system

Abstract

We provide the system. [Solution] A device means for acquiring voice input and converting the voice into text data, A communication module means for sending the converted text data to a server, A server receives text data and a processing device that generates an appropriate response using a generative artificial intelligence model. A device means for converting the generated response into audio data and presenting it through an audio output device, A system that includes a learning algorithm that improves response content based on user feedback.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Modern household appliances are becoming more and more multifunctional, and their operation procedures and troubleshooting are becoming increasingly complicated. Along with this, there is a problem that it is difficult for general users to obtain appropriate information by using thick operation manuals or complicated FAQs. In addition, due to the lack of support means that users can intuitively and quickly access for product operation and problem solving, efficient product use is impaired.

Means for Solving the Problems

[0005] To solve the above-mentioned problems, the present invention provides a device for acquiring voice input and converting the voice into text data. This device includes a communication module that transmits the text data from the device to a server, and the server uses the received text data to generate an appropriate response using a generative artificial intelligence model. Furthermore, by including a device for converting the generated response into voice data and presenting it, and a learning algorithm that improves the response content based on user feedback, the device is configured to allow users to receive intuitive and efficient support regarding home appliances.

[0006] "Voice input" refers to the acoustic signal obtained by a microphone or other acoustic device from the user's voice.

[0007] "Text data" refers to a collection of characters and symbols converted from voice input, and is information in a form that can be processed electronically.

[0008] A "communication module" refers to a function or device that enables the transmission and reception of text data and other information between different devices via digital communication.

[0009] A "server" refers to a computer system equipped with a central processing unit that provides services such as data processing, storage, and responding to requests from other devices.

[0010] A "generative artificial intelligence model" refers to an artificial intelligence algorithm or system that learns from data based on a specific task and generates appropriate responses or predictions.

[0011] A "response" refers to the information or instructions generated and provided in response to inquiries or instructions from the user.

[0012] "Audio data" refers to human voices and sounds obtained by converting text data back into acoustic signals.

[0013] "Audio output device" refers to speakers and other audio equipment that physically reproduce audio data.

[0014] A "learning algorithm" refers to a computational procedure or algorithm that uses past data and feedback information to improve the performance of a system. [Brief explanation of the drawing]

[0015] [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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This 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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Modes for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0019] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0020] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0021] 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).

[0022] 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."

[0023] [First Embodiment]

[0024] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0033] 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.

[0034] 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.

[0035] 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".

[0036] This invention provides a system for interpreting user intent from voice input and providing operation and support for home appliances based on that intent. Specifically, it is implemented in the following form.

[0037] Acquiring and converting voice input:

[0038] The process begins with the user voice-initiating questions or instructions about home appliances. The device is equipped with a high-performance microphone to capture this voice. The device then uses speech recognition technology to convert this voice into text data.

[0039] Data transmission and processing:

[0040] The terminal sends the converted text data to the server. This process is carried out using a secure communication protocol, ensuring data confidentiality. The server analyzes the received text and selects an appropriate generative artificial intelligence model. This AI model is pre-tuned and has learned from consumer electronics instruction manuals and FAQs.

[0041] Response generation and presentation:

[0042] The server inputs text data into an AI model and generates an appropriate response. This response answers the user's question quickly and accurately. For example, in response to a command such as "Turn on the air conditioner," it generates a response like "Turning on the air conditioner with the current settings." The server sends this response back to the terminal, which then uses a text-to-speech (TTS) engine to play the response aloud.

[0043] Learning and personalization:

[0044] By collecting user feedback and usage patterns, the server personalizes the AI ​​model. This allows the system to improve the quality of responses based on user preferences and past behavior. In addition, if the server cannot obtain an appropriate answer to a user's inquiry, it automatically connects the user to the manufacturer's support center. Through this series of processes, the present invention can provide users with intuitive and efficient support.

[0045] For example, if a user asks, "My washing machine is showing error code E04. What should I do?", the device converts this voice into text and sends it to the server. The server's AI model searches for information related to error code E04, generates specific instructions such as "Clean the filter," and communicates this to the user verbally. Through this process, the user can quickly resolve the problem.

[0046] The following describes the processing flow.

[0047] Step 1:

[0048] The user speaks into the device with questions or instructions about home appliances. This voice is captured by the device's microphone.

[0049] Step 2:

[0050] The device processes the captured audio data using a speech recognition engine and converts it into text data. This text data provides the foundational information for interpreting the user's intent.

[0051] Step 3:

[0052] The terminal sends the converted text data to the server. This communication uses a secure protocol to protect the confidentiality of the data.

[0053] Step 4:

[0054] The server analyzes the received text data and selects an appropriate generative artificial intelligence model. This model is pre-tuned based on specific consumer electronics product information and FAQs.

[0055] Step 5:

[0056] The server inputs text data into the AI ​​model and generates a suitable response. This response is designed to provide the most appropriate answer to the user's question.

[0057] Step 6:

[0058] The server sends the generated response to the terminal in text format. This response is intended to answer the user based on the original question.

[0059] Step 7:

[0060] The terminal converts the received text response into speech data using a text-to-speech (TTS) engine. This conversion is necessary for the user to receive the response in voice.

[0061] Step 8:

[0062] The device plays the generated audio data through its speaker, providing the user with a verbal response. This allows the user to intuitively receive instructions and information.

[0063] Step 9:

[0064] The device collects user feedback and behavioral data and sends it to the server. The server uses this data to personalize and improve the AI ​​model.

[0065] Step 10:

[0066] If the server cannot find a valid answer to the user's inquiry, it will automatically attempt to connect with the manufacturer's support center following a pre-configured procedure. The progress of this connection will be notified to the user through their device.

[0067] (Example 1)

[0068] 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."

[0069] Conventional voice recognition systems have struggled to provide quick and accurate responses to control electrical appliances and meet specific user needs. Furthermore, they have inadequate handling of situations where commands are not recognized or incorrect responses are generated. As a result, user convenience is often compromised.

[0070] 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.

[0071] In this invention, the server includes a speech recognition unit means that acquires an audio signal and converts the audio signal into character data; a processing unit means in which a computing device receives the character data and generates a response to the operator's instructions using a trained generative model; and a communication protocol means that automatically connects to the support system of the instruction target if a suitable response is not found. This enables the provision of accurate and rapid responses to voice input, improving user convenience.

[0072] "Audio signal" refers to the electronic signal obtained by converting sound waves acquired by an input device such as a microphone.

[0073] "Character data" refers to data obtained by analyzing an audio signal using speech recognition technology and converting it into digital character format.

[0074] A "speech recognition unit" refers to the technology and equipment used to convert speech signals into text data.

[0075] A "communication module" refers to hardware or software used for data communication when transmitting character data to a computing device.

[0076] A "computational device" refers to a computer or server that receives character data and processes it to generate responses.

[0077] A "pre-trained generative model" refers to an artificial intelligence model that has been trained in advance based on a large amount of data and is designed to generate responses according to a specific task.

[0078] A "processing unit" refers to a component within a computing device that takes character data as input and generates a response using a pre-trained generative model.

[0079] "Audio output means" refers to devices or technologies that convert the generated response into an audio signal and present it to the user as sound.

[0080] A "playback unit" refers to a component that uses an audio output means to reproduce the generated audio signal.

[0081] A "learning algorithm" refers to a data processing method used to improve the response accuracy of a system based on user feedback and usage patterns.

[0082] A "communication protocol" refers to a set of rules and procedures used when exchanging data between a computing device and an external system.

[0083] "Personalization" refers to the process of adapting a system based on the operator's preferences and past behavior, thereby individualizing the content of its responses.

[0084] One embodiment of this invention is a system configured as follows.

[0085] Acquiring and converting voice input:

[0086] The process begins with the user giving instructions or asking questions about home appliances using voice commands. The terminal is equipped with a high-performance audio input device that effectively captures the user's voice. The voice signal is converted into digital text information by a processing program running on the terminal. This conversion can utilize commonly used speech recognition technologies such as Google® Speech-to-Text or Amazon's services.

[0087] Data transmission and processing:

[0088] The terminal transmits the converted character information to the computing device via the internet. The transmission is secured by the SSL / TLS protocol, thus maintaining data confidentiality. The server analyzes this character information, selects a pre-trained generative model, and uses it to generate a response. This pre-trained generative model is trained based on information from sources such as appliance specifications and frequently asked questions.

[0089] Response generation and presentation:

[0090] The server inputs textual information into a trained generative model and generates a response that matches the user's request. This response is fast and accurate, precisely meeting the user's needs. For example, in response to the instruction "Please turn on the air conditioner," the generated response might be output as "The air conditioner will start operating with the current settings." The server sends this response back to the terminal, where it is played back as speech using speech synthesis technology (TTS engine).

[0091] Personalization and learning:

[0092] The server accumulates user feedback and usage history, and uses this data to optimize responses. This is achieved through a learning algorithm, leading to personalization that adapts to user preferences and behavioral patterns. This process improves the accuracy of responses to the information users are seeking.

[0093] As a concrete example, consider a scenario where a user asks aloud, "My washing machine is showing error code E04. What should I do?" The terminal converts this audio into text and sends it to the server. The server's AI model searches for error code E04 and generates specific instructions such as "Please clean the filter," which it then communicates to the user via voice.

[0094] An example of a prompt message would be text such as, "Set the air conditioner temperature to 21 degrees."

[0095] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0096] Step 1:

[0097] Voice input from the user

[0098] The user provides instructions and questions about home appliances using voice. Natural speech is used as input. The terminal acquires this voice signal using a high-performance audio input device. Microphones and related voice capture technologies are used to convert the acquired voice signal into a digital format.

[0099] Step 2:

[0100] Speech-to-text conversion

[0101] The device takes the received audio signal as input and converts it into text data using speech recognition technology. Specifically, services such as Google Speech-to-Text and Amazon's are used. The data processing in this step involves converting the audio signal into text data, and algorithms are used for noise reduction and accuracy improvement. The output is text data.

[0102] Step 3:

[0103] Sending text data

[0104] The terminal sends the converted text data as output to the server. Input is text data, and output is data communication to the server. Transmission is performed using the SSL / TLS protocol to ensure data confidentiality.

[0105] Step 4:

[0106] Text analysis and response generation

[0107] The server receives character data and performs analysis using a trained generative model. The input is character data, which undergoes analysis based on the AI ​​model. The generative AI model receives a prompt and generates a response based on the user's request. The output is the response in generated text.

[0108] Step 5:

[0109] Sending the generated response

[0110] The server sends the generated response to the terminal. The input is the text response generated by the server, and the output is data communication to the terminal. Security is ensured by using SSL / TLS again as the communication protocol.

[0111] Step 6:

[0112] Text-to-speech conversion and presentation

[0113] The terminal receives the transmitted text response and converts it into an audio signal using text-to-speech (TTS) technology. The input is the generated text response, and the output is the audio signal to be presented to the user. Specifically, the terminal uses an audio output device to play the response aloud for the user.

[0114] (Application Example 1)

[0115] 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."

[0116] In modern brick-and-mortar stores, customers require significant support to understand and operate products, but maintaining sufficient staff to provide adequate assistance at all times presents a challenge. Furthermore, customers may not be able to quickly obtain specific product information, resulting in missed sales opportunities. Additionally, there is a need for smart devices that can provide intuitive and useful information to tech-savvy customers.

[0117] 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.

[0118] In this invention, the server includes a device means for acquiring voice information and converting the voice into text data, a transmission module means for transmitting the converted text data to the server, and means equipped with a function to connect to and provide information display devices for physical stores so that users can directly check the content of their questions in the physical store. This makes it possible for customers to receive necessary product information directly and quickly in the physical store while reducing the burden on staff and providing high-quality service.

[0119] "Audio information" refers to information obtained by acquiring sound waves as digital data.

[0120] "Text data" refers to data obtained by analyzing audio information and representing it as a string of characters.

[0121] "Device means" refers to hardware and software components designed and used to achieve a specific function.

[0122] A "transmission module" refers to a means of communication used to send digital data to other computer devices or servers.

[0123] A "generative intelligence system" is a system that uses artificial intelligence technology to generate appropriate responses and information from received data.

[0124] "Audio output data" refers to digital audio data that has been converted to convey information as audio that can be understood by the listener.

[0125] An "audio playback device" refers to a hardware device that outputs digital audio data as physical sound.

[0126] A "user" is an individual or group that operates a system or device and receives its functions or services.

[0127] "Feedback" refers to responses and opinions provided based on the results of using the system, and is information that helps in future improvements.

[0128] A "learning method" refers to an algorithm or process that allows a system to learn from past data and usage patterns to improve its performance.

[0129] A "physical store information display device" refers to a device that includes displays and audio devices for providing product information to customers in a physical store.

[0130] This invention is a system for providing smooth information when users ask questions or perform operations related to products via voice in a physical store. This system combines voice recognition technology and a generative AI model to provide users with quick and accurate responses.

[0131] The system first uses a voice input device equipped with a high-performance microphone to capture the user's speech. This voice information is then converted into text data by speech recognition software (e.g., Google Speech-to-Text) on the device.

[0132] The converted text data is sent to the server in real time using a secure communication protocol. The server utilizes a generative intelligence system (e.g., an AI model on Amazon Web Services) to generate the optimal response from the received text data. This response generation uses a pre-trained model and is particularly suited to datasets appropriate for product operation and information provision in physical stores.

[0133] The generated responses are converted into audio output data and presented to the user via an audio playback device installed in the store. This allows users to instantly obtain the information they need without being affected by other customers or the store environment.

[0134] Furthermore, the system incorporates learning methods based on user feedback and experience, which continuously improves the accuracy and quality of responses. For example, if a customer asks, "How do I use a specific cooking mode on this microwave?", the system will provide detailed instructions that best suit the user's request. An example of a related prompt would be, "Explain the optimal operating procedure for the cooking mode the customer is asking about, based on the specifications."

[0135] Through this series of processes, the present invention provides a powerful support tool to improve the user experience and enable customers to get the most out of the product.

[0136] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0137] Step 1:

[0138] The device acquires user voice information using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is the first thing needed as preparation for speech recognition.

[0139] Step 2:

[0140] The device converts the acquired digital audio data into text data using speech recognition software (e.g., Google Speech-to-Text). During this process, the audio waveform is analyzed and converted into corresponding text. The output is text data representing the user's voice.

[0141] Step 3:

[0142] The terminal sends the converted text data to the server. Here, the input is the text data on the terminal, and the output is the data received on the server side. This transmission is carried out through a secure communication protocol, ensuring the confidentiality of the data.

[0143] Step 4:

[0144] The server processes the received text data and uses a generative intelligence system to generate the optimal response. The input is transcribed conversational data, and the output is an appropriate response text that meets the user's request. In this process, data calculations are performed using a pre-trained model, and specific answers are derived through the use of prompts.

[0145] Step 5:

[0146] The server converts the generated response text into data for speech output and sends it back to the terminal. The input is the response text generated by the server, and the output is data for speech synthesis. Text-to-speech technology is used for this conversion.

[0147] Step 6:

[0148] The terminal receives data for voice output and presents a response to the user via a voice playback device. The input is synthesized speech data, and the output is physical voice playback. This allows the user to obtain the information or instructions they are looking for in voice.

[0149] 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.

[0150] This invention provides a system that operates home appliances using user voice input, and in addition, detects and considers the user's emotional state. Specific embodiments for carrying out this invention are described below.

[0151] Voice input acquisition and emotion recognition:

[0152] Users ask questions or give instructions about home appliances by voice to the device. This voice is captured by the device's high-performance microphone. The device activates its voice recognition engine and converts the captured voice data into text data, while simultaneously using its emotion engine to identify the user's emotions from the voice.

[0153] Data processing:

[0154] The terminal sends the converted text data and emotional state to the server. The server receives the text data and analyzes it along with the emotional identification results. Within the server, a generative artificial intelligence model is executed to generate the optimal response, taking into account the user's emotional state. This generation process involves adjusting the response based on the user's emotions.

[0155] Response generation and feedback:

[0156] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data and present it to the user as voice. This allows the user to quickly obtain information on the spot. Furthermore, the terminal collects user reactions and other feedback and sends it to the server to improve the accuracy of the AI ​​model and emotion engine.

[0157] Specific example:

[0158] For example, if a user grumpily says, "Turn on the air conditioner," the device captures this audio and converts it to text, while its emotion engine identifies the emotion as "grumpy." Based on this information, the server generates a response in a gentle tone, such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The device then plays this response back as audio. If the user requests further assistance and the system detects that their emotional state is unstable, it can also be configured to offer additional support options.

[0159] In the form described above, this invention can provide meticulous service that responds to the user's emotions and improve the user experience.

[0160] The following describes the processing flow.

[0161] Step 1:

[0162] The user gives voice commands or asks questions to the home appliance. These voice commands are captured by a microphone built into the device.

[0163] Step 2:

[0164] The device processes the captured audio data in two ways. First, a speech recognition engine converts the audio into text data. Second, an emotion engine identifies the user's emotions in real time from the audio data.

[0165] Step 3:

[0166] The device sends the converted text data and identified sentiment data to the server. This communication is encrypted using a secure protocol to ensure data security.

[0167] Step 4:

[0168] The server analyzes the received text data and sentiment data. Based on the text data, it selects the optimal generative artificial intelligence model and generates a response that takes into account adjustments based on the sentiment data.

[0169] Step 5:

[0170] The server sends back a response in text format to the terminal. This response includes content that reflects the user's emotional state, enabling more appropriate communication.

[0171] Step 6:

[0172] The text response received by the terminal is converted into audio data using a text-to-speech (TTS) engine. This audio is immediately presented to the user through the terminal's speaker.

[0173] Step 7:

[0174] The device collects responses and feedback to the user's actions. For example, it can make further sentiment inferences based on voice tone and additional instructions.

[0175] Step 8:

[0176] The device sends the collected feedback to the server, which uses it to improve the AI ​​model and enhance the accuracy of the emotion engine. If the response is unclear, it attempts to connect the user to the support center to improve the user's support experience.

[0177] (Example 2)

[0178] 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".

[0179] Conventional voice control systems have limited the user experience because they generate purely businesslike responses without considering the user's emotions. Furthermore, the system sometimes fails to respond appropriately when users make ambiguous requests regarding the operation of home appliances. In addition, there is a lack of means to improve the system's accuracy by utilizing user feedback in real time. As a result, it has been difficult to provide responses that are sensitive to the user's intentions and emotions.

[0180] 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.

[0181] In this invention, the server includes a device means for identifying the user's emotions from acquired voice; an information processing device means for receiving text information and emotion information from an external device and generating a response that takes the user's emotions into account using a generative artificial intelligence model; and a learning device means for improving the response content based on the user's reactions. This enables the generation of personalized responses that take the user's emotions into account and continuous improvement of the system's accuracy.

[0182] A "device for acquiring voice input" is a device that captures the voice emitted by the user and converts it into a digital signal for subsequent processing.

[0183] A "device that converts to text information" is a device that analyzes acquired voice input and converts it into text data as a corresponding string of characters.

[0184] A "device for identifying the user's emotions" is a device that identifies the user's emotional state by analyzing the characteristics of voice input.

[0185] A "communication device" is a device used to send and receive data to and from other devices or equipment via a network.

[0186] A "generative artificial intelligence model" is an algorithm or program trained to generate human-like responses or outputs based on input data.

[0187] An "information processing device" is a device that has the function of analyzing received data and performing appropriate processing or calculations as needed.

[0188] A "device that converts to audio data" is a device that has the function of converting text-based data into audible audio data using speech synthesis technology.

[0189] An "audio output device" is a device that reproduces audio data and transmits it to the user as sound.

[0190] A "learning device" is a device equipped with algorithms that continuously collect feedback from users and contribute to improving the system's performance.

[0191] An "information exchange method" is a set of protocols or guidelines for effectively and securely exchanging data and information between different systems.

[0192] This invention is a system that allows users to operate home appliances using voice commands and provides optimal responses by considering the emotional information detected during the operation. Specific embodiments for carrying out this invention are described below.

[0193] The user gives voice commands to the home appliances through the terminal. The terminal uses a high-performance microphone to capture this voice. The captured voice is converted into text using speech recognition technology, and at the same time, an emotion recognition device analyzes the user's emotions. This results in text data and emotional information.

[0194] Next, the terminal sends text data and emotional information to the server via a communication device. The server processes the received data using a generative AI model and generates an appropriate response that matches the user's emotions. This model is inputted with prompt sentences such as, "If the user says 'Turn on the air conditioner' in a dissatisfied voice, generate a situation-appropriate response in a soft tone."

[0195] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data, which is then presented to the user through an audio output device.

[0196] As a concrete example, consider a scenario where a user grumpily says, "Turn on the air conditioner." In this case, the emotion recognition device detects the emotion of "grumpiness," and the server's AI model generates a gentle response such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The terminal then plays this response as audio and conveys it to the user.

[0197] Furthermore, the device continuously collects user responses and sends them back to the server. The server uses this feedback to act as a learning device, improving the accuracy of its responses.

[0198] In this way, by considering both the user's voice commands and emotional state, a more refined user experience can be provided.

[0199] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0200] Step 1:

[0201] The user gives voice commands to home appliances. The terminal captures this voice in real time using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is sent to the processing unit while noise is reduced. Specifically, the user performs an action such as saying, "Turn on the air conditioner."

[0202] Step 2:

[0203] The terminal inputs the captured audio data into a speech recognition engine, which converts the audio data into text data. At this stage, spectral analysis and phoneme analysis of the audio signal are performed. The output is text data such as "Turn on the air conditioner." In parallel, the same audio data is sent to an emotion recognition device, which analyzes the intonation, speed, and volume of the voice to extract emotional information. The output is data indicating an emotional state, such as "unhappy."

[0204] Step 3:

[0205] The terminal transmits the converted text data and sentiment information to the server via a communication device. The input is the text and sentiment data processed and generated on the terminal, and the output is encrypted data packets that reach the server securely. Specifically, the data is packetized and transmitted using the HTTPS protocol.

[0206] Step 4:

[0207] The server uses a generative AI model to analyze the received text data and sentiment information. The input is the user's text data and sentiment data, and the output is an appropriate response text that takes the user's sentiment into account. The generative AI model generates responses to prompts such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." Natural language processing techniques are used in this process.

[0208] Step 5:

[0209] The response text generated by the server is sent to the terminal. The terminal uses a TTS (Text-to-Speech) engine to synthesize this text into speech and convert it into audio data. The input is the generated response text, and the output is the audio data. The terminal plays this audio to the user through its speaker; specifically, it plays the message, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings."

[0210] Step 6:

[0211] The terminal observes the user's responses and collects feedback as needed. If the user gives further instructions or performs a different action, that data is sent back to the server via the communication device. The input is the user's additional feedback and response data, and the output is improvement data sent to the server. In this way, the entire system improves accuracy through the user's continued use.

[0212] (Application Example 2)

[0213] 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".

[0214] In home appliances and information devices operated via voice input, there is a need for systems that can respond flexibly to the user's emotions. However, conventional systems are limited to converting voice commands to text and provide uniform responses without considering the user's emotional state, which is a problem as it fails to sufficiently increase user satisfaction. Furthermore, in physical stores, it is difficult for sales staff to immediately grasp the customer's emotions and provide appropriate service, which hinders the improvement of the customer experience.

[0215] 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.

[0216] In this invention, the server includes emotion recognition means for identifying the emotional state of voice input and adjusting the response; means for making adjustments before presenting a response based on the identified emotional state; and learning procedure means for improving the response content based on user feedback. This makes it possible to provide individualized responses that are appropriate to the user's emotions and to help improve the customer experience.

[0217] An "emotion recognition mechanism" is a system that identifies the user's emotional state from their voice and adjusts its response based on that information.

[0218] "Voice input" refers to a method by which users provide instructions or information to a system through their voice.

[0219] "Textual information" refers to a data format in which voice input is analyzed and expressed as words.

[0220] "Communication means" refers to a method or device for transmitting converted data to another device or server.

[0221] An "artificial intelligence model for generation" is an algorithm designed to generate an appropriate response based on input information.

[0222] A "sound output mechanism" is a device that presents generated sound data to the user visually or audibly.

[0223] A "learning procedure" is the process of optimizing a system using feedback to provide improved responses in the future.

[0224] This invention constructs the following system to realize interaction using voice input: The terminal uses a high-sensitivity microphone to acquire the user's voice input. The acquired voice is converted into text information by voice recognition software installed on the terminal (e.g., Google Cloud Speech-to-Text API). In addition, sentiment analysis software (e.g., IBM Watson® Tone Analyzer) determines the user's emotional state from the voice. This converted data and sentiment information are quickly transmitted to the server.

[0225] The server uses a generative AI model (e.g., OpenAI® GPT series) based on the received data to generate an emotionally appropriate response. The AI ​​model utilizes prompts that take the user's emotional state into account to generate the most suitable response for each situation. This response is then sent back to the terminal, converted into audio data using text-to-speech synthesis (e.g., Amazon Polly), and communicated to the user through an audio output mechanism.

[0226] For example, in a physical store, when a salesperson wearing smart glasses interacts with a customer, if the customer asks, "Do you have this product in a different color?", and the emotional analysis detects a slight state of anxiety, a response with a gentle tone such as, "Of course, I'd be happy to help," will be generated. In this way, the generated response is naturally suggested to the customer by the salesperson.

[0227] An example of a prompt for a generative AI model is: "The user asks 'AA' and is showing anxiety. Generate an appropriate response in a calming, gentle tone." This allows the system to provide the most appropriate and human-like interaction for the user.

[0228] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0229] Step 1:

[0230] The device acquires the user's voice input. A high-sensitivity microphone built into the device picks up the user's voice and prepares it for processing as digital audio data. This data is used as input for speech recognition.

[0231] Step 2:

[0232] The device launches speech recognition software and converts the audio data into text. The process of converting the audio data into unique text information is performed using the Google Cloud Speech-to-Text API. The output is text information.

[0233] Step 3:

[0234] The device identifies the emotional state from the voice data. Using IBM Watson Tone Analyzer, it analyzes the user's emotions within the voice data to identify their emotional state. This step outputs information indicating the emotional state.

[0235] Step 4:

[0236] The terminal sends the converted text information and sentiment information to the server. The information is transferred to the server via communication means, and the data is prepared for the next processing step.

[0237] Step 5:

[0238] The server generates a response using an AI model based on the received data. Using an AI model such as the OpenAI GPT series, it generates the optimal response based on the received textual and emotional information. The generated response text is obtained as output.

[0239] Step 6:

[0240] The server sends the generated response to the terminal. The response text is returned to the terminal, and preparations for the next speech synthesis are made. The server also performs error checking to ensure the accurate transmission of the response.

[0241] Step 7:

[0242] The device converts the response text into audio data. Using a text-to-speech engine such as Amazon Polly, it converts the generated response text into audio data and prepares it to be conveyed to the user in an easy-to-understand voice format.

[0243] Step 8:

[0244] The device presents a response to the user via its audio output mechanism. The generated audio data is played through the device's speaker, creating an environment where the user can hear the response.

[0245] Step 9:

[0246] The device collects user feedback and sends it to the server. It gathers information about user reactions and usage, and transmits it to the server via communication. This feedback is used to improve future responses.

[0247] 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.

[0248] 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.

[0249] 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.

[0250] [Second Embodiment]

[0251] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0252] 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.

[0253] 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).

[0254] 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.

[0255] 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.

[0256] 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).

[0257] 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.

[0258] 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.

[0259] 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.

[0260] 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.

[0261] 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.

[0262] 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".

[0263] This invention provides a system for interpreting user intent from voice input and providing operation and support for home appliances based on that intent. Specifically, it is implemented in the following form.

[0264] Acquiring and converting voice input:

[0265] The process begins with the user voice-initiating questions or instructions about home appliances. The device is equipped with a high-performance microphone to capture this voice. The device then uses speech recognition technology to convert this voice into text data.

[0266] Data transmission and processing:

[0267] The terminal sends the converted text data to the server. This process is carried out using a secure communication protocol, ensuring data confidentiality. The server analyzes the received text and selects an appropriate generative artificial intelligence model. This AI model is pre-tuned and has learned from consumer electronics instruction manuals and FAQs.

[0268] Response generation and presentation:

[0269] The server inputs text data into an AI model and generates an appropriate response. This response answers the user's question quickly and accurately. For example, in response to a command such as "Turn on the air conditioner," it generates a response like "Turning on the air conditioner with the current settings." The server sends this response back to the terminal, which then uses a text-to-speech (TTS) engine to play the response aloud.

[0270] Learning and personalization:

[0271] By collecting user feedback and usage patterns, the server personalizes the AI ​​model. This allows the system to improve the quality of responses based on user preferences and past behavior. In addition, if the server cannot obtain an appropriate answer to a user's inquiry, it automatically connects the user to the manufacturer's support center. Through this series of processes, the present invention can provide users with intuitive and efficient support.

[0272] For example, if a user asks, "My washing machine is showing error code E04. What should I do?", the device converts this voice into text and sends it to the server. The server's AI model searches for information related to error code E04, generates specific instructions such as "Clean the filter," and communicates this to the user verbally. Through this process, the user can quickly resolve the problem.

[0273] The processing flow will be described below.

[0274] Step 1:

[0275] The user issues a voice query or operation instruction regarding home appliances towards the terminal. This voice is captured by the terminal's microphone.

[0276] Step 2:

[0277] The terminal processes the captured voice data with a voice recognition engine and converts it into text data. This text data is the information that serves as the basis for interpreting the user's intention.

[0278] Step 3:

[0279] The terminal sends the converted text data towards the server. This communication is performed using a secure protocol, protecting the confidentiality of the data.

[0280] Step 4:

[0281] The server analyzes the received text data and selects an appropriate generative artificial intelligence model. This model has been pre-tuned based on specific home appliance product information and FAQs.

[0282] Step 5:

[0283] The server inputs the text data into the AI model and generates a corresponding response. This response is designed to provide the most appropriate answer to the user's question.

[0284] Step 6:

[0285] The server sends the generated response to the terminal in text format. This response is for replying to the user based on the original question.

[0286] Step 7:

[0287] The terminal converts the received text response into voice data using a text-to-speech (TTS) engine. This conversion is necessary for the user to receive the response in voice.

[0288] Step 8:

[0289] The terminal plays the generated voice data from the speaker and provides the response to the user in voice. As a result, the user can intuitively receive instructions and information.

[0290] Step 9:

[0291] The terminal collects the user's feedback and behavior data and sends this to the server. The server uses this data to personalize and improve the AI model.

[0292] Step 10:

[0293] If the server cannot find a valid answer to the user's query, the server attempts to automatically connect to the manufacturer's support center according to a preset procedure. The progress of this connection is notified to the user through the terminal.

[0294] (Example 1)

[0295] Next, Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0296] In a conventional voice recognition system, it was difficult to quickly and accurately provide responses based on the operation of electrical appliances and the specific needs of the user. Also, the response in case the operation was not recognized or an incorrect response was generated was insufficient. Therefore, the convenience of the user was often impaired.

[0297] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following respective means.

[0298] In this invention, the server includes a speech recognition unit means that acquires an audio signal and converts the audio signal into character data; a processing unit means in which a computing device receives the character data and generates a response to the operator's instructions using a trained generative model; and a communication protocol means that automatically connects to the support system of the instruction target if a suitable response is not found. This enables the provision of accurate and rapid responses to voice input, improving user convenience.

[0299] "Audio signal" refers to the electronic signal obtained by converting sound waves acquired by an input device such as a microphone.

[0300] "Character data" refers to data obtained by analyzing an audio signal using speech recognition technology and converting it into digital character format.

[0301] A "speech recognition unit" refers to the technology and equipment used to convert speech signals into text data.

[0302] A "communication module" refers to hardware or software used for data communication when transmitting character data to a computing device.

[0303] A "computational device" refers to a computer or server that receives character data and processes it to generate responses.

[0304] A "pre-trained generative model" refers to an artificial intelligence model that has been trained in advance based on a large amount of data and is designed to generate responses according to a specific task.

[0305] A "processing unit" refers to a component within a computing device that takes character data as input and generates a response using a pre-trained generative model.

[0306] "Audio output means" refers to devices or technologies that convert the generated response into an audio signal and present it to the user as sound.

[0307] The "reproduction unit" refers to a component for reproducing the generated audio signal using the audio output means.

[0308] The "learning algorithm" refers to a data processing method for improving the response accuracy of the system based on user feedback and usage patterns.

[0309] The "communication protocol" refers to a series of rules and procedures used when exchanging data between a computing device and an external system.

[0310] "Personalization" refers to the process of adapting the system based on the operator's preferences and past behaviors and individualizing the content of the response.

[0311] As an embodiment of this invention, a system configured as follows is provided.

[0312] Acquisition and conversion of voice input:

[0313] It starts when the user gives instructions or asks questions about home appliances in voice. The terminal is equipped with a high-performance acoustic input device, and this device effectively acquires the user's voice. The voice signal is converted into digital character information by an arithmetic program operating on the terminal. For this conversion, services such as Google Speech-to-Text and Amazon, which are commonly used as voice recognition technologies, can be applied.

[0314] Data transmission and processing:

[0315] The terminal transmits the converted character information to the computing device via the Internet. Since the transmission is secured by the SSL / TLS protocol, the confidentiality of the data is maintained. The server analyzes this character information, selects a learned generation model, and uses it for generating a response. This learned generation model is trained based on information from the specifications of electrical appliances and common question sets.

[0316] Response generation and presentation:

[0317] The server inputs textual information into a trained generative model and generates a response that matches the user's request. This response is fast and accurate, precisely meeting the user's needs. For example, in response to the instruction "Please turn on the air conditioner," the generated response might be output as "The air conditioner will start operating with the current settings." The server sends this response back to the terminal, where it is played back as speech using speech synthesis technology (TTS engine).

[0318] Personalization and learning:

[0319] The server accumulates user feedback and usage history, and uses this data to optimize responses. This is achieved through a learning algorithm, leading to personalization that adapts to user preferences and behavioral patterns. This process improves the accuracy of responses to the information users are seeking.

[0320] As a concrete example, consider a scenario where a user asks aloud, "My washing machine is showing error code E04. What should I do?" The terminal converts this audio into text and sends it to the server. The server's AI model searches for error code E04 and generates specific instructions such as "Please clean the filter," which it then communicates to the user via voice.

[0321] An example of a prompt message would be text such as, "Set the air conditioner temperature to 21 degrees."

[0322] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0323] Step 1:

[0324] Voice input from the user

[0325] The user provides instructions and questions about home appliances using voice. Natural speech is used as input. The terminal acquires this voice signal using a high-performance audio input device. Microphones and related voice capture technologies are used to convert the acquired voice signal into a digital format.

[0326] Step 2:

[0327] Speech-to-text conversion

[0328] The device takes the received audio signal as input and converts it into text data using speech recognition technology. Specifically, services such as Google Speech-to-Text and Amazon's are used. The data processing in this step involves converting the audio signal into text data, and algorithms are used for noise reduction and accuracy improvement. The output is text data.

[0329] Step 3:

[0330] Sending text data

[0331] The terminal sends the converted text data as output to the server. Input is text data, and output is data communication to the server. Transmission is performed using the SSL / TLS protocol to ensure data confidentiality.

[0332] Step 4:

[0333] Text analysis and response generation

[0334] The server receives character data and performs analysis using a trained generative model. The input is character data, which undergoes analysis based on the AI ​​model. The generative AI model receives a prompt and generates a response based on the user's request. The output is the response in generated text.

[0335] Step 5:

[0336] Sending the generated response

[0337] The server sends the generated response to the terminal. The input is the text response generated by the server, and the output is data communication to the terminal. Security is ensured by using SSL / TLS again as the communication protocol.

[0338] Step 6:

[0339] Text-to-speech conversion and presentation

[0340] The terminal receives the transmitted text response and converts it into an audio signal using text-to-speech (TTS) technology. The input is the generated text response, and the output is the audio signal to be presented to the user. Specifically, the terminal uses an audio output device to play the response aloud for the user.

[0341] (Application Example 1)

[0342] 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."

[0343] In modern brick-and-mortar stores, customers require significant support to understand and operate products, but maintaining sufficient staff to provide adequate assistance at all times presents a challenge. Furthermore, customers may not be able to quickly obtain specific product information, resulting in missed sales opportunities. Additionally, there is a need for smart devices that can provide intuitive and useful information to tech-savvy customers.

[0344] 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.

[0345] In this invention, the server includes a device means for acquiring voice information and converting the voice into text data, a transmission module means for transmitting the converted text data to the server, and means equipped with a function to connect to and provide information display devices for physical stores so that users can directly check the content of their questions in the physical store. This makes it possible for customers to receive necessary product information directly and quickly in the physical store while reducing the burden on staff and providing high-quality service.

[0346] "Audio information" refers to information obtained by acquiring sound waves as digital data.

[0347] "Text data" refers to data obtained by analyzing audio information and representing it as a string of characters.

[0348] "Device means" refers to hardware and software components designed and used to achieve a specific function.

[0349] A "transmission module" refers to a means of communication used to send digital data to other computer devices or servers.

[0350] A "generative intelligence system" is a system that uses artificial intelligence technology to generate appropriate responses and information from received data.

[0351] "Audio output data" refers to digital audio data that has been converted to convey information as audio that can be understood by the listener.

[0352] An "audio playback device" refers to a hardware device that outputs digital audio data as physical sound.

[0353] A "user" is an individual or group that operates a system or device and receives its functions or services.

[0354] "Feedback" refers to responses and opinions provided based on the results of using the system, and is information that helps in future improvements.

[0355] A "learning method" refers to an algorithm or process that allows a system to learn from past data and usage patterns to improve its performance.

[0356] A "physical store information display device" refers to a device that includes displays and audio devices for providing product information to customers in a physical store.

[0357] This invention is a system for providing smooth information when users ask questions or perform operations related to products via voice in a physical store. This system combines voice recognition technology and a generative AI model to provide users with quick and accurate responses.

[0358] The system first uses a voice input device equipped with a high-performance microphone to capture the user's speech. This voice information is then converted into text data by speech recognition software (e.g., Google Speech-to-Text) on the device.

[0359] The converted text data is sent to the server in real time using a secure communication protocol. The server utilizes a generative intelligence system (e.g., an AI model on Amazon Web Services) to generate the optimal response from the received text data. This response generation uses a pre-trained model and is particularly suited to datasets appropriate for product operation and information provision in physical stores.

[0360] The generated responses are converted into audio output data and presented to the user via an audio playback device installed in the store. This allows users to instantly obtain the information they need without being affected by other customers or the store environment.

[0361] Furthermore, the system incorporates learning methods based on user feedback and experience, which continuously improves the accuracy and quality of responses. For example, if a customer asks, "How do I use a specific cooking mode on this microwave?", the system will provide detailed instructions that best suit the user's request. An example of a related prompt would be, "Explain the optimal operating procedure for the cooking mode the customer is asking about, based on the specifications."

[0362] Through this series of processes, the present invention provides a powerful support tool to improve the user experience and enable customers to get the most out of the product.

[0363] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0364] Step 1:

[0365] The device acquires user voice information using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is the first thing needed as preparation for speech recognition.

[0366] Step 2:

[0367] The device converts the acquired digital audio data into text data using speech recognition software (e.g., Google Speech-to-Text). During this process, the audio waveform is analyzed and converted into corresponding text. The output is text data representing the user's voice.

[0368] Step 3:

[0369] The terminal sends the converted text data to the server. Here, the input is the text data on the terminal, and the output is the data received on the server side. This transmission is carried out through a secure communication protocol, ensuring the confidentiality of the data.

[0370] Step 4:

[0371] The server processes the received text data and uses a generative intelligence system to generate the optimal response. The input is transcribed conversational data, and the output is an appropriate response text that meets the user's request. In this process, data calculations are performed using a pre-trained model, and specific answers are derived through the use of prompts.

[0372] Step 5:

[0373] The server converts the generated response text into data for speech output and sends it back to the terminal. The input is the response text generated by the server, and the output is data for speech synthesis. Text-to-speech technology is used for this conversion.

[0374] Step 6:

[0375] The terminal receives data for voice output and presents a response to the user via a voice playback device. The input is synthesized speech data, and the output is physical voice playback. This allows the user to obtain the information or instructions they are looking for in voice.

[0376] 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.

[0377] This invention provides a system that operates home appliances using user voice input, and in addition, detects and considers the user's emotional state. Specific embodiments for carrying out this invention are described below.

[0378] Voice input acquisition and emotion recognition:

[0379] Users ask questions or give instructions about home appliances by voice to the device. This voice is captured by the device's high-performance microphone. The device activates its voice recognition engine and converts the captured voice data into text data, while simultaneously using its emotion engine to identify the user's emotions from the voice.

[0380] Data processing:

[0381] The terminal sends the converted text data and emotional state to the server. The server receives the text data and analyzes it along with the emotional identification results. Within the server, a generative artificial intelligence model is executed to generate the optimal response, taking into account the user's emotional state. This generation process involves adjusting the response based on the user's emotions.

[0382] Response generation and feedback:

[0383] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data and present it to the user as voice. This allows the user to quickly obtain information on the spot. Furthermore, the terminal collects user reactions and other feedback and sends it to the server to improve the accuracy of the AI ​​model and emotion engine.

[0384] Specific example:

[0385] For example, if a user grumpily says, "Turn on the air conditioner," the device captures this audio and converts it to text, while its emotion engine identifies the emotion as "grumpy." Based on this information, the server generates a response in a gentle tone, such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The device then plays this response back as audio. If the user requests further assistance and the system detects that their emotional state is unstable, it can also be configured to offer additional support options.

[0386] In the form described above, this invention can provide meticulous service that responds to the user's emotions and improve the user experience.

[0387] The following describes the processing flow.

[0388] Step 1:

[0389] The user gives voice commands or asks questions to the home appliance. These voice commands are captured by a microphone built into the device.

[0390] Step 2:

[0391] The device processes the captured audio data in two ways. First, a speech recognition engine converts the audio into text data. Second, an emotion engine identifies the user's emotions in real time from the audio data.

[0392] Step 3:

[0393] The device sends the converted text data and identified sentiment data to the server. This communication is encrypted using a secure protocol to ensure data security.

[0394] Step 4:

[0395] The server analyzes the received text data and sentiment data. Based on the text data, it selects the optimal generative artificial intelligence model and generates a response that takes into account adjustments based on the sentiment data.

[0396] Step 5:

[0397] The server sends back a response in text format to the terminal. This response includes content that reflects the user's emotional state, enabling more appropriate communication.

[0398] Step 6:

[0399] The text response received by the terminal is converted into audio data using a text-to-speech (TTS) engine. This audio is immediately presented to the user through the terminal's speaker.

[0400] Step 7:

[0401] The device collects responses and feedback to the user's actions. For example, it can make further sentiment inferences based on voice tone and additional instructions.

[0402] Step 8:

[0403] The device sends the collected feedback to the server, which uses it to improve the AI ​​model and enhance the accuracy of the emotion engine. If the response is unclear, it attempts to connect the user to the support center to improve the user's support experience.

[0404] (Example 2)

[0405] 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".

[0406] Conventional voice control systems have limited the user experience because they generate purely businesslike responses without considering the user's emotions. Furthermore, the system sometimes fails to respond appropriately when users make ambiguous requests regarding the operation of home appliances. In addition, there is a lack of means to improve the system's accuracy by utilizing user feedback in real time. As a result, it has been difficult to provide responses that are sensitive to the user's intentions and emotions.

[0407] 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.

[0408] In this invention, the server includes a device means for identifying the user's emotions from acquired voice; an information processing device means for receiving text information and emotion information from an external device and generating a response that takes the user's emotions into account using a generative artificial intelligence model; and a learning device means for improving the response content based on the user's reactions. This enables the generation of personalized responses that take the user's emotions into account and continuous improvement of the system's accuracy.

[0409] A "device for acquiring voice input" is a device that captures the voice emitted by the user and converts it into a digital signal for subsequent processing.

[0410] A "device that converts to text information" is a device that analyzes acquired voice input and converts it into text data as a corresponding string of characters.

[0411] A "device for identifying the user's emotions" is a device that identifies the user's emotional state by analyzing the characteristics of voice input.

[0412] A "communication device" is a device used to send and receive data to and from other devices or equipment via a network.

[0413] A "generative artificial intelligence model" is an algorithm or program trained to generate human-like responses or outputs based on input data.

[0414] An "information processing device" is a device that has the function of analyzing received data and performing appropriate processing or calculations as needed.

[0415] A "device that converts to audio data" is a device that has the function of converting text-based data into audible audio data using speech synthesis technology.

[0416] An "audio output device" is a device that reproduces audio data and transmits it to the user as sound.

[0417] A "learning device" is a device equipped with algorithms that continuously collect feedback from users and contribute to improving the system's performance.

[0418] An "information exchange method" is a set of protocols or guidelines for effectively and securely exchanging data and information between different systems.

[0419] This invention is a system that allows users to operate home appliances using voice commands and provides optimal responses by considering the emotional information detected during the operation. Specific embodiments for carrying out this invention are described below.

[0420] The user gives voice commands to the home appliances through the terminal. The terminal uses a high-performance microphone to capture this voice. The captured voice is converted into text using speech recognition technology, and at the same time, an emotion recognition device analyzes the user's emotions. This results in text data and emotional information.

[0421] Next, the terminal sends text data and emotional information to the server via a communication device. The server processes the received data using a generative AI model and generates an appropriate response that matches the user's emotions. This model is inputted with prompt sentences such as, "If the user says 'Turn on the air conditioner' in a dissatisfied voice, generate a situation-appropriate response in a soft tone."

[0422] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data, which is then presented to the user through an audio output device.

[0423] As a concrete example, consider a scenario where a user grumpily says, "Turn on the air conditioner." In this case, the emotion recognition device detects the emotion of "grumpiness," and the server's AI model generates a gentle response such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The terminal then plays this response as audio and conveys it to the user.

[0424] Furthermore, the device continuously collects user responses and sends them back to the server. The server uses this feedback to act as a learning device, improving the accuracy of its responses.

[0425] In this way, by considering both the user's voice commands and emotional state, a more refined user experience can be provided.

[0426] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0427] Step 1:

[0428] The user gives voice commands to home appliances. The terminal captures this voice in real time using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is sent to the processing unit while noise is reduced. Specifically, the user performs an action such as saying, "Turn on the air conditioner."

[0429] Step 2:

[0430] The terminal inputs the captured audio data into a speech recognition engine, which converts the audio data into text data. At this stage, spectral analysis and phoneme analysis of the audio signal are performed. The output is text data such as "Turn on the air conditioner." In parallel, the same audio data is sent to an emotion recognition device, which analyzes the intonation, speed, and volume of the voice to extract emotional information. The output is data indicating an emotional state, such as "unhappy."

[0431] Step 3:

[0432] The terminal transmits the converted text data and sentiment information to the server via a communication device. The input is the text and sentiment data processed and generated on the terminal, and the output is encrypted data packets that reach the server securely. Specifically, the data is packetized and transmitted using the HTTPS protocol.

[0433] Step 4:

[0434] The server uses a generative AI model to analyze the received text data and sentiment information. The input is the user's text data and sentiment data, and the output is an appropriate response text that takes the user's sentiment into account. The generative AI model generates responses to prompts such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." Natural language processing techniques are used in this process.

[0435] Step 5:

[0436] The response text generated by the server is sent to the terminal. The terminal uses a TTS (Text-to-Speech) engine to synthesize this text into speech and convert it into audio data. The input is the generated response text, and the output is the audio data. The terminal plays this audio to the user through its speaker; specifically, it plays the message, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings."

[0437] Step 6:

[0438] The terminal observes the user's responses and collects feedback as needed. If the user gives further instructions or performs a different action, that data is sent back to the server via the communication device. The input is the user's additional feedback and response data, and the output is improvement data sent to the server. In this way, the entire system improves accuracy through the user's continued use.

[0439] (Application Example 2)

[0440] 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."

[0441] In home appliances and information devices operated via voice input, there is a need for systems that can respond flexibly to the user's emotions. However, conventional systems are limited to converting voice commands to text and provide uniform responses without considering the user's emotional state, which is a problem as it fails to sufficiently increase user satisfaction. Furthermore, in physical stores, it is difficult for sales staff to immediately grasp the customer's emotions and provide appropriate service, which hinders the improvement of the customer experience.

[0442] 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.

[0443] In this invention, the server includes emotion recognition means for identifying the emotional state of voice input and adjusting the response; means for making adjustments before presenting a response based on the identified emotional state; and learning procedure means for improving the response content based on user feedback. This makes it possible to provide individualized responses that are appropriate to the user's emotions and to help improve the customer experience.

[0444] An "emotion recognition mechanism" is a system that identifies the user's emotional state from their voice and adjusts its response based on that information.

[0445] "Voice input" refers to a method by which users provide instructions or information to a system through their voice.

[0446] "Textual information" refers to a data format in which voice input is analyzed and expressed as words.

[0447] "Communication means" refers to a method or device for transmitting converted data to another device or server.

[0448] An "artificial intelligence model for generation" is an algorithm designed to generate an appropriate response based on input information.

[0449] A "sound output mechanism" is a device that presents generated sound data to the user visually or audibly.

[0450] A "learning procedure" is the process of optimizing a system using feedback to provide improved responses in the future.

[0451] This invention constructs the following system to realize interaction using voice input: The terminal uses a high-sensitivity microphone to acquire the user's voice input. The acquired voice is converted into text information by voice recognition software installed on the terminal (e.g., Google Cloud Speech-to-Text API). In addition, sentiment analysis software (e.g., IBM Watson Tone Analyzer) determines the user's emotional state from the voice. This converted data and sentiment information are quickly transmitted to the server.

[0452] The server uses a generative AI model (e.g., OpenAI GPT series) based on the received data to generate an emotionally appropriate response. The AI ​​model utilizes prompts that take the user's emotional state into account to generate the most suitable response for each situation. This response is then sent back to the terminal, converted into audio data using text-to-speech synthesis (e.g., Amazon Polly), and communicated to the user through an audio output mechanism.

[0453] For example, in a physical store, when a salesperson wearing smart glasses interacts with a customer, if the customer asks, "Do you have this product in a different color?", and the emotional analysis detects a slight state of anxiety, a response with a gentle tone such as, "Of course, I'd be happy to help," will be generated. In this way, the generated response is naturally suggested to the customer by the salesperson.

[0454] An example of a prompt for a generative AI model is: "The user asks 'AA' and is showing anxiety. Generate an appropriate response in a calming, gentle tone." This allows the system to provide the most appropriate and human-like interaction for the user.

[0455] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0456] Step 1:

[0457] The device acquires the user's voice input. A high-sensitivity microphone built into the device picks up the user's voice and prepares it for processing as digital audio data. This data is used as input for speech recognition.

[0458] Step 2:

[0459] The device launches speech recognition software and converts the audio data into text. The process of converting the audio data into unique text information is performed using the Google Cloud Speech-to-Text API. The output is text information.

[0460] Step 3:

[0461] The device identifies the emotional state from the voice data. Using IBM Watson Tone Analyzer, it analyzes the user's emotions within the voice data to identify their emotional state. This step outputs information indicating the emotional state.

[0462] Step 4:

[0463] The terminal sends the converted text information and sentiment information to the server. The information is transferred to the server via communication means, and the data is prepared for the next processing step.

[0464] Step 5:

[0465] The server generates a response using an AI model based on the received data. Using an AI model such as the OpenAI GPT series, it generates the optimal response based on the received textual and emotional information. The generated response text is obtained as output.

[0466] Step 6:

[0467] The server sends the generated response to the terminal. The response text is returned to the terminal, and preparations for the next speech synthesis are made. The server also performs error checking to ensure the accurate transmission of the response.

[0468] Step 7:

[0469] The device converts the response text into audio data. Using a text-to-speech engine such as Amazon Polly, it converts the generated response text into audio data and prepares it to be conveyed to the user in an easy-to-understand voice format.

[0470] Step 8:

[0471] The device presents a response to the user via its audio output mechanism. The generated audio data is played through the device's speaker, creating an environment where the user can hear the response.

[0472] Step 9:

[0473] The device collects user feedback and sends it to the server. It gathers information about user reactions and usage, and transmits it to the server via communication. This feedback is used to improve future responses.

[0474] 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.

[0475] 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.

[0476] 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.

[0477] [Third Embodiment]

[0478] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0479] 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.

[0480] 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).

[0481] 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.

[0482] 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.

[0483] 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).

[0484] 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.

[0485] 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.

[0486] 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.

[0487] 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.

[0488] 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.

[0489] 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".

[0490] This invention provides a system for interpreting user intent from voice input and providing operation and support for home appliances based on that intent. Specifically, it is implemented in the following form.

[0491] Acquiring and converting voice input:

[0492] The process begins with the user voice-initiating questions or instructions about home appliances. The device is equipped with a high-performance microphone to capture this voice. The device then uses speech recognition technology to convert this voice into text data.

[0493] Data transmission and processing:

[0494] The terminal sends the converted text data to the server. This process is carried out using a secure communication protocol, ensuring data confidentiality. The server analyzes the received text and selects an appropriate generative artificial intelligence model. This AI model is pre-tuned and has learned from consumer electronics instruction manuals and FAQs.

[0495] Response generation and presentation:

[0496] The server inputs text data into an AI model and generates an appropriate response. This response answers the user's question quickly and accurately. For example, in response to a command such as "Turn on the air conditioner," it generates a response like "Turning on the air conditioner with the current settings." The server sends this response back to the terminal, which then uses a text-to-speech (TTS) engine to play the response aloud.

[0497] Learning and personalization:

[0498] By collecting user feedback and usage patterns, the server personalizes the AI ​​model. This allows the system to improve the quality of responses based on user preferences and past behavior. In addition, if the server cannot obtain an appropriate answer to a user's inquiry, it automatically connects the user to the manufacturer's support center. Through this series of processes, the present invention can provide users with intuitive and efficient support.

[0499] For example, if a user asks, "My washing machine is showing error code E04. What should I do?", the device converts this voice into text and sends it to the server. The server's AI model searches for information related to error code E04, generates specific instructions such as "Clean the filter," and communicates this to the user verbally. Through this process, the user can quickly resolve the problem.

[0500] The following describes the processing flow.

[0501] Step 1:

[0502] The user speaks into the device with questions or instructions about home appliances. This voice is captured by the device's microphone.

[0503] Step 2:

[0504] The device processes the captured audio data using a speech recognition engine and converts it into text data. This text data provides the foundational information for interpreting the user's intent.

[0505] Step 3:

[0506] The terminal sends the converted text data to the server. This communication uses a secure protocol to protect the confidentiality of the data.

[0507] Step 4:

[0508] The server analyzes the received text data and selects an appropriate generative artificial intelligence model. This model is pre-tuned based on specific consumer electronics product information and FAQs.

[0509] Step 5:

[0510] The server inputs text data into the AI ​​model and generates a suitable response. This response is designed to provide the most appropriate answer to the user's question.

[0511] Step 6:

[0512] The server sends the generated response to the terminal in text format. This response is intended to answer the user based on the original question.

[0513] Step 7:

[0514] The terminal converts the received text response into speech data using a text-to-speech (TTS) engine. This conversion is necessary for the user to receive the response in voice.

[0515] Step 8:

[0516] The device plays the generated audio data through its speaker, providing the user with a verbal response. This allows the user to intuitively receive instructions and information.

[0517] Step 9:

[0518] The device collects user feedback and behavioral data and sends it to the server. The server uses this data to personalize and improve the AI ​​model.

[0519] Step 10:

[0520] If the server cannot find a valid answer to the user's inquiry, it will automatically attempt to connect with the manufacturer's support center following a pre-configured procedure. The progress of this connection will be notified to the user through their device.

[0521] (Example 1)

[0522] 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."

[0523] Conventional voice recognition systems have struggled to provide quick and accurate responses to control electrical appliances and meet specific user needs. Furthermore, they have inadequate handling of situations where commands are not recognized or incorrect responses are generated. As a result, user convenience is often compromised.

[0524] 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.

[0525] In this invention, the server includes a speech recognition unit means that acquires an audio signal and converts the audio signal into character data; a processing unit means in which a computing device receives the character data and generates a response to the operator's instructions using a trained generative model; and a communication protocol means that automatically connects to the support system of the instruction target if a suitable response is not found. This enables the provision of accurate and rapid responses to voice input, improving user convenience.

[0526] "Audio signal" refers to the electronic signal obtained by converting sound waves acquired by an input device such as a microphone.

[0527] "Character data" refers to data obtained by analyzing an audio signal using speech recognition technology and converting it into digital character format.

[0528] A "speech recognition unit" refers to the technology and equipment used to convert speech signals into text data.

[0529] A "communication module" refers to hardware or software used for data communication when transmitting character data to a computing device.

[0530] A "computational device" refers to a computer or server that receives character data and processes it to generate responses.

[0531] A "pre-trained generative model" refers to an artificial intelligence model that has been trained in advance based on a large amount of data and is designed to generate responses according to a specific task.

[0532] A "processing unit" refers to a component within a computing device that takes character data as input and generates a response using a pre-trained generative model.

[0533] "Audio output means" refers to devices or technologies that convert the generated response into an audio signal and present it to the user as sound.

[0534] A "playback unit" refers to a component that uses an audio output means to reproduce the generated audio signal.

[0535] A "learning algorithm" refers to a data processing method used to improve the response accuracy of a system based on user feedback and usage patterns.

[0536] A "communication protocol" refers to a set of rules and procedures used when exchanging data between a computing device and an external system.

[0537] "Personalization" refers to the process of adapting a system based on the operator's preferences and past behavior, thereby individualizing the content of its responses.

[0538] One embodiment of this invention is a system configured as follows.

[0539] Acquiring and converting voice input:

[0540] The process begins with the user giving instructions or asking questions about home appliances using voice commands. The terminal is equipped with a high-performance audio input device that effectively captures the user's voice. The voice signal is converted into digital text information by a computation program running on the terminal. Commonly used speech recognition technologies such as Google Speech-to-Text or Amazon's service can be applied to this conversion.

[0541] Data transmission and processing:

[0542] The terminal transmits the converted character information to the computing device via the internet. The transmission is secured by the SSL / TLS protocol, thus maintaining data confidentiality. The server analyzes this character information, selects a pre-trained generative model, and uses it to generate a response. This pre-trained generative model is trained based on information from sources such as appliance specifications and frequently asked questions.

[0543] Response generation and presentation:

[0544] The server inputs textual information into a trained generative model and generates a response that matches the user's request. This response is fast and accurate, precisely meeting the user's needs. For example, in response to the instruction "Please turn on the air conditioner," the generated response might be output as "The air conditioner will start operating with the current settings." The server sends this response back to the terminal, where it is played back as speech using speech synthesis technology (TTS engine).

[0545] Personalization and learning:

[0546] The server accumulates user feedback and usage history, and uses this data to optimize responses. This is achieved through a learning algorithm, leading to personalization that adapts to user preferences and behavioral patterns. This process improves the accuracy of responses to the information users are seeking.

[0547] As a concrete example, consider a scenario where a user asks aloud, "My washing machine is showing error code E04. What should I do?" The terminal converts this audio into text and sends it to the server. The server's AI model searches for error code E04 and generates specific instructions such as "Please clean the filter," which it then communicates to the user via voice.

[0548] An example of a prompt message would be text such as, "Set the air conditioner temperature to 21 degrees."

[0549] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0550] Step 1:

[0551] Voice input from the user

[0552] The user provides instructions and questions about home appliances using voice. Natural speech is used as input. The terminal acquires this voice signal using a high-performance audio input device. Microphones and related voice capture technologies are used to convert the acquired voice signal into a digital format.

[0553] Step 2:

[0554] Speech-to-text conversion

[0555] The device takes the received audio signal as input and converts it into text data using speech recognition technology. Specifically, services such as Google Speech-to-Text and Amazon's are used. The data processing in this step involves converting the audio signal into text data, and algorithms are used for noise reduction and accuracy improvement. The output is text data.

[0556] Step 3:

[0557] Sending text data

[0558] The terminal sends the converted text data as output to the server. Input is text data, and output is data communication to the server. Transmission is performed using the SSL / TLS protocol to ensure data confidentiality.

[0559] Step 4:

[0560] Text analysis and response generation

[0561] The server receives character data and performs analysis using a trained generative model. The input is character data, which undergoes analysis based on the AI ​​model. The generative AI model receives a prompt and generates a response based on the user's request. The output is the response in generated text.

[0562] Step 5:

[0563] Sending the generated response

[0564] The server sends the generated response to the terminal. The input is the text response generated by the server, and the output is data communication to the terminal. Security is ensured by using SSL / TLS again as the communication protocol.

[0565] Step 6:

[0566] Text-to-speech conversion and presentation

[0567] The terminal receives the transmitted text response and converts it into an audio signal using text-to-speech (TTS) technology. The input is the generated text response, and the output is the audio signal to be presented to the user. Specifically, the terminal uses an audio output device to play the response aloud for the user.

[0568] (Application Example 1)

[0569] 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."

[0570] In modern brick-and-mortar stores, customers require significant support to understand and operate products, but maintaining sufficient staff to provide adequate assistance at all times presents a challenge. Furthermore, customers may not be able to quickly obtain specific product information, resulting in missed sales opportunities. Additionally, there is a need for smart devices that can provide intuitive and useful information to tech-savvy customers.

[0571] 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.

[0572] In this invention, the server includes a device means for acquiring voice information and converting the voice into text data, a transmission module means for transmitting the converted text data to the server, and means equipped with a function to connect to and provide information display devices for physical stores so that users can directly check the content of their questions in the physical store. This makes it possible for customers to receive necessary product information directly and quickly in the physical store while reducing the burden on staff and providing high-quality service.

[0573] "Audio information" refers to information obtained by acquiring sound waves as digital data.

[0574] "Text data" refers to data obtained by analyzing audio information and representing it as a string of characters.

[0575] "Device means" refers to hardware and software components designed and used to achieve a specific function.

[0576] A "transmission module" refers to a means of communication used to send digital data to other computer devices or servers.

[0577] A "generative intelligence system" is a system that uses artificial intelligence technology to generate appropriate responses and information from received data.

[0578] "Audio output data" refers to digital audio data that has been converted to convey information as audio that can be understood by the listener.

[0579] An "audio playback device" refers to a hardware device that outputs digital audio data as physical sound.

[0580] A "user" is an individual or group that operates a system or device and receives its functions or services.

[0581] "Feedback" refers to responses and opinions provided based on the results of using the system, and is information that helps in future improvements.

[0582] A "learning method" refers to an algorithm or process that allows a system to learn from past data and usage patterns to improve its performance.

[0583] A "physical store information display device" refers to a device that includes displays and audio devices for providing product information to customers in a physical store.

[0584] This invention is a system for providing smooth information when users ask questions or perform operations related to products via voice in a physical store. This system combines voice recognition technology and a generative AI model to provide users with quick and accurate responses.

[0585] The system first uses a voice input device equipped with a high-performance microphone to capture the user's speech. This voice information is then converted into text data by speech recognition software (e.g., Google Speech-to-Text) on the device.

[0586] The converted text data is sent to the server in real time using a secure communication protocol. The server utilizes a generative intelligence system (e.g., an AI model on Amazon Web Services) to generate the optimal response from the received text data. This response generation uses a pre-trained model and is particularly suited to datasets appropriate for product operation and information provision in physical stores.

[0587] The generated responses are converted into audio output data and presented to the user via an audio playback device installed in the store. This allows users to instantly obtain the information they need without being affected by other customers or the store environment.

[0588] Furthermore, the system incorporates learning methods based on user feedback and experience, which continuously improves the accuracy and quality of responses. For example, if a customer asks, "How do I use a specific cooking mode on this microwave?", the system will provide detailed instructions that best suit the user's request. An example of a related prompt would be, "Explain the optimal operating procedure for the cooking mode the customer is asking about, based on the specifications."

[0589] Through this series of processes, the present invention provides a powerful support tool to improve the user experience and enable customers to get the most out of the product.

[0590] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0591] Step 1:

[0592] The device acquires user voice information using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is the first thing needed as preparation for speech recognition.

[0593] Step 2:

[0594] The device converts the acquired digital audio data into text data using speech recognition software (e.g., Google Speech-to-Text). During this process, the audio waveform is analyzed and converted into corresponding text. The output is text data representing the user's voice.

[0595] Step 3:

[0596] The terminal sends the converted text data to the server. Here, the input is the text data on the terminal, and the output is the data received on the server side. This transmission is carried out through a secure communication protocol, ensuring the confidentiality of the data.

[0597] Step 4:

[0598] The server processes the received text data and uses a generative intelligence system to generate the optimal response. The input is transcribed conversational data, and the output is an appropriate response text that meets the user's request. In this process, data calculations are performed using a pre-trained model, and specific answers are derived through the use of prompts.

[0599] Step 5:

[0600] The server converts the generated response text into data for speech output and sends it back to the terminal. The input is the response text generated by the server, and the output is data for speech synthesis. Text-to-speech technology is used for this conversion.

[0601] Step 6:

[0602] The terminal receives data for voice output and presents a response to the user via a voice playback device. The input is synthesized speech data, and the output is physical voice playback. This allows the user to obtain the information or instructions they are looking for in voice.

[0603] 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.

[0604] This invention provides a system that operates home appliances using user voice input, and in addition, detects and considers the user's emotional state. Specific embodiments for carrying out this invention are described below.

[0605] Voice input acquisition and emotion recognition:

[0606] Users ask questions or give instructions about home appliances by voice to the device. This voice is captured by the device's high-performance microphone. The device activates its voice recognition engine and converts the captured voice data into text data, while simultaneously using its emotion engine to identify the user's emotions from the voice.

[0607] Data processing:

[0608] The terminal sends the converted text data and emotional state to the server. The server receives the text data and analyzes it along with the emotional identification results. Within the server, a generative artificial intelligence model is executed to generate the optimal response, taking into account the user's emotional state. This generation process involves adjusting the response based on the user's emotions.

[0609] Response generation and feedback:

[0610] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data and present it to the user as voice. This allows the user to quickly obtain information on the spot. Furthermore, the terminal collects user reactions and other feedback and sends it to the server to improve the accuracy of the AI ​​model and emotion engine.

[0611] Specific example:

[0612] For example, if a user grumpily says, "Turn on the air conditioner," the device captures this audio and converts it to text, while its emotion engine identifies the emotion as "grumpy." Based on this information, the server generates a response in a gentle tone, such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The device then plays this response back as audio. If the user requests further assistance and the system detects that their emotional state is unstable, it can also be configured to offer additional support options.

[0613] In the form described above, this invention can provide meticulous service that responds to the user's emotions and improve the user experience.

[0614] The following describes the processing flow.

[0615] Step 1:

[0616] The user gives voice commands or asks questions to the home appliance. These voice commands are captured by a microphone built into the device.

[0617] Step 2:

[0618] The device processes the captured audio data in two ways. First, a speech recognition engine converts the audio into text data. Second, an emotion engine identifies the user's emotions in real time from the audio data.

[0619] Step 3:

[0620] The device sends the converted text data and identified sentiment data to the server. This communication is encrypted using a secure protocol to ensure data security.

[0621] Step 4:

[0622] The server analyzes the received text data and sentiment data. Based on the text data, it selects the optimal generative artificial intelligence model and generates a response that takes into account adjustments based on the sentiment data.

[0623] Step 5:

[0624] The server sends back a response in text format to the terminal. This response includes content that reflects the user's emotional state, enabling more appropriate communication.

[0625] Step 6:

[0626] The text response received by the terminal is converted into audio data using a text-to-speech (TTS) engine. This audio is immediately presented to the user through the terminal's speaker.

[0627] Step 7:

[0628] The device collects responses and feedback to the user's actions. For example, it can make further sentiment inferences based on voice tone and additional instructions.

[0629] Step 8:

[0630] The device sends the collected feedback to the server, which uses it to improve the AI ​​model and enhance the accuracy of the emotion engine. If the response is unclear, it attempts to connect the user to the support center to improve the user's support experience.

[0631] (Example 2)

[0632] 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."

[0633] Conventional voice control systems have limited the user experience because they generate purely businesslike responses without considering the user's emotions. Furthermore, the system sometimes fails to respond appropriately when users make ambiguous requests regarding the operation of home appliances. In addition, there is a lack of means to improve the system's accuracy by utilizing user feedback in real time. As a result, it has been difficult to provide responses that are sensitive to the user's intentions and emotions.

[0634] 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.

[0635] In this invention, the server includes a device means for identifying the user's emotions from acquired voice; an information processing device means for receiving text information and emotion information from an external device and generating a response that takes the user's emotions into account using a generative artificial intelligence model; and a learning device means for improving the response content based on the user's reactions. This enables the generation of personalized responses that take the user's emotions into account and continuous improvement of the system's accuracy.

[0636] A "device for acquiring voice input" is a device that captures the voice emitted by the user and converts it into a digital signal for subsequent processing.

[0637] A "device that converts to text information" is a device that analyzes acquired voice input and converts it into text data as a corresponding string of characters.

[0638] A "device for identifying the user's emotions" is a device that identifies the user's emotional state by analyzing the characteristics of voice input.

[0639] A "communication device" is a device used to send and receive data to and from other devices or equipment via a network.

[0640] A "generative artificial intelligence model" is an algorithm or program trained to generate human-like responses or outputs based on input data.

[0641] An "information processing device" is a device that has the function of analyzing received data and performing appropriate processing or calculations as needed.

[0642] A "device that converts to audio data" is a device that has the function of converting text-based data into audible audio data using speech synthesis technology.

[0643] An "audio output device" is a device that reproduces audio data and transmits it to the user as sound.

[0644] A "learning device" is a device equipped with algorithms that continuously collect feedback from users and contribute to improving the system's performance.

[0645] An "information exchange method" is a set of protocols or guidelines for effectively and securely exchanging data and information between different systems.

[0646] This invention is a system that allows users to operate home appliances using voice commands and provides optimal responses by considering the emotional information detected during the operation. Specific embodiments for carrying out this invention are described below.

[0647] The user gives voice commands to the home appliances through the terminal. The terminal uses a high-performance microphone to capture this voice. The captured voice is converted into text using speech recognition technology, and at the same time, an emotion recognition device analyzes the user's emotions. This results in text data and emotional information.

[0648] Next, the terminal sends text data and emotional information to the server via a communication device. The server processes the received data using a generative AI model and generates an appropriate response that matches the user's emotions. This model is inputted with prompt sentences such as, "If the user says 'Turn on the air conditioner' in a dissatisfied voice, generate a situation-appropriate response in a soft tone."

[0649] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data, which is then presented to the user through an audio output device.

[0650] As a concrete example, consider a scenario where a user grumpily says, "Turn on the air conditioner." In this case, the emotion recognition device detects the emotion of "grumpiness," and the server's AI model generates a gentle response such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The terminal then plays this response as audio and conveys it to the user.

[0651] Furthermore, the device continuously collects user responses and sends them back to the server. The server uses this feedback to act as a learning device, improving the accuracy of its responses.

[0652] In this way, by considering both the user's voice commands and emotional state, a more refined user experience can be provided.

[0653] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0654] Step 1:

[0655] The user gives voice commands to home appliances. The terminal captures this voice in real time using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is sent to the processing unit while noise is reduced. Specifically, the user performs an action such as saying, "Turn on the air conditioner."

[0656] Step 2:

[0657] The terminal inputs the captured audio data into a speech recognition engine, which converts the audio data into text data. At this stage, spectral analysis and phoneme analysis of the audio signal are performed. The output is text data such as "Turn on the air conditioner." In parallel, the same audio data is sent to an emotion recognition device, which analyzes the intonation, speed, and volume of the voice to extract emotional information. The output is data indicating an emotional state, such as "unhappy."

[0658] Step 3:

[0659] The terminal transmits the converted text data and sentiment information to the server via a communication device. The input is the text and sentiment data processed and generated on the terminal, and the output is encrypted data packets that reach the server securely. Specifically, the data is packetized and transmitted using the HTTPS protocol.

[0660] Step 4:

[0661] The server uses a generative AI model to analyze the received text data and sentiment information. The input is the user's text data and sentiment data, and the output is an appropriate response text that takes the user's sentiment into account. The generative AI model generates responses to prompts such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." Natural language processing techniques are used in this process.

[0662] Step 5:

[0663] The response text generated by the server is sent to the terminal. The terminal uses a TTS (Text-to-Speech) engine to synthesize this text into speech and convert it into audio data. The input is the generated response text, and the output is the audio data. The terminal plays this audio to the user through its speaker; specifically, it plays the message, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings."

[0664] Step 6:

[0665] The terminal observes the user's responses and collects feedback as needed. If the user gives further instructions or performs a different action, that data is sent back to the server via the communication device. The input is the user's additional feedback and response data, and the output is improvement data sent to the server. In this way, the entire system improves accuracy through the user's continued use.

[0666] (Application Example 2)

[0667] 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."

[0668] In home appliances and information devices operated via voice input, there is a need for systems that can respond flexibly to the user's emotions. However, conventional systems are limited to converting voice commands to text and provide uniform responses without considering the user's emotional state, which is a problem as it fails to sufficiently increase user satisfaction. Furthermore, in physical stores, it is difficult for sales staff to immediately grasp the customer's emotions and provide appropriate service, which hinders the improvement of the customer experience.

[0669] 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.

[0670] In this invention, the server includes emotion recognition means for identifying the emotional state of voice input and adjusting the response; means for making adjustments before presenting a response based on the identified emotional state; and learning procedure means for improving the response content based on user feedback. This makes it possible to provide individualized responses that are appropriate to the user's emotions and to help improve the customer experience.

[0671] An "emotion recognition mechanism" is a system that identifies the user's emotional state from their voice and adjusts its response based on that information.

[0672] "Voice input" refers to a method by which users provide instructions or information to a system through their voice.

[0673] "Textual information" refers to a data format in which voice input is analyzed and expressed as words.

[0674] "Communication means" refers to a method or device for transmitting converted data to another device or server.

[0675] An "artificial intelligence model for generation" is an algorithm designed to generate an appropriate response based on input information.

[0676] A "sound output mechanism" is a device that presents generated sound data to the user visually or audibly.

[0677] A "learning procedure" is the process of optimizing a system using feedback to provide improved responses in the future.

[0678] This invention constructs the following system to realize interaction using voice input: The terminal uses a high-sensitivity microphone to acquire the user's voice input. The acquired voice is converted into text information by voice recognition software installed on the terminal (e.g., Google Cloud Speech-to-Text API). In addition, sentiment analysis software (e.g., IBM Watson Tone Analyzer) determines the user's emotional state from the voice. This converted data and sentiment information are quickly transmitted to the server.

[0679] The server uses a generative AI model (e.g., OpenAI GPT series) based on the received data to generate an emotionally appropriate response. The AI ​​model utilizes prompts that take the user's emotional state into account to generate the most suitable response for each situation. This response is then sent back to the terminal, converted into audio data using text-to-speech synthesis (e.g., Amazon Polly), and communicated to the user through an audio output mechanism.

[0680] For example, in a physical store, when a salesperson wearing smart glasses interacts with a customer, if the customer asks, "Do you have this product in a different color?", and the emotional analysis detects a slight state of anxiety, a response with a gentle tone such as, "Of course, I'd be happy to help," will be generated. In this way, the generated response is naturally suggested to the customer by the salesperson.

[0681] An example of a prompt for a generative AI model is: "The user asks 'AA' and is showing anxiety. Generate an appropriate response in a calming, gentle tone." This allows the system to provide the most appropriate and human-like interaction for the user.

[0682] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0683] Step 1:

[0684] The device acquires the user's voice input. A high-sensitivity microphone built into the device picks up the user's voice and prepares it for processing as digital audio data. This data is used as input for speech recognition.

[0685] Step 2:

[0686] The device launches speech recognition software and converts the audio data into text. The process of converting the audio data into unique text information is performed using the Google Cloud Speech-to-Text API. The output is text information.

[0687] Step 3:

[0688] The device identifies the emotional state from the voice data. Using IBM Watson Tone Analyzer, it analyzes the user's emotions within the voice data to identify their emotional state. This step outputs information indicating the emotional state.

[0689] Step 4:

[0690] The terminal sends the converted text information and sentiment information to the server. The information is transferred to the server via communication means, and the data is prepared for the next processing step.

[0691] Step 5:

[0692] The server generates a response using an AI model based on the received data. Using an AI model such as the OpenAI GPT series, it generates the optimal response based on the received textual and emotional information. The generated response text is obtained as output.

[0693] Step 6:

[0694] The server sends the generated response to the terminal. The response text is returned to the terminal, and preparations for the next speech synthesis are made. The server also performs error checking to ensure the accurate transmission of the response.

[0695] Step 7:

[0696] The device converts the response text into audio data. Using a text-to-speech engine such as Amazon Polly, it converts the generated response text into audio data and prepares it to be conveyed to the user in an easy-to-understand voice format.

[0697] Step 8:

[0698] The device presents a response to the user via its audio output mechanism. The generated audio data is played through the device's speaker, creating an environment where the user can hear the response.

[0699] Step 9:

[0700] The device collects user feedback and sends it to the server. It gathers information about user reactions and usage, and transmits it to the server via communication. This feedback is used to improve future responses.

[0701] 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.

[0702] 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.

[0703] 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.

[0704] [Fourth Embodiment]

[0705] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0706] 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.

[0707] 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).

[0708] 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.

[0709] 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.

[0710] 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).

[0711] 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.

[0712] 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.

[0713] 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.

[0714] 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.

[0715] 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.

[0716] 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.

[0717] 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".

[0718] This invention provides a system for interpreting user intent from voice input and providing operation and support for home appliances based on that intent. Specifically, it is implemented in the following form.

[0719] Acquiring and converting voice input:

[0720] The process begins with the user voice-initiating questions or instructions about home appliances. The device is equipped with a high-performance microphone to capture this voice. The device then uses speech recognition technology to convert this voice into text data.

[0721] Data transmission and processing:

[0722] The terminal sends the converted text data to the server. This process is carried out using a secure communication protocol, ensuring data confidentiality. The server analyzes the received text and selects an appropriate generative artificial intelligence model. This AI model is pre-tuned and has learned from consumer electronics instruction manuals and FAQs.

[0723] Response generation and presentation:

[0724] The server inputs text data into an AI model and generates an appropriate response. This response answers the user's question quickly and accurately. For example, in response to a command such as "Turn on the air conditioner," it generates a response like "Turning on the air conditioner with the current settings." The server sends this response back to the terminal, which then uses a text-to-speech (TTS) engine to play the response aloud.

[0725] Learning and personalization:

[0726] By collecting user feedback and usage patterns, the server personalizes the AI ​​model. This allows the system to improve the quality of responses based on user preferences and past behavior. In addition, if the server cannot obtain an appropriate answer to a user's inquiry, it automatically connects the user to the manufacturer's support center. Through this series of processes, the present invention can provide users with intuitive and efficient support.

[0727] For example, if a user asks, "My washing machine is showing error code E04. What should I do?", the device converts this voice into text and sends it to the server. The server's AI model searches for information related to error code E04, generates specific instructions such as "Clean the filter," and communicates this to the user verbally. Through this process, the user can quickly resolve the problem.

[0728] The following describes the processing flow.

[0729] Step 1:

[0730] The user speaks into the device with questions or instructions about home appliances. This voice is captured by the device's microphone.

[0731] Step 2:

[0732] The device processes the captured audio data using a speech recognition engine and converts it into text data. This text data provides the foundational information for interpreting the user's intent.

[0733] Step 3:

[0734] The terminal sends the converted text data to the server. This communication uses a secure protocol to protect the confidentiality of the data.

[0735] Step 4:

[0736] The server analyzes the received text data and selects an appropriate generative artificial intelligence model. This model is pre-tuned based on specific consumer electronics product information and FAQs.

[0737] Step 5:

[0738] The server inputs text data into the AI ​​model and generates a suitable response. This response is designed to provide the most appropriate answer to the user's question.

[0739] Step 6:

[0740] The server sends the generated response to the terminal in text format. This response is intended to answer the user based on the original question.

[0741] Step 7:

[0742] The terminal converts the received text response into speech data using a text-to-speech (TTS) engine. This conversion is necessary for the user to receive the response in voice.

[0743] Step 8:

[0744] The device plays the generated audio data through its speaker, providing the user with a verbal response. This allows the user to intuitively receive instructions and information.

[0745] Step 9:

[0746] The device collects user feedback and behavioral data and sends it to the server. The server uses this data to personalize and improve the AI ​​model.

[0747] Step 10:

[0748] If the server cannot find a valid answer to the user's inquiry, it will automatically attempt to connect with the manufacturer's support center following a pre-configured procedure. The progress of this connection will be notified to the user through their device.

[0749] (Example 1)

[0750] 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".

[0751] Conventional voice recognition systems have struggled to provide quick and accurate responses to control electrical appliances and meet specific user needs. Furthermore, they have inadequate handling of situations where commands are not recognized or incorrect responses are generated. As a result, user convenience is often compromised.

[0752] 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.

[0753] In this invention, the server includes a speech recognition unit means that acquires an audio signal and converts the audio signal into character data; a processing unit means in which a computing device receives the character data and generates a response to the operator's instructions using a trained generative model; and a communication protocol means that automatically connects to the support system of the instruction target if a suitable response is not found. This enables the provision of accurate and rapid responses to voice input, improving user convenience.

[0754] "Audio signal" refers to the electronic signal obtained by converting sound waves acquired by an input device such as a microphone.

[0755] "Character data" refers to data obtained by analyzing an audio signal using speech recognition technology and converting it into digital character format.

[0756] A "speech recognition unit" refers to the technology and equipment used to convert speech signals into text data.

[0757] A "communication module" refers to hardware or software used for data communication when transmitting character data to a computing device.

[0758] A "computational device" refers to a computer or server that receives character data and processes it to generate responses.

[0759] A "pre-trained generative model" refers to an artificial intelligence model that has been trained in advance based on a large amount of data and is designed to generate responses according to a specific task.

[0760] A "processing unit" refers to a component within a computing device that takes character data as input and generates a response using a pre-trained generative model.

[0761] "Audio output means" refers to devices or technologies that convert the generated response into an audio signal and present it to the user as sound.

[0762] A "playback unit" refers to a component that uses an audio output means to reproduce the generated audio signal.

[0763] A "learning algorithm" refers to a data processing method used to improve the response accuracy of a system based on user feedback and usage patterns.

[0764] A "communication protocol" refers to a set of rules and procedures used when exchanging data between a computing device and an external system.

[0765] "Personalization" refers to the process of adapting a system based on the operator's preferences and past behavior, thereby individualizing the content of its responses.

[0766] One embodiment of this invention is a system configured as follows.

[0767] Acquiring and converting voice input:

[0768] The process begins with the user giving instructions or asking questions about home appliances using voice commands. The terminal is equipped with a high-performance audio input device that effectively captures the user's voice. The voice signal is converted into digital text information by a computation program running on the terminal. Commonly used speech recognition technologies such as Google Speech-to-Text or Amazon's service can be applied to this conversion.

[0769] Data transmission and processing:

[0770] The terminal transmits the converted character information to the computing device via the internet. The transmission is secured by the SSL / TLS protocol, thus maintaining data confidentiality. The server analyzes this character information, selects a pre-trained generative model, and uses it to generate a response. This pre-trained generative model is trained based on information from sources such as appliance specifications and frequently asked questions.

[0771] Response generation and presentation:

[0772] The server inputs textual information into a trained generative model and generates a response that matches the user's request. This response is fast and accurate, precisely meeting the user's needs. For example, in response to the instruction "Please turn on the air conditioner," the generated response might be output as "The air conditioner will start operating with the current settings." The server sends this response back to the terminal, where it is played back as speech using speech synthesis technology (TTS engine).

[0773] Personalization and learning:

[0774] The server accumulates user feedback and usage history, and uses this data to optimize responses. This is achieved through a learning algorithm, leading to personalization that adapts to user preferences and behavioral patterns. This process improves the accuracy of responses to the information users are seeking.

[0775] As a concrete example, consider a scenario where a user asks aloud, "My washing machine is showing error code E04. What should I do?" The terminal converts this audio into text and sends it to the server. The server's AI model searches for error code E04 and generates specific instructions such as "Please clean the filter," which it then communicates to the user via voice.

[0776] An example of a prompt message would be text such as, "Set the air conditioner temperature to 21 degrees."

[0777] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0778] Step 1:

[0779] Voice input from the user

[0780] The user provides instructions and questions about home appliances using voice. Natural speech is used as input. The terminal acquires this voice signal using a high-performance audio input device. Microphones and related voice capture technologies are used to convert the acquired voice signal into a digital format.

[0781] Step 2:

[0782] Speech-to-text conversion

[0783] The device takes the received audio signal as input and converts it into text data using speech recognition technology. Specifically, services such as Google Speech-to-Text and Amazon's are used. The data processing in this step involves converting the audio signal into text data, and algorithms are used for noise reduction and accuracy improvement. The output is text data.

[0784] Step 3:

[0785] Sending text data

[0786] The terminal sends the converted text data as output to the server. Input is text data, and output is data communication to the server. Transmission is performed using the SSL / TLS protocol to ensure data confidentiality.

[0787] Step 4:

[0788] Text analysis and response generation

[0789] The server receives character data and performs analysis using a trained generative model. The input is character data, which undergoes analysis based on the AI ​​model. The generative AI model receives a prompt and generates a response based on the user's request. The output is the response in generated text.

[0790] Step 5:

[0791] Sending the generated response

[0792] The server sends the generated response to the terminal. The input is the text response generated by the server, and the output is data communication to the terminal. Security is ensured by using SSL / TLS again as the communication protocol.

[0793] Step 6:

[0794] Text-to-speech conversion and presentation

[0795] The terminal receives the transmitted text response and converts it into an audio signal using text-to-speech (TTS) technology. The input is the generated text response, and the output is the audio signal to be presented to the user. Specifically, the terminal uses an audio output device to play the response aloud for the user.

[0796] (Application Example 1)

[0797] 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".

[0798] In modern brick-and-mortar stores, customers require significant support to understand and operate products, but maintaining sufficient staff to provide adequate assistance at all times presents a challenge. Furthermore, customers may not be able to quickly obtain specific product information, resulting in missed sales opportunities. Additionally, there is a need for smart devices that can provide intuitive and useful information to tech-savvy customers.

[0799] 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.

[0800] In this invention, the server includes a device means for acquiring voice information and converting the voice into text data, a transmission module means for transmitting the converted text data to the server, and means equipped with a function to connect to and provide information display devices for physical stores so that users can directly check the content of their questions in the physical store. This makes it possible for customers to receive necessary product information directly and quickly in the physical store while reducing the burden on staff and providing high-quality service.

[0801] "Audio information" refers to information obtained by acquiring sound waves as digital data.

[0802] "Text data" refers to data obtained by analyzing audio information and representing it as a string of characters.

[0803] "Device means" refers to hardware and software components designed and used to achieve a specific function.

[0804] A "transmission module" refers to a means of communication used to send digital data to other computer devices or servers.

[0805] A "generative intelligence system" is a system that uses artificial intelligence technology to generate appropriate responses and information from received data.

[0806] "Audio output data" refers to digital audio data that has been converted to convey information as audio that can be understood by the listener.

[0807] An "audio playback device" refers to a hardware device that outputs digital audio data as physical sound.

[0808] A "user" is an individual or group that operates a system or device and receives its functions or services.

[0809] "Feedback" refers to responses and opinions provided based on the results of using the system, and is information that helps in future improvements.

[0810] A "learning method" refers to an algorithm or process that allows a system to learn from past data and usage patterns to improve its performance.

[0811] A "physical store information display device" refers to a device that includes displays and audio devices for providing product information to customers in a physical store.

[0812] This invention is a system for providing smooth information when users ask questions or perform operations related to products via voice in a physical store. This system combines voice recognition technology and a generative AI model to provide users with quick and accurate responses.

[0813] The system first uses a voice input device equipped with a high-performance microphone to capture the user's speech. This voice information is then converted into text data by speech recognition software (e.g., Google Speech-to-Text) on the device.

[0814] The converted text data is sent to the server in real time using a secure communication protocol. The server utilizes a generative intelligence system (e.g., an AI model on Amazon Web Services) to generate the optimal response from the received text data. This response generation uses a pre-trained model and is particularly suited to datasets appropriate for product operation and information provision in physical stores.

[0815] The generated responses are converted into audio output data and presented to the user via an audio playback device installed in the store. This allows users to instantly obtain the information they need without being affected by other customers or the store environment.

[0816] Furthermore, the system incorporates learning methods based on user feedback and experience, which continuously improves the accuracy and quality of responses. For example, if a customer asks, "How do I use a specific cooking mode on this microwave?", the system will provide detailed instructions that best suit the user's request. An example of a related prompt would be, "Explain the optimal operating procedure for the cooking mode the customer is asking about, based on the specifications."

[0817] Through this series of processes, the present invention provides a powerful support tool to improve the user experience and enable customers to get the most out of the product.

[0818] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0819] Step 1:

[0820] The device acquires user voice information using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is the first thing needed as preparation for speech recognition.

[0821] Step 2:

[0822] The device converts the acquired digital audio data into text data using speech recognition software (e.g., Google Speech-to-Text). During this process, the audio waveform is analyzed and converted into corresponding text. The output is text data representing the user's voice.

[0823] Step 3:

[0824] The terminal sends the converted text data to the server. Here, the input is the text data on the terminal, and the output is the data received on the server side. This transmission is carried out through a secure communication protocol, ensuring the confidentiality of the data.

[0825] Step 4:

[0826] The server processes the received text data and uses a generative intelligence system to generate the optimal response. The input is transcribed conversational data, and the output is an appropriate response text that meets the user's request. In this process, data calculations are performed using a pre-trained model, and specific answers are derived through the use of prompts.

[0827] Step 5:

[0828] The server converts the generated response text into data for speech output and sends it back to the terminal. The input is the response text generated by the server, and the output is data for speech synthesis. Text-to-speech technology is used for this conversion.

[0829] Step 6:

[0830] The terminal receives data for voice output and presents a response to the user via a voice playback device. The input is synthesized speech data, and the output is physical voice playback. This allows the user to obtain the information or instructions they are looking for in voice.

[0831] 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.

[0832] This invention provides a system that operates home appliances using user voice input, and in addition, detects and considers the user's emotional state. Specific embodiments for carrying out this invention are described below.

[0833] Voice input acquisition and emotion recognition:

[0834] Users ask questions or give instructions about home appliances by voice to the device. This voice is captured by the device's high-performance microphone. The device activates its voice recognition engine and converts the captured voice data into text data, while simultaneously using its emotion engine to identify the user's emotions from the voice.

[0835] Data processing:

[0836] The terminal sends the converted text data and emotional state to the server. The server receives the text data and analyzes it along with the emotional identification results. Within the server, a generative artificial intelligence model is executed to generate the optimal response, taking into account the user's emotional state. This generation process involves adjusting the response based on the user's emotions.

[0837] Response generation and feedback:

[0838] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data and present it to the user as voice. This allows the user to quickly obtain information on the spot. Furthermore, the terminal collects user reactions and other feedback and sends it to the server to improve the accuracy of the AI ​​model and emotion engine.

[0839] Specific example:

[0840] For example, if a user grumpily says, "Turn on the air conditioner," the device captures this audio and converts it to text, while its emotion engine identifies the emotion as "grumpy." Based on this information, the server generates a response in a gentle tone, such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The device then plays this response back as audio. If the user requests further assistance and the system detects that their emotional state is unstable, it can also be configured to offer additional support options.

[0841] In the form described above, this invention can provide meticulous service that responds to the user's emotions and improve the user experience.

[0842] The following describes the processing flow.

[0843] Step 1:

[0844] The user gives voice commands or asks questions to the home appliance. These voice commands are captured by a microphone built into the device.

[0845] Step 2:

[0846] The device processes the captured audio data in two ways. First, a speech recognition engine converts the audio into text data. Second, an emotion engine identifies the user's emotions in real time from the audio data.

[0847] Step 3:

[0848] The device sends the converted text data and identified sentiment data to the server. This communication is encrypted using a secure protocol to ensure data security.

[0849] Step 4:

[0850] The server analyzes the received text data and sentiment data. Based on the text data, it selects the optimal generative artificial intelligence model and generates a response that takes into account adjustments based on the sentiment data.

[0851] Step 5:

[0852] The server sends back a response in text format to the terminal. This response includes content that reflects the user's emotional state, enabling more appropriate communication.

[0853] Step 6:

[0854] The text response received by the terminal is converted into audio data using a text-to-speech (TTS) engine. This audio is immediately presented to the user through the terminal's speaker.

[0855] Step 7:

[0856] The device collects responses and feedback to the user's actions. For example, it can make further sentiment inferences based on voice tone and additional instructions.

[0857] Step 8:

[0858] The device sends the collected feedback to the server, which uses it to improve the AI ​​model and enhance the accuracy of the emotion engine. If the response is unclear, it attempts to connect the user to the support center to improve the user's support experience.

[0859] (Example 2)

[0860] 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".

[0861] Conventional voice control systems have limited the user experience because they generate purely businesslike responses without considering the user's emotions. Furthermore, the system sometimes fails to respond appropriately when users make ambiguous requests regarding the operation of home appliances. In addition, there is a lack of means to improve the system's accuracy by utilizing user feedback in real time. As a result, it has been difficult to provide responses that are sensitive to the user's intentions and emotions.

[0862] 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.

[0863] In this invention, the server includes a device means for identifying the user's emotions from acquired voice; an information processing device means for receiving text information and emotion information from an external device and generating a response that takes the user's emotions into account using a generative artificial intelligence model; and a learning device means for improving the response content based on the user's reactions. This enables the generation of personalized responses that take the user's emotions into account and continuous improvement of the system's accuracy.

[0864] A "device for acquiring voice input" is a device that captures the voice emitted by the user and converts it into a digital signal for subsequent processing.

[0865] A "device that converts to text information" is a device that analyzes acquired voice input and converts it into text data as a corresponding string of characters.

[0866] A "device for identifying the user's emotions" is a device that identifies the user's emotional state by analyzing the characteristics of voice input.

[0867] A "communication device" is a device used to send and receive data to and from other devices or equipment via a network.

[0868] A "generative artificial intelligence model" is an algorithm or program trained to generate human-like responses or outputs based on input data.

[0869] An "information processing device" is a device that has the function of analyzing received data and performing appropriate processing or calculations as needed.

[0870] A "device that converts to audio data" is a device that has the function of converting text-based data into audible audio data using speech synthesis technology.

[0871] An "audio output device" is a device that reproduces audio data and transmits it to the user as sound.

[0872] A "learning device" is a device equipped with algorithms that continuously collect feedback from users and contribute to improving the system's performance.

[0873] An "information exchange method" is a set of protocols or guidelines for effectively and securely exchanging data and information between different systems.

[0874] This invention is a system that allows users to operate home appliances using voice commands and provides optimal responses by considering the emotional information detected during the operation. Specific embodiments for carrying out this invention are described below.

[0875] The user gives voice commands to the home appliances through the terminal. The terminal uses a high-performance microphone to capture this voice. The captured voice is converted into text using speech recognition technology, and at the same time, an emotion recognition device analyzes the user's emotions. This results in text data and emotional information.

[0876] Next, the terminal sends text data and emotional information to the server via a communication device. The server processes the received data using a generative AI model and generates an appropriate response that matches the user's emotions. This model is inputted with prompt sentences such as, "If the user says 'Turn on the air conditioner' in a dissatisfied voice, generate a situation-appropriate response in a soft tone."

[0877] The server generates a response, which is sent to the terminal in text format. The terminal uses a text-to-speech (TTS) engine to convert this response into audio data, which is then presented to the user through an audio output device.

[0878] As a concrete example, consider a scenario where a user grumpily says, "Turn on the air conditioner." In this case, the emotion recognition device detects the emotion of "grumpiness," and the server's AI model generates a gentle response such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." The terminal then plays this response as audio and conveys it to the user.

[0879] Furthermore, the device continuously collects user responses and sends them back to the server. The server uses this feedback to act as a learning device, improving the accuracy of its responses.

[0880] In this way, by considering both the user's voice commands and emotional state, a more refined user experience can be provided.

[0881] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0882] Step 1:

[0883] The user gives voice commands to home appliances. The terminal captures this voice in real time using a high-performance microphone. The input is the user's voice, and the output is digital voice data. This data is sent to the processing unit while noise is reduced. Specifically, the user performs an action such as saying, "Turn on the air conditioner."

[0884] Step 2:

[0885] The terminal inputs the captured audio data into a speech recognition engine, which converts the audio data into text data. At this stage, spectral analysis and phoneme analysis of the audio signal are performed. The output is text data such as "Turn on the air conditioner." In parallel, the same audio data is sent to an emotion recognition device, which analyzes the intonation, speed, and volume of the voice to extract emotional information. The output is data indicating an emotional state, such as "unhappy."

[0886] Step 3:

[0887] The terminal transmits the converted text data and sentiment information to the server via a communication device. The input is the text and sentiment data processed and generated on the terminal, and the output is encrypted data packets that reach the server securely. Specifically, the data is packetized and transmitted using the HTTPS protocol.

[0888] Step 4:

[0889] The server uses a generative AI model to analyze the received text data and sentiment information. The input is the user's text data and sentiment data, and the output is an appropriate response text that takes the user's sentiment into account. The generative AI model generates responses to prompts such as, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings." Natural language processing techniques are used in this process.

[0890] Step 5:

[0891] The response text generated by the server is sent to the terminal. The terminal uses a TTS (Text-to-Speech) engine to synthesize this text into speech and convert it into audio data. The input is the generated response text, and the output is the audio data. The terminal plays this audio to the user through its speaker; specifically, it plays the message, "It's a little hot, isn't it? I'll turn on the air conditioner with the current settings."

[0892] Step 6:

[0893] The terminal observes the user's responses and collects feedback as needed. If the user gives further instructions or performs a different action, that data is sent back to the server via the communication device. The input is the user's additional feedback and response data, and the output is improvement data sent to the server. In this way, the entire system improves accuracy through the user's continued use.

[0894] (Application Example 2)

[0895] 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".

[0896] In home appliances and information devices operated via voice input, there is a need for systems that can respond flexibly to the user's emotions. However, conventional systems are limited to converting voice commands to text and provide uniform responses without considering the user's emotional state, which is a problem as it fails to sufficiently increase user satisfaction. Furthermore, in physical stores, it is difficult for sales staff to immediately grasp the customer's emotions and provide appropriate service, which hinders the improvement of the customer experience.

[0897] 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.

[0898] In this invention, the server includes emotion recognition means for identifying the emotional state of voice input and adjusting the response; means for making adjustments before presenting a response based on the identified emotional state; and learning procedure means for improving the response content based on user feedback. This makes it possible to provide individualized responses that are appropriate to the user's emotions and to help improve the customer experience.

[0899] An "emotion recognition mechanism" is a system that identifies the user's emotional state from their voice and adjusts its response based on that information.

[0900] "Voice input" refers to a method by which users provide instructions or information to a system through their voice.

[0901] "Textual information" refers to a data format in which voice input is analyzed and expressed as words.

[0902] "Communication means" refers to a method or device for transmitting converted data to another device or server.

[0903] An "artificial intelligence model for generation" is an algorithm designed to generate an appropriate response based on input information.

[0904] A "sound output mechanism" is a device that presents generated sound data to the user visually or audibly.

[0905] A "learning procedure" is the process of optimizing a system using feedback to provide improved responses in the future.

[0906] This invention constructs the following system to realize interaction using voice input: The terminal uses a high-sensitivity microphone to acquire the user's voice input. The acquired voice is converted into text information by voice recognition software installed on the terminal (e.g., Google Cloud Speech-to-Text API). In addition, sentiment analysis software (e.g., IBM Watson Tone Analyzer) determines the user's emotional state from the voice. This converted data and sentiment information are quickly transmitted to the server.

[0907] The server uses a generative AI model (e.g., OpenAI GPT series) based on the received data to generate an emotionally appropriate response. The AI ​​model utilizes prompts that take the user's emotional state into account to generate the most suitable response for each situation. This response is then sent back to the terminal, converted into audio data using text-to-speech synthesis (e.g., Amazon Polly), and communicated to the user through an audio output mechanism.

[0908] For example, in a physical store, when a salesperson wearing smart glasses interacts with a customer, if the customer asks, "Do you have this product in a different color?", and the emotional analysis detects a slight state of anxiety, a response with a gentle tone such as, "Of course, I'd be happy to help," will be generated. In this way, the generated response is naturally suggested to the customer by the salesperson.

[0909] An example of a prompt for a generative AI model is: "The user asks 'AA' and is showing anxiety. Generate an appropriate response in a calming, gentle tone." This allows the system to provide the most appropriate and human-like interaction for the user.

[0910] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0911] Step 1:

[0912] The device acquires the user's voice input. A high-sensitivity microphone built into the device picks up the user's voice and prepares it for processing as digital audio data. This data is used as input for speech recognition.

[0913] Step 2:

[0914] The device launches speech recognition software and converts the audio data into text. The process of converting the audio data into unique text information is performed using the Google Cloud Speech-to-Text API. The output is text information.

[0915] Step 3:

[0916] The device identifies the emotional state from the voice data. Using IBM Watson Tone Analyzer, it analyzes the user's emotions within the voice data to identify their emotional state. This step outputs information indicating the emotional state.

[0917] Step 4:

[0918] The terminal sends the converted text information and sentiment information to the server. The information is transferred to the server via communication means, and the data is prepared for the next processing step.

[0919] Step 5:

[0920] The server generates a response using an AI model based on the received data. Using an AI model such as the OpenAI GPT series, it generates the optimal response based on the received textual and emotional information. The generated response text is obtained as output.

[0921] Step 6:

[0922] The server sends the generated response to the terminal. The response text is returned to the terminal, and preparations for the next speech synthesis are made. The server also performs error checking to ensure the accurate transmission of the response.

[0923] Step 7:

[0924] The device converts the response text into audio data. Using a text-to-speech engine such as Amazon Polly, it converts the generated response text into audio data and prepares it to be conveyed to the user in an easy-to-understand voice format.

[0925] Step 8:

[0926] The device presents a response to the user via its audio output mechanism. The generated audio data is played through the device's speaker, creating an environment where the user can hear the response.

[0927] Step 9:

[0928] The device collects user feedback and sends it to the server. It gathers information about user reactions and usage, and transmits it to the server via communication. This feedback is used to improve future responses.

[0929] 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.

[0930] 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.

[0931] 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.

[0932] 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.

[0933] 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.

[0934] 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.

[0935] 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.

[0936] 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.

[0937] 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."

[0938] 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.

[0939] 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.

[0940] 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.

[0941] 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.

[0942] 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.

[0943] 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.

[0944] 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.

[0945] 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.

[0946] 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.

[0947] 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.

[0948] 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.

[0949] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0950] The following is further disclosed regarding the embodiments described above.

[0951] (Claim 1)

[0952] A device means for acquiring voice input and converting the voice into text data,

[0953] A communication module means for sending the converted text data to a server,

[0954] A server receives text data and a processing device that generates an appropriate response using a generative artificial intelligence model.

[0955] A device means for converting the generated response into audio data and presenting it through an audio output device,

[0956] A system that includes a learning algorithm that improves response content based on user feedback.

[0957] (Claim 2)

[0958] The system according to claim 1, which uses a trained generative artificial intelligence model configured to generate a response based on information specific to a home appliance from text data.

[0959] (Claim 3)

[0960] The system according to claim 1, comprising a communication protocol for automatically connecting to manufacturer support if no response is found.

[0961] "Example 1"

[0962] (Claim 1)

[0963] A speech recognition unit means that acquires an audio signal and converts the audio signal into text data,

[0964] A communication module means for transmitting the converted character data to a computing device,

[0965] A computing device receives character data and generates a processing unit means that uses a trained generative model to generate a response to the operator's instructions,

[0966] A playback unit means that converts the generated response into an audio signal and presents it through an audio output means,

[0967] A learning algorithm means that improves and personalizes response content based on operator feedback and usage patterns,

[0968] A system including a communication protocol means that automatically connects to a designated support system if an appropriate response is not found.

[0969] (Claim 2)

[0970] The system according to claim 1, which uses a trained generative model configured to generate a response based on information specific to an electrical appliance from text data.

[0971] (Claim 3)

[0972] The system according to claim 1, comprising a personalization process that trains a pre-trained generative model based on operation history and generates responses that reflect the operator's preferences and tendencies.

[0973] "Application Example 1"

[0974] (Claim 1)

[0975] A device means for acquiring audio information and converting the audio into text data,

[0976] A transmission module means for sending the converted text data to a server,

[0977] A processing device means that receives text data from a server and generates an appropriate response using a generative intelligence system,

[0978] A device means that converts the generated response into data for audio output and presents it through an audio playback device,

[0979] A learning method that improves response content based on user feedback,

[0980] A system that includes a means to connect to and provide information display devices for physical stores, enabling users to directly check the content of their questions at the physical store.

[0981] (Claim 2)

[0982] The system according to claim 1, which uses a trained generative intelligence system configured to generate a response based on information specific to an electrical device from text data.

[0983] (Claim 3)

[0984] The system according to claim 1, further comprising means for automatically connecting to manufacturer support if no response is found.

[0985] "Example 2 of combining an emotion engine"

[0986] (Claim 1)

[0987] A device means for acquiring voice input and converting the voice into text information,

[0988] A device that identifies the user's emotions from acquired audio,

[0989] A communication device means for transmitting converted character information and emotional information to an external device,

[0990] An information processing device means that receives textual information and emotional information from an external device and generates a response that takes the user's emotions into account using a generative artificial intelligence model,

[0991] A device means for converting the generated response into acoustic data and presenting it through an acoustic output device,

[0992] A system including a learning device that improves response content based on user reactions.

[0993] (Claim 2)

[0994] The system according to claim 1, which utilizes a trained generative artificial intelligence model configured to generate a response based on information about a home appliance from textual data.

[0995] (Claim 3)

[0996] The system according to claim 1, comprising an information exchange method for automatically connecting to manufacturer support if no response is found.

[0997] "Application example 2 when combining with an emotional engine"

[0998] (Claim 1)

[0999] A means for capturing voice input and converting the voice into text information,

[1000] A communication means for transmitting the converted character information to a computer device,

[1001] A computer device receives character information and generates an optimal response using an artificial intelligence model for generation;

[1002] A means for converting the generated response into audio data and presenting it through an audio output mechanism,

[1003] A learning procedure that improves response content based on user feedback,

[1004] An emotion recognition means that identifies the emotional state of voice input and adjusts the response,

[1005] A means of making adjustments before presenting a response based on identified emotional states.

[1006] A system that includes this.

[1007] (Claim 2)

[1008] The system according to claim 1, which uses a trained artificial intelligence model for generation configured to generate a response based on information specific to an electrical product from text information.

[1009] (Claim 3)

[1010] The system according to claim 1, comprising a communication protocol for automatically connecting to manufacturer support if no response is found. [Explanation of Symbols]

[1011] 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 device means for acquiring voice input and converting the voice into text data, A communication module means for sending the converted text data to a server, A server receives text data and a processing device that generates an appropriate response using a generative artificial intelligence model. A device means for converting the generated response into audio data and presenting it through an audio output device, A system that includes a learning algorithm that improves response content based on user feedback.

2. The system according to claim 1, which uses a trained generative artificial intelligence model configured to generate a response based on information specific to a home appliance from text data.

3. The system according to claim 1, comprising a communication protocol for automatically connecting to manufacturer support if no response is found.

Images