses. At this time, a generative AI model is utilized to process the audio or text received by the input processing mechanism from the child or customer. The voice generation mechanism then mimics the voices of parents or staff to create the optimal response. The output mechanism provides the generated audio to the child or customer, enabling natural and human-like dialogue.

[0607] For example, if a user is wearing smart glasses, the system uses voice recognition to transcribe customer requests into text and suggests products based on that content. The server uses data management to save conversation logs and organizes the analytical data for later access by the user.

[0608] A concrete example of a prompt would be, "Think of the best way to interact with a customer in a relaxed manner. For example, how can you make a friendly suggestion when recommending a new product?" This prompt is used to provide guidance for the generative AI model to generate responses that improve the quality of customer service.

[0609] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0610] Step 1:

[0611] The user records the voice of a parent or staff member using an audio acquisition device and uploads the audio data to the server via the terminal. The input is audio data, and the output is digital audio data stored on the server. The server performs format conversion to prepare this data.

[0612] Step 2:

[0613] The server uses speech learning methods to analyze uploaded audio data and extract speech features. Here, it generates speech feature vectors using Mel-frequency cepstrum coefficients (MFCCs) and speech pitch, among other things. The input is digital audio data, and the output is a speech feature vector.

[0614] Step 3:

[0615] The server uses emotion adjustment mechanisms to analyze emotional states from speech feature vectors. In this process, it determines an emotional category based on each feature, classifying them into categories such as relaxation or stress. The input is a speech feature vector, and the output is emotional state information.

[0616] Step 4:

[0617] The terminal receives input from a child or customer. The input processing device converts the speech into text data. The input is customer speech data, and the output is text data. The terminal uses speech recognition software to perform this conversion.

[0618] Step 5:

[0619] The server uses a generative AI model to generate the optimal response from emotional state information and text data. The generated response is output as text containing expressions appropriate to the emotion. The AI ​​model used here is a pre-trained natural language generation model.

[0620] Step 6:

[0621] The server uses speech generation means to convert the generated text responses into speech. In this process, speech synthesis technology is used to mimic the voices of parents or staff, and to generate speech with emotionally appropriate tones. The input is text responses, and the output is synthesized speech data.

[0622] Step 7:

[0623] The device uses an output mechanism to provide the generated synthesized voice to the child or customer. Through this voice, the user can experience natural and human-like dialogue. The output is delivered via a speaker or headphones.

[0624] This series of steps enables the system to engage in natural conversations that respond to the emotions of parents and staff, thereby improving efficiency in real-world customer service situations.

[0625] 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.

[0626] 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.

[0627] 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.

[0628] [Fourth Embodiment]

[0629] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0630] 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.

[0631] 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).

[0632] 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.

[0633] 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.

[0634] 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).

[0635] 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.

[0636] 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.

[0637] 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.

[0638] 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.

[0639] 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.

[0640] 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.

[0641] 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".

[0642] This invention is a system aimed at reducing feelings of isolation and alleviating parental anxiety in dual-income households where children are home alone. This system uses a generative AI to reproduce the parent's voice, enabling natural conversation with the child. Specific embodiments of this invention are described below.

[0643] The user (parent) uses a dedicated application to record their voice and upload it to the server. This voice data is analyzed on the server to learn the characteristics of the parent's voice. The learning process includes parameters such as the parent's voice pitch, speed, and intonation, and these characteristics are integrated into the voice model.

[0644] The device receives voice or text input from the child. The received voice is converted into text data on the server using speech recognition technology. Based on this text data, the server generates an appropriate response and creates that response as a voice that resembles the parent's voice.

[0645] The generated audio is played from the device to the child, enabling a real-time conversational experience. The child can speak further through the device, and each time a new response is generated, sustaining a natural conversation.

[0646] As a concrete example of this system, imagine a scenario where a child returns home from school and the device speaks to them in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, saying, "That sounds interesting, what did you make?" and plays it back from the device.

[0647] Furthermore, the server records all conversations and performs periodic analysis. Parents can later review these conversation logs and use the information to improve future communication. In this way, the present invention is a system that technically complements parent-child conversations even in the absence of parents, providing a sense of security and connection.

[0648] The following describes the processing flow.

[0649] Step 1:

[0650] The user (parent) launches the app and records their voice. The recorded audio data is uploaded directly from the device to the server.

[0651] Step 2:

[0652] The server analyzes the received audio data and extracts characteristics of the parent's voice. This includes information such as voice pitch, speed, intonation, and word patterns, and uses this data to train an AI model.

[0653] Step 3:

[0654] The device receives information entered by the child via voice or text. In the case of voice input, it uses speech recognition technology to convert it into text data.

[0655] Step 4:

[0656] The server performs natural language processing on the received text data to generate appropriate responses to questions and conversations. These responses are designed to mimic the tone and phrasing of the parent.

[0657] Step 5:

[0658] Based on the generated text response, the server uses a pre-trained speech model to generate speech data that reproduces the parent's voice.

[0659] Step 6:

[0660] The device plays the generated audio data to the child in real time. This allows the child to have an experience similar to interacting with their parent.

[0661] Step 7:

[0662] If the conversation continues, return to step 3 and receive new input from the child. This maintains a continuous and natural flow of dialogue.

[0663] Step 8:

[0664] The server logs all conversations and analyzes them later. This helps identify frequently occurring phrases and emotional states in conversations, and the results are fed back to parents for review.

[0665] (Example 1)

[0666] 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".

[0667] In dual-income households, children often feel isolated when their parents are absent, and parents themselves feel anxious because they are unaware of their children's situation. There was a need for technological solutions to compensate for parental absence, alleviate children's feelings of isolation, and provide parents with a sense of security.

[0668] 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.

[0669] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing and learning the characteristics of the parent's voice, voice generation means for generating arbitrary responses that resemble the parent's voice using a generative AI model, and means for outputting the voice to the child. This makes it possible to have natural conversations with the child using the parent's voice even when the parent is absent, thereby alleviating the child's sense of isolation and providing the parent with a sense of security.

[0670] "Voice acquisition means" refers to a device or function for recording the voice of a parent.

[0671] "Voice learning methods" refer to a function that analyzes the characteristics of a parent's voice, such as pitch, speed, and intonation, and learns them as digital information.

[0672] "Voice generation means" refers to a function that uses a generative AI model to generate arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature information.

[0673] "Means of reproducing sound" refers to functions including speakers and audio playback devices for outputting generated sound to children.

[0674] The "input processing means" is a function that receives input from a child, either as voice or text, and generates an appropriate response.

[0675] "Voice recognition means" refers to a function that converts a child's voice input into text data and generates a response.

[0676] A "conversation management system" is a data management function that records all conversations to enable future analysis.

[0677] "Emotional analysis means" refers to a function that performs emotional analysis on the generated voice to mimic the parent's emotional state and provide a more natural response.

[0678] A "conversation recording device" is a function that records the process by which generated responses are output to the child in real time, and uses this information for future analysis.

[0679] This invention is a system designed to alleviate feelings of isolation in children and reduce parental anxiety during the absence of parents in dual-income households. The system aims to enable natural conversations with children by faithfully reproducing the parent's voice using a generative AI model. Specific embodiments of this invention are described below.

[0680] First, the user (parent) records their own voice using a dedicated application. A device with a microphone is used for this recording. The recorded audio data is sent to a server. The server uses voice analysis software to extract features such as pitch, speed, and intonation from the audio data. Based on this data, a generative AI model is trained to create a voice generation model that reproduces the parent's voice.

[0681] The device receives voice and text input from the child. The received voice data is sent to a server. On the server, speech recognition technology is used to convert the voice into text. Based on this text, an appropriate response is generated by a generation AI. This response is synthesized to sound like the parent's voice.

[0682] This generated audio data is sent to the device and played back to the child. This allows the child to have a conversational experience as if their parent were present.

[0683] As a concrete example, imagine a scenario where, upon a child returning home from school, the device automatically speaks in the parent's voice, asking, "What did you do today?" If the child replies, "I did art class today," the server generates a response in the parent's voice, such as, "That sounds interesting, what did you make?", and plays it back from the device. This enables natural, real-time conversation.

[0684] As a concrete example of a prompt, an instruction such as "Ask the child about their thoughts today in a parent's voice, and generate a response that mimics the parent's voice" can be input into a generative AI model, thereby facilitating natural dialogue between parent and child.

[0685] In this way, the present invention is a system that uses technical means to compensate for the absence of parents and maintain the bond between parents and children.

[0686] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0687] Step 1:

[0688] The user (parent) records their voice using a dedicated application. Here, a device with a microphone is used to record the parent's speech as a digital audio file. This audio data serves as input data for later analysis of the parent's voice characteristics. The completed audio recording is temporarily stored for use in the next step.

[0689] Step 2:

[0690] The user uploads recorded audio data to the server. Based on this uploaded audio data, the server uses speech analysis software to extract features such as pitch, speed, and intonation. These extracted features are used as training data for a speech generation AI model. This allows the server to create a speech generation model that reproduces the parent's voice. The output is digital data that retains the features of the parent's voice.

[0691] Step 3:

[0692] The device receives voice or text input from the child. Here, the child inputs voice or text through the device's microphone or keyboard. The input is in real time, and the device temporarily stores this data before sending it to the server. In the case of voice input, the input data is directly processed for speech recognition in the next step.

[0693] Step 4:

[0694] The server processes the audio data sent by the child using speech recognition software and converts it into text data. This text data is then used as new input for response generation. Preprocessing, such as noise reduction and speech intensity normalization, is performed. The output is the audio converted to text.

[0695] Step 5:

[0696] The server generates an appropriate response based on the text data. It uses a generative AI model to analyze the input text and create a response based on the prompt. This response is synthesized as speech using parental voice features. The output of this step is the generated response as speech data.

[0697] Step 6:

[0698] The server sends the generated audio data to the device. The device plays the received audio through its speaker and delivers it to the child. The played audio is very similar to the parent's voice, providing the child with a natural conversational experience.

[0699] Step 7:

[0700] The server records the entire conversation and saves it to a database for analysis. The data saved as a conversation log can be reviewed later by parents to help improve parent-child communication. The output is data recording the content of the conversation.

[0701] (Application Example 1)

[0702] 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".

[0703] In dual-income households, a key challenge is how to enhance the safety and emotional security of children when they are home alone. It is necessary to alleviate children's feelings of isolation while their parents are away, while also providing emergency preparedness information and safety tips to reduce parental anxiety.

[0704] 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.

[0705] In this invention, the server includes voice acquisition means for recording the parent's voice, voice learning means for analyzing the parent's voice and learning its characteristics as digital data, voice generation means for generating arbitrary responses that resemble the parent's voice based on the recorded voice and learned characteristic data, and information provision means for providing safety information and emergency response methods. This makes it possible to provide safety information and enhance a sense of security by engaging in natural conversation with the child on behalf of the parent.

[0706] "Voice acquisition means" refers to devices or methods for recording the voices of parents.

[0707] "Voice learning methods" refer to devices or methods that analyze the acquired voice of a parent and learn its characteristics as digital data.

[0708] "Voice generation means" refers to a device or method that generates an arbitrary response in a manner resembling the voice of a parent, based on recorded voice and learned feature data.

[0709] "Output means" refers to a device or method for outputting the generated sound to a child.

[0710] "Input processing means" refers to a device or method that receives input from a child as voice or text and generates a response.

[0711] "Data management means" refers to devices or methods for saving and analyzing conversation logs.

[0712] "Information provision means" refers to devices or methods for providing children with safety information and emergency response procedures.

[0713] This system provides a sense of security to children even when the parents are absent by acquiring the parents' voices and generating responses based on them. Specifically, voice acquisition is performed using the user's (parent's) smart device. Using this device, the parent records their voice and uploads it to the server. The server uses a cloud service as a voice learning tool, analyzing the voice data to learn the parents' voice characteristics. The Google Cloud Speech-to-Text API is used for this learning process.

[0714] The system uses OpenAI's speech synthesis API to generate responses that mimic the parent's voice based on learned feature data. The server is always running, quickly generating and outputting responses to requests from the device. This generated voice is delivered to the child through a dedicated application running on the device.

[0715] The terminal transmits received audio and text to the server as input processing. Furthermore, it can provide safety information and emergency response measures through information provision mechanisms. This is not only a simulation of everyday communication, but also a crucial function for protecting children's safety.

[0716] In this way, the device speaks to the child on behalf of the parent, and the conversation logs are saved to a cloud service using data management tools for analysis. Parents can later review these logs to improve family communication and security. For example, when the child returns home, the device asks in the parent's voice, "What did you do at school today?", and the generated voice responds with the child's answer, "It was fun, let's do our best tomorrow too." An example of a prompt to support such a conversation is, "Please respond in a way that shows interest as a parent in what your child has said."

[0717] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0718] Step 1:

[0719] Users use a dedicated application to record their voice via their smart device and upload this data to a server. The input is the user's voice, and the output is an audio file stored on cloud storage. The recorded audio is sent to the server by an audio acquisition method and saved as an audio file.

[0720] Step 2:

[0721] The server analyzes uploaded audio data using speech learning methods. The input is an audio file, and internally, the Google Cloud Speech-to-Text API is used to extract digital feature data such as pitch, speed, and intonation. This digital feature data is integrated into a generative AI model and stored as training data. The output is a data model that has learned the parent's speech features.

[0722] Step 3:

[0723] The device receives voice or text input from the child. The input is either the child's voice or text; in the case of voice input, it is converted to text using speech recognition technology. The converted text data is output and sent to the server.

[0724] Step 4:

[0725] The server generates a response based on the received text data. It uses prompts to instruct the AI ​​model, which then plans an appropriate response. The input is text data based on the child's speech, and the output is the text data to be used as the response. An example of a prompt is, "As a parent, please respond to what your child has said with interest."

[0726] Step 5:

[0727] The server uses speech generation technology to convert the planned text data into speech that resembles the parent's voice. Using the OpenAI speech synthesis API, it applies speech feature data to the input text to generate a response in the parent's voice. The output is the generated speech data.

[0728] Step 6:

[0729] The generated audio data is sent to the device and played back to the child through the device's speaker as the output. The input is audio data, and the output is the child's response via the audio interface.

[0730] Step 7:

[0731] The server stores all conversation logs using data management tools and performs periodic analysis. The input is conversation data recorded via communication, which is then analyzed to extract key points. This allows parents to review the conversation logs later and use the information to improve safety and communication. The output is the data resulting from the analysis.

[0732] 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.

[0733] This invention is a system that can recognize the emotions of the user (parent) and generate an optimal response accordingly. By considering the parent's emotions, this system improves the quality of responses, making conversations with children more natural and humane.

[0734] The user (parent) records voice using a dedicated app and uploads the voice data to the server from their device. The server receives this data, analyzes it, and extracts voice characteristics. Furthermore, it uses an emotion engine to recognize the parent's emotional state from this voice. Emotion recognition includes a process of classifying emotions into categories such as happiness, anger, and surprise.

[0735] When a child inputs information via voice or text through the device, the device sends it to the server. The server generates an appropriate response based on the input, taking into account the parent's emotional state. The emotion engine detects emotional fluctuations in real time and adjusts the response accordingly. For example, if the parent is relaxed, the response will be in a calm tone; if the parent is excited, it will be in an energetic tone.

[0736] The generated responses are provided to the child as speech synthesized to mimic the characteristics of the parent's voice. This allows the child to enjoy conversations that reflect the parent's emotions. As the interaction with the child progresses, the server manages all conversations and associated emotional data as logs, which the user can review later.

[0737] For example, suppose a child inputs, "I had a bad day at school today." If the system recognizes that the parent's current emotional state is calm and that encouragement is needed, it will generate an encouraging response such as, "That must have been tough. But it's okay, tomorrow will surely be a better day," and play it back in the parent's voice.

[0738] Thus, the present invention functions as a system that enables empathetic dialogue with children by generating responses that include the parent's emotions.

[0739] The following describes the processing flow.

[0740] Step 1:

[0741] The user (parent) launches a dedicated app and records audio. The recorded audio data is then sent directly to the server by the device.

[0742] Step 2:

[0743] The server analyzes the received audio data and extracts voice characteristics. In this process, it uses an emotion engine to recognize the emotions contained in the audio. The emotion engine infers emotions from the tone, speed, and intonation of the voice and records them as data.

[0744] Step 3:

[0745] When a child inputs voice or text into the device, the device forwards this input to a server. The server converts this input into text format and performs natural language processing.

[0746] Step 4:

[0747] The server generates an appropriate response to the child's input. When generating the response, it takes into account the parent's emotional data, which has been analyzed in advance, and sets the message tone to match the parent's current emotional state.

[0748] Step 5:

[0749] The server performs speech synthesis based on the generated response and creates speech data that resembles the parent's voice using pre-learned characteristics of the parent's voice.

[0750] Step 6:

[0751] The device plays this generated audio back to the child in real time. This allows the child to experience a conversation that reflects the parent's emotions.

[0752] Step 7:

[0753] If the conversation continues, return to step 3 and repeat the same process with the new input.

[0754] Step 8:

[0755] The server logs all conversation data and emotional states. This saved data can be reviewed later by parents to help improve communication with their children.

[0756] (Example 2)

[0757] 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".

[0758] In conversations with children, mechanical responses that disregard the parent's emotional state are often produced, leading to a problem where children do not feel adequately cared for. In particular, when a parent's emotions do not translate into appropriate responses to the child's needs or situation, parent-child communication may become weak. Therefore, the challenge is to achieve natural dialogue that takes the parent's emotions into consideration.

[0759] 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.

[0760] In this invention, the server includes emotion recognition means for analyzing the parent's emotional state and reflecting it in the response, speech synthesis means for generating arbitrary responses that resemble the parent's speech characteristics based on recorded voice data and learned feature data, and data management means for storing conversation history and related data for later analysis. This makes it possible to provide children with personalized and interactive responses that take the parent's emotions into consideration.

[0761] "Data acquisition means" refers to a device or algorithm for acquiring and recording the voice of a parent.

[0762] "Analysis means" refers to a device or program for extracting and learning speech features in digital format from acquired speech data.

[0763] A "speech synthesis means" is a device or algorithm for generating arbitrary responses that resemble the speech features of the parent, based on learned feature data.

[0764] "Output control means" refers to a device or program for outputting generated audio to a child under selected conditions.

[0765] "Input processing means" refers to a device or program for receiving inquiries from children in voice or text format and generating appropriate responses.

[0766] An "emotion recognition tool" is a device or algorithm that analyzes a parent's emotional state and uses the results to inform the response.

[0767] "Data management means" refers to a device or program for properly storing conversation history and related data so that it can be analyzed later.

[0768] "Speech recognition means" refers to a device or software for converting speech input into text data.

[0769] "Emotional analysis means" refers to a device or program that analyzes emotional states in real time and adjusts responses so that the generated voice can mimic the emotional state of the parent.

[0770] This system takes into account the parent's emotional state and is configured to generate natural and humane responses in interactions with children. Its embodiments are described in detail below.

[0771] The user (parent) records their voice using a dedicated software application. This recorded voice is processed as hardware by a mobile device such as a smartphone or tablet and uploaded to a server via a data acquisition method. Secure communication protocols such as HTTPS are used to ensure the security of the communication.

[0772] The server uses audio signal processing libraries such as "OpenSMILE" to extract audio features such as pitch, energy, and periodicity from the acquired audio data. Next, it uses emotion recognition algorithms such as "DeepMoji" to analyze the emotional state contained in the audio data. This information allows the system to understand the parent's current emotions in real time.

[0773] The child inputs their situation and feelings into the device via voice or text. The device receives this input and sends it to a server running a natural language generation model such as "GPT-3," which generates an appropriate response tailored to the parent's emotional state. In this process, the prompt sentence is constructed for the generation AI model. For example, a prompt sentence might be set to, "How would you respond if the parent's mood is happy and the child says they have made a new friend?"

[0774] The generated responses are converted into speech using speech synthesis technology such as "Google Text-to-Speech." By mimicking the parent's voice characteristics, the child can experience a conversation that feels as if they are actually talking to their parent. The device then plays the synthesized speech, providing the child with responses that reflect the parent's emotions.

[0775] Furthermore, the server saves all conversation data and associated sentiment data as logs. This allows users to review this data later and use it to reflect on and improve their communication with their children.

[0776] This system aims to facilitate smooth communication between parents and children by generating personalized and interactive responses that take into account the parent's feelings.

[0777] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0778] Step 1:

[0779] The user (parent) records their voice using a dedicated application. At this stage, the input is the parent's voice, and the output is a digital audio file. The recorded audio file is stored on the device and sent to the server by a data acquisition means. Specifically, the device uploads the file to the server using a secure protocol (e.g., HTTPS) via Wi-Fi or mobile data communication.

[0780] Step 2:

[0781] The server uses an audio signal processing library such as "OpenSMILE" to analyze the received audio data. The input is the previously uploaded audio file, and the output is extracted audio feature data (pitch, energy, periodicity, etc.). Specifically, the server passes the audio data to an analysis algorithm, generates feature vectors, captures changes in each time frame, and records them as features.

[0782] Step 3:

[0783] The server uses emotion recognition algorithms such as "DeepMoji" to determine the parent's emotional state based on voice feature data. The input is voice feature data, and the output is a specific emotion category (e.g., happiness, anger, surprise). In this process, the server inputs the feature data into a neural network model, calculates the probability of each emotion category, and determines the most likely emotional state.

[0784] Step 4:

[0785] Children input voice or text through a dedicated terminal application. The input is the text or voice entered by the child, and the output is text data sent to the server. Specifically, the child's voice is converted to text by the speech recognition function on the terminal and sent to the server as input.

[0786] Step 5:

[0787] The server generates a response using a generative AI model such as "GPT-3" based on the child's text input it receives. In this case, the input is the child's text and the parent's emotional state, and the output is the generated response sentence. Specifically, the server provides the input sentence as a prompt sentence to the generative AI model and obtains a text response that takes the parent's emotional state into account. For example, a prompt such as "How would you respond if the parent's emotional state is calm and the child says they have made a new friend?" might be used.

[0788] Step 6:

[0789] The server converts the generated response into speech using a text-to-speech engine (e.g., "Google Text-to-Speech"). The input is the generated response sentence, and the output is the audio data. Specifically, the synthesized speech is parameterized to mimic the parent's voice and encoded as an audio file.

[0790] Step 7:

[0791] The device receives the synthesized speech and plays it back to the child through the speaker. At this stage, the input is the audio data received from the server, and the output is the speech as physical sound. Specifically, the device loads the audio file into a buffer and plays it back through the speaker at a normal volume using playback software.

[0792] Step 8:

[0793] The server records all parent-child dialogue data and associated emotional information in a database, retaining it for later user access. The input is all data generated during the conversation, and the output is a structured log dataset. Specifically, it stores timestamps, text, voice features, emotional states, etc., for each dialogue session as database records, providing a basis for future analysis.

[0794] (Application Example 2)

[0795] 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".

[0796] It is difficult for parents and store staff to automatically generate responses that accurately reflect their emotional state when interacting with children or customers, making it challenging to maintain natural and human-centered communication. Therefore, there is a need for improved communication in close relationships and enhanced customer satisfaction in store services. Furthermore, there is a demand for systems that can support appropriate responses based on the staff's emotional state during customer service.

[0797] 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.

[0798] In this invention, the server includes voice acquisition means for recording the voice of a parent or staff member; voice learning means for analyzing the voice of a parent or staff member and learning its characteristics as digital data; and voice generation means for generating appropriate responses that resemble the voice of a parent or staff member based on the recorded voice and learned characteristic data. This makes it possible to provide a system that enables appropriate responses according to the emotional state of the parent or staff member, and facilitates natural and humane communication with children and customers.

[0799] "Voice acquisition means" refers to devices or technologies for recording the voices of parents or staff.

[0800] "Speech learning methods" refer to technologies that analyze acquired speech and learn its characteristics as digital data.

[0801] "Voice generation means" refers to a technology that generates appropriate responses in a manner similar to the voice of a parent or staff member, based on recorded voice and learned feature data.

[0802] "Output means" refers to a device or method for providing the generated audio to a child or customer.

[0803] "Input processing means" refers to technology that receives input from a child or customer as voice or text and generates a response.

[0804] "Data management methods" refer to technologies for saving and analyzing conversation logs.

[0805] "Emotional adjustment techniques" refer to methods that suggest appropriate responses during customer service based on the emotional state of the staff member.

[0806] "Speech recognition means" refers to technology for converting speech into text data.

[0807] "Emotional analysis methods" are methods of analyzing emotions so that the generated voice can mimic the emotional state of a parent or staff member.

[0808] The system for implementing this invention is initiated when a parent or staff member records speech using a speech acquisition means and uploads the speech data to a server via a terminal. The server analyzes the speech using a speech learning means and extracts its features as digital data. In this process, the pitch, tone, and speed of the speech are taken into consideration.

[0809] The server uses emotion adjustment mechanisms to recognize the emotional state of staff and generate real-time, adjustable responses. At this time, a generative AI model is utilized to process the audio or text received by the input processing mechanism from the child or customer. The voice generation mechanism then mimics the voices of parents or staff to create the optimal response. The output mechanism provides the generated audio to the child or customer, enabling natural and human-like dialogue.

[0810] For example, if a user is wearing smart glasses, the system uses voice recognition to transcribe customer requests into text and suggests products based on that content. The server uses data management to save conversation logs and organizes the analytical data for later access by the user.

[0811] A concrete example of a prompt would be, "Think of the best way to interact with a customer in a relaxed manner. For example, how can you make a friendly suggestion when recommending a new product?" This prompt is used to provide guidance for the generative AI model to generate responses that improve the quality of customer service.

[0812] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0813] Step 1:

[0814] The user records the voice of a parent or staff member using an audio acquisition device and uploads the audio data to the server via the terminal. The input is audio data, and the output is digital audio data stored on the server. The server performs format conversion to prepare this data.

[0815] Step 2:

[0816] The server uses speech learning methods to analyze uploaded audio data and extract speech features. Here, it generates speech feature vectors using Mel-frequency cepstrum coefficients (MFCCs) and speech pitch, among other things. The input is digital audio data, and the output is a speech feature vector.

[0817] Step 3:

[0818] The server uses emotion adjustment mechanisms to analyze emotional states from speech feature vectors. In this process, it determines an emotional category based on each feature, classifying them into categories such as relaxation or stress. The input is a speech feature vector, and the output is emotional state information.

[0819] Step 4:

[0820] The terminal receives input from a child or customer. The input processing device converts the speech into text data. The input is customer speech data, and the output is text data. The terminal uses speech recognition software to perform this conversion.

[0821] Step 5:

[0822] The server uses a generative AI model to generate the optimal response from emotional state information and text data. The generated response is output as text containing expressions appropriate to the emotion. The AI ​​model used here is a pre-trained natural language generation model.

[0823] Step 6:

[0824] The server uses speech generation means to convert the generated text responses into speech. In this process, speech synthesis technology is used to mimic the voices of parents or staff, and to generate speech with emotionally appropriate tones. The input is text responses, and the output is synthesized speech data.

[0825] Step 7:

[0826] The device uses an output mechanism to provide the generated synthesized voice to the child or customer. Through this voice, the user can experience natural and human-like dialogue. The output is delivered via a speaker or headphones.

[0827] This series of steps enables the system to engage in natural conversations that respond to the emotions of parents and staff, thereby improving efficiency in real-world customer service situations.

[0828] 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.

[0829] 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.

[0830] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0831] 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.

[0832] 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.

[0833] 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.

[0834] 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.

[0835] 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.

[0836] 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."

[0837] 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.

[0838] 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.

[0839] 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.

[0840] 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.

[0841] 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.

[0842] 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.

[0843] 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.

[0844] 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.

[0845] 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.

[0846] 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.

[0847] 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.

[0848] 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 as being incorporated by reference.

[0849] The following is further disclosed regarding the embodiments described above.

[0850] (Claim 1)

[0851] A means of acquiring voice for recording the voice of the parent,

[0852] A voice learning method that analyzes the parent's voice and learns its characteristics as digital data,

[0853] A voice generation means for generating arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature data,

[0854] An output means for outputting the generated audio to a child,

[0855] An input processing means for receiving input from a child as voice or text and generating a response,

[0856] A data management system for saving and analyzing conversation logs,

[0857] A system that includes this.

[0858] (Claim 2)

[0859] The system according to claim 1, comprising a speech recognition means that converts a child's voice input into text data using speech recognition.

[0860] (Claim 3)

[0861] The system according to claim 1, which performs emotion analysis to mimic the emotional state of the parent in the generated voice.

[0862] "Example 1"

[0863] (Claim 1)

[0864] A means of acquiring voice for recording the voice of a parent,

[0865] A voice learning method that analyzes the voice characteristics of parents, including pitch, speed, and intonation, and learns them as digital information,

[0866] A voice generation means that generates arbitrary responses that resemble the parent's voice using a generative AI model based on recorded audio and learned feature information,

[0867] A means of playing back the sound for outputting the generated audio to a child,

[0868] An input processing means for receiving input from a child as voice or text and generating a response,

[0869] A speech recognition means for converting a child's voice input into text and generating a response,

[0870] A conversation management system for recording all conversations and analyzing them later,

[0871] A system that includes this.

[0872] (Claim 2)

[0873] The system according to claim 1, which provides a more natural response by performing emotion analysis on the generated voice to mimic the emotional state of the parent.

[0874] (Claim 3)

[0875] The system according to claim 1, wherein the generated response is output to the child in real time, and the conversation process is recorded for future analysis.

[0876] "Application Example 1"

[0877] (Claim 1)

[0878] A means of acquiring voice for recording the voice of the parent,

[0879] A voice learning method that analyzes the parent's voice and learns its characteristics as digital data,

[0880] A voice generation means for generating arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature data,

[0881] An output means for outputting the generated audio to a child,

[0882] An input processing means for receiving input from a child as voice or text and generating a response,

[0883] A data management system for saving and analyzing conversation logs,

[0884] Information provision means for providing safety information and emergency response methods,

[0885] A system that includes this.

[0886] (Claim 2)

[0887] The system according to claim 1, comprising a speech recognition means that converts a child's voice input into text data using speech recognition.

[0888] (Claim 3)

[0889] The system according to claim 1, which performs emotion analysis to mimic the emotional state of the parent in the generated voice.

[0890] "Example 2 of combining an emotion engine"

[0891] (Claim 1)

[0892] A data acquisition method for acquiring and recording the voice of the parent,

[0893] An analysis method for analyzing acquired audio data and learning audio features in digital format,

[0894] A speech synthesis means for generating arbitrary responses that resemble the parent's speech features based on recorded speech data and learned feature data,

[0895] Output control means for outputting the generated audio to the child,

[0896] An input processing means for receiving inquiries from children in voice or text format and generating appropriate responses,

[0897] A means of recognizing emotions to analyze the emotional state of parents and reflect it in responses,

[0898] A data management system for saving conversation history and related data, and for making it analyzable later.

[0899] A system that includes this.

[0900] (Claim 2)

[0901] The system according to claim 1, comprising a speech recognition means that converts a child's voice input into text data using speech recognition.

[0902] (Claim 3)

[0903] The system according to claim 1, which includes an emotion analysis means for analyzing and adjusting the emotional state in real time so that the generated voice can mimic the emotional state of the parent.

[0904] "Application example 2 when combining with an emotional engine"

[0905] (Claim 1)

[0906] A means for recording the voice of a parent or staff member,

[0907] A voice learning method that analyzes the voices of parents or staff and learns their characteristics as digital data,

[0908] A voice generation means for generating an appropriate response that resembles the voice of a parent or staff member, based on recorded voice and learned feature data,

[0909] An output means for outputting the generated audio to a child or customer,

[0910] An input processing means for receiving input from a child or customer as voice or text and generating a response,

[0911] A data management system for saving and analyzing conversation logs,

[0912] An emotional adjustment mechanism that suggests appropriate responses during customer service based on the emotional state of the staff,

[0913] A system that includes this.

[0914] (Claim 2)

[0915] The system according to claim 1, comprising a speech recognition means that converts voice input from a child or customer into text data.

[0916] (Claim 3)

[0917] The system according to claim 1, comprising sentiment analysis means that performs sentiment analysis in order for the generated voice to mimic the emotional state of a parent or staff member. [Explanation of Symbols]

[0918] 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. A means of acquiring voice for recording the voice of the parent, A voice learning method that analyzes the parent's voice and learns its characteristics as digital data, A voice generation means for generating arbitrary responses that resemble the parent's voice, based on recorded voice and learned feature data, An output means for outputting the generated audio to a child, An input processing means for receiving input from a child as voice or text and generating a response, A data management system for saving and analyzing conversation logs, A system that includes this.

2. The system according to claim 1, comprising a speech recognition means that converts a child's voice input into text data using speech recognition.

3. The system according to claim 1, which performs emotion analysis to make the generated voice mimic the emotional state of the parent.

Images