system

Abstract

Provide a system. 【Solution means】Means for obtaining voice input from a user, Means for converting voice input into text data, Means for analyzing the text data by natural language processing to identify the user's intention, Means for executing a query on a database to obtain search results, Means for organizing the search results and generating text in natural language, Means for converting the text into voice data, Means for outputting the voice data to the user, Means for displaying supplementary information on a display, A system including the above.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In recent years, there has been an increasing demand for an automated system for improving customer service in unmanned commercial facilities and stores. However, the current system has a problem that it is difficult to provide an accurate response to a user's question and appropriate information. As a result, it is difficult for customers to quickly and easily obtain the necessary information, and there are restrictions on improving the quality of the service. In particular, in large-scale commercial facilities, since location information, business hours, etc. of products and services are frequently updated, accurate information provision in real time is required. The purpose of the present invention is to provide an unmanned search tool that efficiently and accurately provides information to customers in order to solve such problems.

Means for Solving the Problems

[0005] The present invention provides a system that includes means for acquiring voice input from a user, means for converting voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for executing queries against a database to acquire search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, and means for displaying supplementary information on a display. This enables an unmanned search tool that allows customers to easily acquire information in commercial facilities and stores. For example, if a user's question concerns the location information of a product or service, the system can acquire the relevant location information from the database and provide it to the user. Also, if a user's question concerns the opening hours of a store, the system can acquire the relevant opening hours information from the database and provide it to the user. This system enables users to acquire necessary information quickly and accurately, thereby improving the quality of service.

[0006] A "user" is a customer or user who uses a system to obtain information.

[0007] "Means for acquiring voice input" refers to system components, including microphones, that are devices or technologies for capturing the user's voice.

[0008] "Means for converting voice input into text data" refers to system components, including speech recognition engines and algorithms, that convert acquired speech into text data, and the devices and technologies used for this purpose.

[0009] "Natural language processing" refers to the techniques and algorithms used to analyze text data and understand its context and meaning.

[0010] "Means for identifying user intent" refers to technologies and algorithms that use natural language processing to identify the specific content of user questions and requests.

[0011] A "database" is a system or storage device that stores the information to be searched.

[0012] "Means of executing queries" refer to devices or technologies that access a database and perform search operations based on user requests.

[0013] "Means of obtaining search results" refer to devices and technologies that receive the results of queries executed from a database and extract the necessary information.

[0014] "Methods for organizing search results" refer to technologies and algorithms for organizing and structuring acquired data into a format that is easy for users to understand.

[0015] "Methods for generating text in natural language" refer to technologies and algorithms for generating appropriate response sentences for users based on organized data.

[0016] "Means for converting text to audio data" refers to devices or technologies for converting generated text into audio data using a speech synthesis engine.

[0017] "Means of outputting audio data to the user" refers to devices or technologies that play back generated audio data through a speaker and provide information to the user.

[0018] "Means for displaying supplementary information on a display" refers to devices or technologies that display supplementary information, such as text or images, on a screen or monitor to provide users with information visually. [Brief explanation of the drawing]

[0019] [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]It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0020] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0021] First, the language used in the following description will be explained.

[0022] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0023] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0024] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0025] 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).

[0026] 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."

[0027] [First Embodiment]

[0028] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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".

[0040] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. The system consists of the following parts:

[0041] (1) Acquisition of voice input

[0042] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is sent to the system through the microphone.

[0043] (2) Text conversion of voice input

[0044] Terminal: The acquired voice data is sent to the internal speech recognition engine. This engine converts the voice into text data. For example, the user's voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0045] (3) Analysis of text data

[0046] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[0047] (4) Querying the database

[0048] Server: Based on the results of parsing the text data, the server executes queries against the database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the location information of coffee makers from the database.

[0049] (5) Organizing search results and generating responses

[0050] Server: Search results retrieved from the database are organized by the server, and a user-friendly text response is generated. For example, based on the retrieved location information, a text response such as "The coffee maker is on the second floor in the home appliance section" is generated.

[0051] (6) Text to speech conversion

[0052] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[0053] (7) Output of audio data to the user

[0054] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0055] Specific example

[0056] For example, if a user asks a robot in a large commercial facility, "Please tell me the opening hours of this store," the process would be as follows:

[0057] 1. Terminal: The microphone captures the user's voice and sends the voice data to the server.

[0058] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0059] 3. Server: A natural language processing engine analyzes the text and identifies the user's intention, such as "I want to know the business hours information."

[0060] 4. Server: Execute an appropriate query on the database (e.g., SELECT hours FROM stores WHERE name='this store';) to retrieve the business hours information.

[0061] 5. Server: Based on the search results, it generates a text response that says, "This store is open every day from 9 am to 9 pm."

[0062] 6. Server: Converts text responses into audio data and sends the audio data to the terminal.

[0063] 7. Terminal: Plays audio data and tells the user through the speaker, "This store is open every day from 9 am to 9 pm." The opening hours are also displayed on the screen.

[0064] This system allows users to quickly and accurately obtain the information they need within commercial facilities.

[0065] The following describes the processing flow.

[0066] Step 1:

[0067] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[0068] Step 2:

[0069] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[0070] Step 3:

[0071] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[0072] Step 4:

[0073] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[0074] Step 5:

[0075] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[0076] Step 6:

[0077] The server identifies the user's intent as "I want to know the location of the product," and then generates a query to the database. For example, it creates an SQL query such as "SELECT location FROM products WHERE name='coffee maker';".

[0078] Step 7:

[0079] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[0080] Step 8:

[0081] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. For example, it might generate text such as, "The coffee maker is located in the home appliance section on the second floor."

[0082] Step 9:

[0083] The server sends the generated text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[0084] Step 10:

[0085] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[0086] Step 11:

[0087] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[0088] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility.

[0089] (Example 1)

[0090] 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."

[0091] Traditional information retrieval systems in commercial facilities and stores had problems that made it difficult for users to obtain information quickly and accurately. In particular, voice-based information retrieval was insufficient, and users often had to perform manual searches. Furthermore, the accuracy of natural language processing was low, and it was sometimes impossible to accurately analyze user intent. In addition, the means of effectively delivering search results to users were limited, resulting in a system that was difficult for users to use.

[0092] 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.

[0093] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for querying a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, and means for displaying supplementary information on a display device. This enables the user to quickly and accurately acquire the necessary information within a commercial facility.

[0094] "Voice input" refers to audio data that a user provides to the system via a microphone.

[0095] "Text data" refers to character information converted from voice input by a speech recognition engine.

[0096] "Natural language processing" is a general term for technologies and methods used to analyze text data and identify user intent.

[0097] A "database" refers to a management system for managing and retrieving information, as well as a collection of structured data used to store that information.

[0098] A "query" is a set of instructions sent to a database to retrieve or manipulate information.

[0099] "Search results" refer to the data obtained when a query is executed against a database.

[0100] "Voice data" refers to digital data used to store the voice responses generated by the system.

[0101] A "display device" is a device used to visually display text data and supplementary information.

[0102] "Supplemental information" refers to additional text, images, or other information displayed on a display device to complement the voice response.

[0103] A "user" refers to a person who provides input for using the system to search for information.

[0104] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. This system automates a series of processes from acquiring voice input to outputting voice responses, and is specifically configured as follows.

[0105] Hardware and software configuration

[0106] Acquiring voice input

[0107] Terminal: The terminal is equipped with a high-sensitivity microphone. When the user provides voice input through this microphone, the voice is stored as digital audio data within the terminal. The terminal has the capability to process the audio data in real time.

[0108] Voice input to text conversion

[0109] Terminal: Voice data is converted into text data using a speech recognition engine such as Google® Cloud Speech-to-Text API. For example, the voice saying "Where is the coffee maker?" is converted into text data and sent to the server.

[0110] Text data analysis

[0111] Server: The server is equipped with the BERT model as its natural language processing engine. Text data is sent to the server and parsed by the BERT model. This identifies the intent behind the user's question.

[0112] Database query

[0113] Server: Based on the results of parsing the text data, the server executes SQL queries against the PostgreSQL database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the necessary information from the database.

[0114] Organizing search results and generating responses

[0115] Server: The server organizes the acquired information and generates a text response in a user-friendly format. For example, it might generate a specific response such as, "The coffee maker is located in the home appliance section on the second floor."

[0116] Text to speech conversion

[0117] Server: The organized text responses are converted into speech data using a speech synthesis engine such as Amazon Polly. This speech data is sent to the terminal in a natural speech format.

[0118] Output of audio data to the user

[0119] Terminal: The generated audio data is played back to the user through the terminal's speaker. Supplementary information (e.g., maps and location information) is also displayed on the display device.

[0120] Specific example

[0121] For example, if a user asks a robot in a large commercial facility, "Could you tell me the opening hours of this store?", the following process will occur.

[0122] 1. Terminal: A high-sensitivity microphone captures the user's voice and stores it as digital audio data within the terminal.

[0123] 2. Terminal: The Google Cloud Speech-to-Text API converts the audio into text data, "Please tell me the opening hours of this store," and sends it to the server.

[0124] 3. Server: The BERT model analyzes the text data and identifies the user's intention, such as "I want to know the business hours information."

[0125] 4. Server: Execute the SQL query "SELECT hours FROM stores WHERE name='this store';" against the PostgreSQL database to retrieve the relevant business hours information.

[0126] 5. Server: Based on the acquired business hours information, it generates a text response that says, "This store is open every day from 9:00 to 21:00."

[0127] 6. Server: Use Amazon Polly to convert text responses into audio data and send it to the terminal.

[0128] 7. Terminal: Plays audio data and provides responses to the user through the speaker. The display shows supplementary information about business hours.

[0129] Example of a prompt

[0130] "Design an information retrieval system for a commercial facility. This system should include seven processing steps: voice input to text conversion, natural language processing analysis, database querying, and voice response generation. Please also specify the names of any relevant hardware and software."

[0131] This system will enable users to quickly and accurately obtain the information they need within commercial facilities.

[0132] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0133] Step 1:

[0134] Terminal: The terminal is equipped with a highly sensitive microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is stored on the terminal as digital audio data. Once this audio data is acquired, the system is ready to proceed to the next processing step. For example, if the user says, "Where is the coffee maker?", their voice will be picked up by the microphone.

[0135] Input: User's voice

[0136] Output: Digital audio data

[0137] Step 2:

[0138] Terminal: Digital voice data is converted to text data using the Google Cloud Speech-to-Text API. This API analyzes the voice input and converts it into text information. In this process, the voice data is transformed into the text "Where is the coffee maker?". After being converted to text data, the data is sent to the server.

[0139] Input: Digital audio data

[0140] Output: Text data

[0141] Step 3:

[0142] Server: The server analyzes the received text data using a BERT model. The server uses a natural language processing engine to identify the user's intent from the text data. For example, the intent "I want to know the location of the coffee maker" is identified. Once the analysis is complete, the analysis results are generated.

[0143] Input: Text data

[0144] Output: Analysis results (user intent)

[0145] Step 4:

[0146] Server: Based on the analysis results, the server executes an SQL query against the PostgreSQL database. During the SQL query generation stage, for example, a query like "SELECT location FROM products WHERE name='coffee maker';" is created. This query is sent to the database, and the necessary information is retrieved. The information returned is the location information of the coffee maker.

[0147] Input: Analysis results (user intent)

[0148] Output: Search results from the database (location information)

[0149] Step 5:

[0150] Server: The server organizes the retrieved search results. Based on the organized data, a user-friendly text response is generated. For example, a response such as "The coffee maker is on the second floor in the home appliance section" is created. Once this response is generated, the process moves on to the next step of speech conversion.

[0151] Input: Search results from the database (location information)

[0152] Output: Text response

[0153] Step 6:

[0154] Server: The server uses a speech synthesis engine such as Amazon Polly to convert the generated text response into speech data. This conversion produces the speech data "The coffee maker is on the second floor in the home appliance section." After this speech data is generated, it is sent to the terminal.

[0155] Input: Text response

[0156] Output: Audio data

[0157] Step 7:

[0158] Terminal: The terminal plays audio data transmitted from the server through its speaker, providing information to the user. Simultaneously, it uses a display device to show a map indicating the coffee maker's location and any necessary supplementary information. The user can hear the audio message, "The coffee maker is in the home appliance section on the second floor," and confirm its specific location on the display.

[0159] Input: Audio data

[0160] Output: Audio output to the user and display of supplementary information.

[0161] (Application Example 1)

[0162] 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."

[0163] In current commercial facilities and stores, customers spend a significant amount of time and effort searching for information such as the location of products and services, opening hours, and promotional campaigns. This compromises customer convenience and prevents an efficient shopping experience within stores. Furthermore, there are limitations to what staff can do to handle searches, and further efficiency improvements are needed, especially in large commercial facilities.

[0164] 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.

[0165] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for executing queries against a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, means for displaying supplementary information on a display, and means for responding to customer questions in a physical store and providing necessary information both audibly and visually. This enables customers to efficiently acquire the necessary information and improve their in-store shopping experience.

[0166] "Means for acquiring voice input" refers to devices or functions for acquiring voice from a user.

[0167] "Means for converting voice input into text data" refers to systems and algorithms for converting acquired voice data into text data.

[0168] "A means of analyzing text data using natural language processing to identify user intent" refers to a technology that analyzes converted text data to identify the information or requests that the user is seeking.

[0169] "A means of executing queries against a database and retrieving search results" refers to a system that queries a database based on a user's request and retrieves the necessary information.

[0170] "Methods for organizing search results and generating text in natural language" refers to technologies for formatting acquired information into an easily understandable form and generating responses in natural language.

[0171] "Means for converting text to audio data" refers to a function for converting generated text data into speech.

[0172] "Means of outputting audio data to the user" refers to devices or functions that provide the converted audio data to the user.

[0173] "Means for displaying supplementary information on a display" refers to technologies for displaying text and image information on a display as an auxiliary means, in addition to voice response.

[0174] "Means of responding to customer questions in physical stores and providing necessary information via voice and visual means" refers to systems or devices that respond to user questions within a physical store and provide necessary information using voice and visual information.

[0175] The embodiments for carrying out this invention will be described in detail below.

[0176] System Overview

[0177] Hardware configuration

[0178] 1. Microphone: Used to obtain user voice input.

[0179] 2. Speaker: Used to provide audio output to the user.

[0180] 3. Display: Used to display auxiliary information.

[0181] 4. Network connection: Used to communicate with the server.

[0182] Software Configuration

[0183] 1. Speech recognition engine: As an example, we will use the Google Cloud Speech-to-Text API.

[0184] 2. Natural Language Processing Engine: As an example, we will use OpenAI® GPT-4®.

[0185] 3. Database management system: PostgreSQL will be used as an example.

[0186] 4. Text-to-Speech Engine: As an example, we will use the Google Cloud Text-to-Speech API.

[0187] System operation

[0188] 1. Acquisition of voice input

[0189] When a user speaks to the robot, the microphone captures the audio and sends the audio data to the server.

[0190] 2. Text conversion of voice input

[0191] The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API).

[0192] 3. Analysis of text data

[0193] The text data is analyzed by a natural language processing engine (e.g., OpenAI GPT-4) on the server to identify the user's intent.

[0194] 4. Querying the database

[0195] Based on the results of the text data analysis, the server executes queries against the database (e.g., PostgreSQL) to retrieve the necessary information.

[0196] 5. Organizing search results and generating responses

[0197] The server organizes the acquired information and generates a text response in natural language.

[0198] 6. Text to Speech Conversion

[0199] The server converts the text response into speech data using a speech synthesis engine (e.g., Google Cloud Text-to-Speech API).

[0200] 7. Output of audio data to the user

[0201] The converted audio data is played back to the user through the speaker. Supplementary information is also displayed on the screen.

[0202] Specific example

[0203] For example, if a user asks a supermarket robot, "Where can I find the sale milk?", the robot would process it as follows:

[0204] Example of a prompt

[0205] User: "Where can I find milk on sale?"

[0206] System: "Recognizes speech data and converts it to text."

[0207] GPT-4: "Users want to know where the discounted milk is located."

[0208] Server: "SELECT location FROM products WHERE name='Special Sale Milk';"

[0209] Database: "It's located at the far left of the refrigerated section."

[0210] Natural language generation: "The discounted milk is on the far left of the refrigerated section."

[0211] Google Cloud Text-to-Speech API: "The discounted milk is on the far left of the refrigerated section."

[0212] Speaker: (Voice) "The sale milk is located at the far left of the refrigerated section."

[0213] Display: (Shows the location of the refrigerated section on a map)

[0214] In this way, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[0215] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0216] Step 1:

[0217] The user speaks into the device. The device's microphone picks up the user's voice input. For example, if the user says, "Where can I find the sale milk?", that voice data is acquired. Input: User's voice. Output: Voice data.

[0218] Step 2:

[0219] The device sends the acquired audio data to the server. The server uses a speech recognition engine (Google Cloud Speech-to-Text API) to convert the audio data into text data. Specifically, the server processes the audio data and generates the text "Where can I find the discounted milk?". Input: Audio data. Output: Text data.

[0220] Step 3:

[0221] The server passes the acquired text data to a natural language processing engine (OpenAI GPT-4) for analysis. Here, the user's intent is identified as "I want to know the location of the discounted milk." Input: Text data. Output: User's intent.

[0222] Step 4:

[0223] The server generates and executes a query against the database (PostgreSQL) based on the identified user's intent. In this case, the query executed is "SELECT location FROM products WHERE name='Special Offer Milk';". Input: User intent. Output: Query result.

[0224] Step 5:

[0225] The server organizes the search results retrieved from the database and generates a text response in natural language. Specifically, it generates a text response such as, "The discounted milk is on the far left of the refrigerated section." Input: Query results. Output: Text response.

[0226] Step 6:

[0227] The server converts the generated text response into speech data using a text-to-speech engine (Google Cloud Text-to-Speech API). Specifically, the server generates the speech data "The discounted milk is on the far left of the refrigerated section." Input: Text response. Output: Speech data.

[0228] Step 7:

[0229] The terminal plays audio data received from the server through its speaker and provides a response to the user. It also displays the location of the refrigerated section on a map as supplementary information. Input: Audio data and supplementary information. Output: Audio response and visual information.

[0230] Through these steps, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[0231] 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.

[0232] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. The system consists of the following parts:

[0233] (1) Acquisition of voice input

[0234] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone.

[0235] (2) Text conversion of voice input

[0236] Terminal: The acquired voice data is sent to the speech recognition engine. This engine converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0237] (3) Analysis of text data

[0238] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[0239] (4) Recognition of emotions

[0240] Server: Text and audio data are sent to the emotion engine. The emotion engine analyzes the user's emotions from their voice tone and text content. For example, it recognizes emotions such as when the user is in a hurry or is frustrated.

[0241] (5) Querying the database

[0242] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[0243] (6) Organizing search results and generating responses

[0244] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[0245] (7) Text to speech conversion

[0246] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[0247] (8) Output of audio data to the user

[0248] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0249] Specific example

[0250] Example 1: Product Search

[0251] When a user asks, "Where is the coffee maker?", the following process takes place:

[0252] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[0253] 2. Server: Converts audio data into text data and analyzes it.

[0254] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[0255] 4. Server: Executes queries on the database to obtain location information.

[0256] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[0257] 6. Terminal: Plays audio data and displays supplementary information on the display.

[0258] Example 2: Inquiry about business hours

[0259] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[0260] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[0261] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0262] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[0263] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[0264] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[0265] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[0266] This system will not only allow users to obtain quick and accurate information within commercial facilities, but also provide them with friendly, emotionally resonant responses.

[0267] The following describes the processing flow.

[0268] Step 1:

[0269] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[0270] Step 2:

[0271] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[0272] Step 3:

[0273] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[0274] Step 4:

[0275] The server receives voice data from the voice recognition engine and converts it into text data. Specifically, text data such as "Where is the coffee maker?" is generated.

[0276] Step 5:

[0277] The server sends the generated text data to the natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intention.

[0278] Step 6:

[0279] The server sends the text data to the emotion engine to analyze the user's emotion. The emotion engine identifies the user's emotion from the tone of the voice and the content of the text.

[0280] Step 7:

[0281] Based on the user's intention and emotion, the server generates a query to the database. For example, create an SQL query like "SELECT location FROM products WHERE name='coffee maker';".

[0282] Step 8:

[0283] The server sends the generated SQL query to the database and executes the query. Location information of the coffee maker is returned from the database.

[0284] Step 9:

[0285] The server organizes the location information obtained from the database and generates a text response in an easy-to-understand form for the user. Considering the emotion recognition result, the tone and content of the response are adjusted.

[0286] Step 10:

[0287] The server sends the text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[0288] Step 11:

[0289] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[0290] Step 12:

[0291] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[0292] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility. Furthermore, the emotion engine provides friendly responses that are tailored to the user's emotions.

[0293] (Example 2)

[0294] 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".

[0295] Conventional unmanned search tools provide location and business hours information based on user voice input, but they lack the ability to generate responses that take user emotions into account, resulting in a limited user experience. Improving the speed and accuracy of information delivery is also a challenge.

[0296] 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.

[0297] In this invention, the server includes means for recognizing the user's emotions from voice input and text data, means for organizing search results and generating text in natural language based on the emotion recognition results, and means for converting text into voice data. This enables detailed responses tailored to the user's situation and emotions, improving the user experience and providing information quickly and accurately.

[0298] A "user" refers to the end user who attempts to obtain information using this system.

[0299] "Voice input" refers to the process of acquiring the voice a user speaks into a device as digital data.

[0300] "Text data" refers to data in string format converted by speech recognition.

[0301] "Natural language processing" refers to the technology of analyzing text data to identify its meaning and intent.

[0302] "Emotion recognition" refers to the technology that analyzes and identifies a user's emotional state from voice data and text data.

[0303] A "database" refers to a computerized information management system in which various types of information are systematically organized and stored.

[0304] A "query" refers to a set of commands executed on a database for searching or performing operations.

[0305] "Search results" refer to information obtained by executing queries against a database.

[0306] "Speech synthesis" refers to the technology that converts text data into speech data.

[0307] "Audio data" refers to audio information generated by speech synthesis and output to the user.

[0308] "Supplementary information" refers to additional information (e.g., maps and location information) that is visually displayed and added to the voice response.

[0309] The system of the present invention is a self-service search tool aimed at improving customer service in commercial facilities and stores, and also incorporates an emotion engine that recognizes and responds to the emotions of users. This system is composed of the following parts.

[0310] (1) Acquisition of voice input

[0311] Terminal: A microphone is installed in the terminal, and this is used to acquire the voice input of the user. When the user states a question or request, the voice is transmitted to the system through the microphone. For example, when the user asks, "Where is the coffee maker?", the voice is acquired by the terminal.

[0312] (2) Text conversion of voice input

[0313] Terminal: The acquired voice data is transmitted to a voice recognition engine (e.g., a voice recognition API), and this engine converts the voice into character data. For example, the voice "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0314] (3) Analysis of text data

[0315] Server: The text data is transmitted to the server and analyzed by a natural language processing engine (e.g., a natural language processing API). Through the analysis, the intention of the user's question is identified. For example, the identification "Want to know the location of the coffee maker" is made.

[0316] (4) Recognition of emotions

[0317] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes the user's emotions from the tone of their voice and the content of their text. For example, it can recognize emotions such as when the user is in a hurry or is frustrated.

[0318] (5) Querying the database

[0319] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[0320] (6) Organizing search results and generating responses

[0321] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[0322] (7) Text to speech conversion

[0323] Server: The organized text responses are converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This data is expressed as natural speech corresponding to the user's question.

[0324] (8) Output of audio data to the user

[0325] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0326] Specific example

[0327] Example 1: Product Search

[0328] When a user asks, "Where is the coffee maker?", the following process takes place:

[0329] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[0330] 2. Server: Converts audio data into text data and analyzes it.

[0331] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[0332] 4. Server: Executes queries on the database to obtain location information.

[0333] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[0334] 6. Terminal: Plays audio data and displays supplementary information on the display.

[0335] Example 2: Inquiry about business hours

[0336] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[0337] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[0338] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0339] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[0340] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[0341] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[0342] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[0343] Example prompts for use with a generative AI model

[0344] "Please explain the specific processing flow when a user asks, 'Where is the coffee maker?'"

[0345] "Please explain how the system recognizes the user's emotions and how it generates responses based on those emotions."

[0346] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0347] Step 1:

[0348] Terminal: The terminal is equipped with a microphone, and the user provides voice input. For example, the user might ask, "Where is the coffee maker?" This voice data is acquired in digital format and sent to the next processing step. The input is the user's voice data, and the output is voice data in digital format.

[0349] Step 2:

[0350] Terminal: The acquired digital voice data is sent to a speech recognition engine (e.g., a speech recognition API). This engine analyzes the voice data and converts it into text data. For example, the voice data "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?". The input is voice data, and the output is text data.

[0351] Step 3:

[0352] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker. The input is text data, and the output is the intent of the user's question.

[0353] Step 4:

[0354] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes emotions from the tone of voice and the content of the text. For example, it recognizes emotions such as when the user is in a hurry or is frustrated. The input is text and audio data, and the output is the user's emotional state.

[0355] Step 5:

[0356] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it generates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database. The input is the analysis results and sentiment recognition results, and the output is the location information.

[0357] Step 6:

[0358] Server: Organizes location information obtained from the database and generates text responses in natural language based on sentiment recognition results. For example, it might generate a text response such as, "We will let you know shortly. The coffee maker is on the second floor in the home appliance section." The input is location information and sentiment recognition results, and the output is a text response.

[0359] Step 7:

[0360] Server: The generated text response is converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This conversion creates speech data that is natural-sounding, corresponding to the user's question. The input is a text response, and the output is speech data.

[0361] Step 8:

[0362] Terminal: The created audio data is sent to the terminal and played back to the user through the speaker. Additionally, supplementary information such as location data is displayed on the screen. Input is audio data and supplementary information, while output is playback of the audio response and display on the screen.

[0363] (Application Example 2)

[0364] 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".

[0365] In modern commercial facilities and stores, enabling users to quickly and accurately obtain information about the products and services they need is crucial for improving customer satisfaction. However, many customers may prefer not to ask store staff directly, and staff are not always available to assist. Therefore, there is a need for unmanned systems that efficiently answer customer questions. Furthermore, systems that can respond in accordance with customer emotions are expected to further enhance the customer experience.

[0366] 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. In this invention, the server includes means for analyzing the user's voice data and text data and recognizing the user's emotions, means for executing queries against the database and obtaining search results, and means for organizing the search results and generating text in natural language according to the emotions. This enables appropriate and rapid responses to user questions, as well as friendly responses that are tailored to the user's emotions.

[0367] "Means for obtaining voice input from the user" refers to devices or programs for collecting voices spoken by the user, and includes microphones and voice input interfaces.

[0368] "Means of converting voice input into text data" refers to technologies and programs that convert acquired voice into string information, and this includes speech recognition engines and their algorithms.

[0369] "Methods for analyzing text data using natural language processing to identify user intent" refers to technologies and programs that linguistically analyze string information to understand the intent behind user requests and questions, and includes natural language processing (NLP) engines.

[0370] "Means of analyzing user voice and text data to recognize user emotions" refers to technologies and programs that estimate psychological states and emotions based on the characteristics of voice and text, and emotion recognition engines are included in this.

[0371] "A means of executing queries against a database and retrieving search results" refers to the technology or program that sends commands to a database to obtain specific information and receives the response, and this includes SQL queries.

[0372] "Means for organizing search results and generating text in natural language according to emotions" refers to technologies and programs that organize acquired information into a certain structure and generate text in a way that aligns with the user's emotions, and response generation engines are included in this.

[0373] "Means of converting text into audio data" refers to technologies and programs that convert textual information into audio information, and this includes text-to-speech (TTS) engines.

[0374] "Means of outputting audio data to the user" refers to devices or programs that play back the generated audio information, and this includes speakers and audio playback functions.

[0375] "Means of displaying supplementary information on a display" refers to devices or programs that visually represent content in addition to information provided by voice, and this includes displays and screen display functions.

[0376] This invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. A detailed embodiment of this system is described below.

[0377] System Configuration

[0378] The system primarily consists of terminals that acquire user voice input and servers that perform data analysis and response generation.

[0379] Hardware and software to be used

[0380] Hardware: Smartphones, microphones, speakers, displays, servers

[0381] software:

[0382] Speech recognition: Google Cloud Speech-to-Text API

[0383] Natural Language Processing (NLP): NLTK (Python's natural language processing library)

[0384] Emotion Recognition: IBM Watson® Natural Language Understanding

[0385] Database: MySQL (registered trademark)

[0386] Text-to-Speech (TTS): Amazon Polly

[0387] System Operation Overview

[0388] 1. Acquisition of voice input

[0389] The user speaks into a smartphone app, and the voice is captured by the microphone.

[0390] Example: The user asks, "Where is the coffee maker?"

[0391] 2. Text conversion of voice input

[0392] The acquired audio is converted into text data using the Google Cloud Speech-to-Text API.

[0393] This means that the audio "Where is the coffee maker?" will be converted to text "Where is the coffee maker?".

[0394] 3. Analysis of text data

[0395] The server analyzes the generated text data using NLTK to identify the user's intent.

[0396] The analysis results identify the user's intention: "I want to know the location of the product (coffee maker)."

[0397] 4. Recognition of emotions

[0398] The server uses IBM Watson to recognize emotions from the user's text data and voice tone.

[0399] Emotion recognition can sometimes reveal that the user is in a hurry.

[0400] 5. Querying the database

[0401] The server queries the MySQL database to retrieve location information for the relevant product or service.

[0402] The query result retrieves the location information of the coffee maker (e.g., the home appliance section on the second floor).

[0403] 6. Organizing search results and generating responses

[0404] The server organizes the acquired location information and generates natural responses based on the user's emotions.

[0405] For example, for a user in a hurry, a response such as "I'll be right there. The coffee maker is on the second floor in the home appliance section" will be generated.

[0406] 7. Text to Speech Conversion

[0407] The server converts the generated text response into audio data using Amazon Polly.

[0408] This converts the text "The coffee maker is in the home appliance section on the second floor" into audio data.

[0409] 8. Output of audio data to the user

[0410] Audio data is played through the smartphone's speaker, providing information to the user.

[0411] Furthermore, supplementary information (e.g., maps and location information) is displayed on the smartphone screen.

[0412] Specific example

[0413] As a concrete example, the following prompt statements could be used:

[0414] Example of a prompt:

[0415] The user is asking "Where is the coffee maker?" via voice. Convert the voice to text, analyze it, and obtain the product's location information. Also, recognize the user's emotions and respond appropriately.

[0416] Based on this prompt, the system can generate appropriate answers to the user's questions and provide information quickly and accurately.

[0417] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0418] Step 1:

[0419] The user speaks into their smartphone, and voice input is obtained. The user states a question or request, and the device's (smartphone's) microphone captures the voice. The input is the user's voice, and the output is the captured voice data.

[0420] Step 2:

[0421] The acquired audio data is converted into text data using the Google Cloud Speech-to-Text API. The device sends the audio data to the speech recognition engine in the cloud and receives the text data in return. The input is audio data, and the output is the converted text data.

[0422] Step 3:

[0423] The server analyzes the received text data using NLTK to identify the user's intent. The server sends the text data to a natural language processing engine, which then identifies the user's request (e.g., "I want to know the location of the product") as a result of the analysis. The input is text data, and the output is the analysis result (user's intent).

[0424] Step 4:

[0425] The server analyzes text and audio data using IBM Watson to recognize the user's emotions. The server estimates emotions from the tone of voice and the content of the text. The input is text and audio data, and the output is recognized emotion data.

[0426] Step 5:

[0427] The server executes queries against the MySQL database and retrieves search results. Based on the user's intent and sentiment, the server executes queries to request information from the database. The input is the query information, and the output is the search results retrieved from the database (e.g., product location information).

[0428] Step 6:

[0429] The server organizes search results and generates text responses in natural language based on sentiment. The server then creates appropriate responses for the user based on the information it retrieves. Inputs are search results and sentiment data, while output is the generated text response.

[0430] Step 7:

[0431] The generated text response is converted into speech data using Amazon Polly. The server sends the text response to the speech synthesis engine and receives the speech data. The input is the text response, and the output is the converted speech data.

[0432] Step 8:

[0433] The device plays the received audio data and displays supplementary information on the display. The device plays audio through the speaker and displays related information on the display. The input is audio data and supplementary information, and the output is the audio and display content provided to the user.

[0434] 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.

[0435] 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.

[0436] 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.

[0437] [Second Embodiment]

[0438] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0439] 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.

[0440] 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).

[0441] 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.

[0442] 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.

[0443] 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).

[0444] 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.

[0445] 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.

[0446] 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.

[0447] 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.

[0448] 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.

[0449] 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".

[0450] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. The system consists of the following parts:

[0451] (1) Acquisition of voice input

[0452] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is sent to the system through the microphone.

[0453] (2) Text conversion of voice input

[0454] Terminal: The acquired voice data is sent to the internal speech recognition engine. This engine converts the voice into text data. For example, the user's voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0455] (3) Analysis of text data

[0456] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[0457] (4) Querying the database

[0458] Server: Based on the results of parsing the text data, the server executes queries against the database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the location information of coffee makers from the database.

[0459] (5) Organizing search results and generating responses

[0460] Server: Search results retrieved from the database are organized by the server, and a user-friendly text response is generated. For example, based on the retrieved location information, a text response such as "The coffee maker is on the second floor in the home appliance section" is generated.

[0461] (6) Text to speech conversion

[0462] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[0463] (7) Output of audio data to the user

[0464] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0465] Specific example

[0466] For example, if a user asks a robot in a large commercial facility, "Please tell me the opening hours of this store," the process would be as follows:

[0467] 1. Terminal: The microphone captures the user's voice and sends the voice data to the server.

[0468] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0469] 3. Server: A natural language processing engine analyzes the text and identifies the user's intention, such as "I want to know the business hours information."

[0470] 4. Server: Execute an appropriate query on the database (e.g., SELECT hours FROM stores WHERE name='this store';) to retrieve the business hours information.

[0471] 5. Server: Based on the search results, it generates a text response that says, "This store is open every day from 9 am to 9 pm."

[0472] 6. Server: Converts text responses into audio data and sends the audio data to the terminal.

[0473] 7. Terminal: Plays audio data and tells the user through the speaker, "This store is open every day from 9 am to 9 pm." The opening hours are also displayed on the screen.

[0474] This system allows users to quickly and accurately obtain the information they need within commercial facilities.

[0475] The following describes the processing flow.

[0476] Step 1:

[0477] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[0478] Step 2:

[0479] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[0480] Step 3:

[0481] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[0482] Step 4:

[0483] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[0484] Step 5:

[0485] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[0486] Step 6:

[0487] The server identifies the user's intent as "I want to know the location of the product," and then generates a query to the database. For example, it creates an SQL query such as "SELECT location FROM products WHERE name='coffee maker';".

[0488] Step 7:

[0489] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[0490] Step 8:

[0491] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. For example, it might generate text such as, "The coffee maker is located in the home appliance section on the second floor."

[0492] Step 9:

[0493] The server sends the generated text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[0494] Step 10:

[0495] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[0496] Step 11:

[0497] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[0498] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility.

[0499] (Example 1)

[0500] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0501] Traditional information retrieval systems in commercial facilities and stores had problems that made it difficult for users to obtain information quickly and accurately. In particular, voice-based information retrieval was insufficient, and users often had to perform manual searches. Furthermore, the accuracy of natural language processing was low, and it was sometimes impossible to accurately analyze user intent. In addition, the means of effectively delivering search results to users were limited, resulting in a system that was difficult for users to use.

[0502] 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.

[0503] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for querying a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, and means for displaying supplementary information on a display device. This enables the user to quickly and accurately acquire the necessary information within a commercial facility.

[0504] "Voice input" refers to audio data that a user provides to the system via a microphone.

[0505] "Text data" refers to character information converted from voice input by a speech recognition engine.

[0506] "Natural language processing" is a general term for technologies and methods used to analyze text data and identify user intent.

[0507] A "database" refers to a management system for managing and retrieving information, as well as a collection of structured data used to store that information.

[0508] A "query" is a set of instructions sent to a database to retrieve or manipulate information.

[0509] "Search results" refer to the data obtained when a query is executed against a database.

[0510] "Voice data" refers to digital data used to store the voice responses generated by the system.

[0511] A "display device" is a device used to visually display text data and supplementary information.

[0512] "Supplemental information" refers to additional text, images, or other information displayed on a display device to complement the voice response.

[0513] A "user" refers to a person who provides input for using the system to search for information.

[0514] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. This system automates a series of processes from acquiring voice input to outputting voice responses, and is specifically configured as follows.

[0515] Hardware and software configuration

[0516] Acquiring voice input

[0517] Terminal: The terminal is equipped with a high-sensitivity microphone. When the user provides voice input through this microphone, the voice is stored as digital audio data within the terminal. The terminal has the capability to process the audio data in real time.

[0518] Voice input to text conversion

[0519] Terminal: Voice data is converted into text data using a speech recognition engine such as the Google Cloud Speech-to-Text API. For example, the voice saying "Where is the coffee maker?" is converted into text data and sent to the server.

[0520] Text data analysis

[0521] Server: The server is equipped with the BERT model as its natural language processing engine. Text data is sent to the server and parsed by the BERT model. This identifies the intent behind the user's question.

[0522] Database query

[0523] Server: Based on the results of parsing the text data, the server executes SQL queries against the PostgreSQL database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the necessary information from the database.

[0524] Organizing search results and generating responses

[0525] Server: The server organizes the acquired information and generates a text response in a user-friendly format. For example, it might generate a specific response such as, "The coffee maker is located in the home appliance section on the second floor."

[0526] Text to speech conversion

[0527] Server: The organized text responses are converted into speech data using a speech synthesis engine such as Amazon Polly. This speech data is sent to the terminal in a natural speech format.

[0528] Output of audio data to the user

[0529] Terminal: The generated audio data is played back to the user through the terminal's speaker. Supplementary information (e.g., maps and location information) is also displayed on the display device.

[0530] Specific example

[0531] For example, if a user asks a robot in a large commercial facility, "Could you tell me the opening hours of this store?", the following process will occur.

[0532] 1. Terminal: A high-sensitivity microphone captures the user's voice and stores it as digital audio data within the terminal.

[0533] 2. Terminal: The Google Cloud Speech-to-Text API converts the audio into text data, "Please tell me the opening hours of this store," and sends it to the server.

[0534] 3. Server: The BERT model analyzes the text data and identifies the user's intention, such as "I want to know the business hours information."

[0535] 4. Server: Execute the SQL query "SELECT hours FROM stores WHERE name='this store';" against the PostgreSQL database to retrieve the relevant business hours information.

[0536] 5. Server: Based on the acquired business hours information, it generates a text response that says, "This store is open every day from 9:00 to 21:00."

[0537] 6. Server: Use Amazon Polly to convert text responses into audio data and send it to the terminal.

[0538] 7. Terminal: Plays audio data and provides responses to the user through the speaker. The display shows supplementary information about business hours.

[0539] Example of a prompt

[0540] "Design an information retrieval system for a commercial facility. This system should include seven processing steps: voice input to text conversion, natural language processing analysis, database querying, and voice response generation. Please also specify the names of any relevant hardware and software."

[0541] This system will enable users to quickly and accurately obtain the information they need within commercial facilities.

[0542] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0543] Step 1:

[0544] Terminal: The terminal is equipped with a highly sensitive microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is stored on the terminal as digital audio data. Once this audio data is acquired, the system is ready to proceed to the next processing step. For example, if the user says, "Where is the coffee maker?", their voice will be picked up by the microphone.

[0545] Input: User's voice

[0546] Output: Digital audio data

[0547] Step 2:

[0548] Terminal: Digital voice data is converted to text data using the Google Cloud Speech-to-Text API. This API analyzes the voice input and converts it into text information. In this process, the voice data is transformed into the text "Where is the coffee maker?". After being converted to text data, the data is sent to the server.

[0549] Input: Digital audio data

[0550] Output: Text data

[0551] Step 3:

[0552] Server: The server analyzes the received text data using a BERT model. The server uses a natural language processing engine to identify the user's intent from the text data. For example, the intent "I want to know the location of the coffee maker" is identified. Once the analysis is complete, the analysis results are generated.

[0553] Input: Text data

[0554] Output: Analysis results (user intent)

[0555] Step 4:

[0556] Server: Based on the analysis results, the server executes an SQL query against the PostgreSQL database. During the SQL query generation stage, for example, a query like "SELECT location FROM products WHERE name='coffee maker';" is created. This query is sent to the database, and the necessary information is retrieved. The information returned is the location information of the coffee maker.

[0557] Input: Analysis results (user intent)

[0558] Output: Search results from the database (location information)

[0559] Step 5:

[0560] Server: The server organizes the retrieved search results. Based on the organized data, a user-friendly text response is generated. For example, a response such as "The coffee maker is on the second floor in the home appliance section" is created. Once this response is generated, the process moves on to the next step of speech conversion.

[0561] Input: Search results from the database (location information)

[0562] Output: Text response

[0563] Step 6:

[0564] Server: The server uses a speech synthesis engine such as Amazon Polly to convert the generated text response into speech data. This conversion produces the speech data "The coffee maker is on the second floor in the home appliance section." After this speech data is generated, it is sent to the terminal.

[0565] Input: Text response

[0566] Output: Audio data

[0567] Step 7:

[0568] Terminal: The terminal plays audio data transmitted from the server through its speaker, providing information to the user. Simultaneously, it uses a display device to show a map indicating the coffee maker's location and any necessary supplementary information. The user can hear the audio message, "The coffee maker is in the home appliance section on the second floor," and confirm its specific location on the display.

[0569] Input: Audio data

[0570] Output: Audio output to the user and display of supplementary information.

[0571] (Application Example 1)

[0572] 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."

[0573] In current commercial facilities and stores, customers spend a significant amount of time and effort searching for information such as the location of products and services, opening hours, and promotional campaigns. This compromises customer convenience and prevents an efficient shopping experience within stores. Furthermore, there are limitations to what staff can do to handle searches, and further efficiency improvements are needed, especially in large commercial facilities.

[0574] 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.

[0575] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for executing queries against a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, means for displaying supplementary information on a display, and means for responding to customer questions in a physical store and providing necessary information both audibly and visually. This enables customers to efficiently acquire the necessary information and improve their in-store shopping experience.

[0576] "Means for acquiring voice input" refers to devices or functions for acquiring voice from a user.

[0577] "Means for converting voice input into text data" refers to systems and algorithms for converting acquired voice data into text data.

[0578] "A means of analyzing text data using natural language processing to identify user intent" refers to a technology that analyzes converted text data to identify the information or requests that the user is seeking.

[0579] "A means of executing queries against a database and retrieving search results" refers to a system that queries a database based on a user's request and retrieves the necessary information.

[0580] "Methods for organizing search results and generating text in natural language" refers to technologies for formatting acquired information into an easily understandable form and generating responses in natural language.

[0581] "Means for converting text to audio data" refers to a function for converting generated text data into speech.

[0582] "Means of outputting audio data to the user" refers to devices or functions that provide the converted audio data to the user.

[0583] "Means for displaying supplementary information on a display" refers to technologies for displaying text and image information on a display as an auxiliary means, in addition to voice response.

[0584] "Means of responding to customer questions in physical stores and providing necessary information via voice and visual means" refers to systems or devices that respond to user questions within a physical store and provide necessary information using voice and visual information.

[0585] The embodiments for carrying out this invention will be described in detail below.

[0586] System Overview

[0587] Hardware configuration

[0588] 1. Microphone: Used to obtain user voice input.

[0589] 2. Speaker: Used to provide audio output to the user.

[0590] 3. Display: Used to display auxiliary information.

[0591] 4. Network connection: Used to communicate with the server.

[0592] Software Configuration

[0593] 1. Speech recognition engine: As an example, we will use the Google Cloud Speech-to-Text API.

[0594] 2. Natural Language Processing Engine: OpenAI GPT-4 will be used as an example.

[0595] 3. Database management system: PostgreSQL will be used as an example.

[0596] 4. Text-to-Speech Engine: As an example, we will use the Google Cloud Text-to-Speech API.

[0597] System operation

[0598] 1. Acquisition of voice input

[0599] When a user speaks to the robot, the microphone captures the audio and sends the audio data to the server.

[0600] 2. Text conversion of voice input

[0601] The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API).

[0602] 3. Analysis of text data

[0603] The text data is analyzed by a natural language processing engine (e.g., OpenAI GPT-4) on the server to identify the user's intent.

[0604] 4. Querying the database

[0605] Based on the results of the text data analysis, the server executes queries against the database (e.g., PostgreSQL) to retrieve the necessary information.

[0606] 5. Organizing search results and generating responses

[0607] The server organizes the acquired information and generates a text response in natural language.

[0608] 6. Text to Speech Conversion

[0609] The server converts the text response into speech data using a speech synthesis engine (e.g., Google Cloud Text-to-Speech API).

[0610] 7. Output of audio data to the user

[0611] The converted audio data is played back to the user through the speaker. Supplementary information is also displayed on the screen.

[0612] Specific example

[0613] For example, if a user asks a supermarket robot, "Where can I find the sale milk?", the robot would process it as follows:

[0614] Example of a prompt

[0615] User: "Where can I find milk on sale?"

[0616] System: "Recognizes speech data and converts it to text."

[0617] GPT-4: "Users want to know where the discounted milk is located."

[0618] Server: "SELECT location FROM products WHERE name='Special Sale Milk';"

[0619] Database: "It's located at the far left of the refrigerated section."

[0620] Natural language generation: "The discounted milk is on the far left of the refrigerated section."

[0621] Google Cloud Text-to-Speech API: "The discounted milk is on the far left of the refrigerated section."

[0622] Speaker: (Voice) "The sale milk is located at the far left of the refrigerated section."

[0623] Display: (Shows the location of the refrigerated section on a map)

[0624] In this way, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[0625] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0626] Step 1:

[0627] The user speaks into the device. The device's microphone picks up the user's voice input. For example, if the user says, "Where can I find the sale milk?", that voice data is acquired. Input: User's voice. Output: Voice data.

[0628] Step 2:

[0629] The device sends the acquired audio data to the server. The server uses a speech recognition engine (Google Cloud Speech-to-Text API) to convert the audio data into text data. Specifically, the server processes the audio data and generates the text "Where can I find the discounted milk?". Input: Audio data. Output: Text data.

[0630] Step 3:

[0631] The server passes the acquired text data to a natural language processing engine (OpenAI GPT-4) for analysis. Here, the user's intent is identified as "I want to know the location of the discounted milk." Input: Text data. Output: User's intent.

[0632] Step 4:

[0633] The server generates and executes a query against the database (PostgreSQL) based on the identified user's intent. In this case, the query executed is "SELECT location FROM products WHERE name='Special Offer Milk';". Input: User intent. Output: Query result.

[0634] Step 5:

[0635] The server organizes the search results retrieved from the database and generates a text response in natural language. Specifically, it generates a text response such as, "The discounted milk is on the far left of the refrigerated section." Input: Query results. Output: Text response.

[0636] Step 6:

[0637] The server converts the generated text response into speech data using a text-to-speech engine (Google Cloud Text-to-Speech API). Specifically, the server generates the speech data "The discounted milk is on the far left of the refrigerated section." Input: Text response. Output: Speech data.

[0638] Step 7:

[0639] The terminal plays audio data received from the server through its speaker and provides a response to the user. It also displays the location of the refrigerated section on a map as supplementary information. Input: Audio data and supplementary information. Output: Audio response and visual information.

[0640] Through these steps, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[0641] 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.

[0642] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. The system consists of the following parts:

[0643] (1) Acquisition of voice input

[0644] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone.

[0645] (2) Text conversion of voice input

[0646] Terminal: The acquired voice data is sent to the speech recognition engine. This engine converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0647] (3) Analysis of text data

[0648] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[0649] (4) Recognition of emotions

[0650] Server: Text and audio data are sent to the emotion engine. The emotion engine analyzes the user's emotions from their voice tone and text content. For example, it recognizes emotions such as when the user is in a hurry or is frustrated.

[0651] (5) Querying the database

[0652] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[0653] (6) Organizing search results and generating responses

[0654] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[0655] (7) Text to speech conversion

[0656] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[0657] (8) Output of audio data to the user

[0658] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0659] Specific example

[0660] Example 1: Product Search

[0661] When a user asks, "Where is the coffee maker?", the following process takes place:

[0662] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[0663] 2. Server: Converts audio data into text data and analyzes it.

[0664] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[0665] 4. Server: Executes queries on the database to obtain location information.

[0666] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[0667] 6. Terminal: Plays audio data and displays supplementary information on the display.

[0668] Example 2: Inquiry about business hours

[0669] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[0670] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[0671] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0672] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[0673] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[0674] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[0675] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[0676] This system will not only allow users to obtain quick and accurate information within commercial facilities, but also provide them with friendly, emotionally resonant responses.

[0677] The following describes the processing flow.

[0678] Step 1:

[0679] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[0680] Step 2:

[0681] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[0682] Step 3:

[0683] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[0684] Step 4:

[0685] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[0686] Step 5:

[0687] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[0688] Step 6:

[0689] The server sends text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies the user's emotions from the tone of voice and the content of the text.

[0690] Step 7:

[0691] The server generates queries to the database based on the user's intent and feelings. For example, it might create an SQL query like "SELECT location FROM products WHERE name='coffee maker';".

[0692] Step 8:

[0693] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[0694] Step 9:

[0695] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. It also adjusts the tone and content of the response, taking into account the sentiment recognition results.

[0696] Step 10:

[0697] The server sends the text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[0698] Step 11:

[0699] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[0700] Step 12:

[0701] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[0702] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility. Furthermore, the emotion engine provides friendly responses that are tailored to the user's emotions.

[0703] (Example 2)

[0704] 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".

[0705] Conventional unmanned search tools provide location and business hours information based on user voice input, but they lack the ability to generate responses that take user emotions into account, resulting in a limited user experience. Improving the speed and accuracy of information delivery is also a challenge.

[0706] 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.

[0707] In this invention, the server includes means for recognizing the user's emotions from voice input and text data, means for organizing search results and generating text in natural language based on the emotion recognition results, and means for converting text into voice data. This enables detailed responses tailored to the user's situation and emotions, improving the user experience and providing information quickly and accurately.

[0708] A "user" refers to the end user who attempts to obtain information using this system.

[0709] "Voice input" refers to the process of acquiring the voice a user speaks into a device as digital data.

[0710] "Text data" refers to data in string format converted by speech recognition.

[0711] "Natural language processing" refers to the technology of analyzing text data to identify its meaning and intent.

[0712] "Emotion recognition" refers to the technology that analyzes and identifies a user's emotional state from voice data and text data.

[0713] A "database" refers to a computerized information management system in which various types of information are systematically organized and stored.

[0714] A "query" refers to a set of commands executed on a database for searching or performing operations.

[0715] "Search results" refer to information obtained by executing queries against a database.

[0716] "Speech synthesis" refers to the technology that converts text data into speech data.

[0717] "Audio data" refers to audio information generated by speech synthesis and output to the user.

[0718] "Supplementary information" refers to additional visually displayed information (e.g., maps or location information) that is added to the voice response.

[0719] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to the user's emotions. This system consists of the following parts:

[0720] (1) Acquisition of voice input

[0721] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone. For example, if the user asks, "Where is the coffee maker?", that voice will be acquired by the terminal.

[0722] (2) Text conversion of voice input

[0723] Terminal: The acquired voice data is sent to a speech recognition engine (e.g., a speech recognition API), which converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0724] (3) Analysis of text data

[0725] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[0726] (4) Recognition of emotions

[0727] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes the user's emotions from the tone of their voice and the content of their text. For example, it can recognize emotions such as when the user is in a hurry or is frustrated.

[0728] (5) Querying the database

[0729] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[0730] (6) Organizing search results and generating responses

[0731] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[0732] (7) Text to speech conversion

[0733] Server: The organized text responses are converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This data is expressed as natural speech corresponding to the user's question.

[0734] (8) Output of audio data to the user

[0735] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0736] Specific example

[0737] Example 1: Product Search

[0738] When a user asks, "Where is the coffee maker?", the following process takes place:

[0739] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[0740] 2. Server: Converts audio data into text data and analyzes it.

[0741] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[0742] 4. Server: Executes queries on the database to obtain location information.

[0743] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[0744] 6. Terminal: Plays audio data and displays supplementary information on the display.

[0745] Example 2: Inquiry about business hours

[0746] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[0747] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[0748] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0749] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[0750] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[0751] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[0752] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[0753] Example prompts for use with a generative AI model

[0754] "Please explain the specific processing flow when a user asks, 'Where is the coffee maker?'"

[0755] "Please explain how the system recognizes the user's emotions and how it generates responses based on those emotions."

[0756] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0757] Step 1:

[0758] Terminal: The terminal is equipped with a microphone, and the user provides voice input. For example, the user might ask, "Where is the coffee maker?" This voice data is acquired in digital format and sent to the next processing step. The input is the user's voice data, and the output is voice data in digital format.

[0759] Step 2:

[0760] Terminal: The acquired digital voice data is sent to a speech recognition engine (e.g., a speech recognition API). This engine analyzes the voice data and converts it into text data. For example, the voice data "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?". The input is voice data, and the output is text data.

[0761] Step 3:

[0762] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker. The input is text data, and the output is the intent of the user's question.

[0763] Step 4:

[0764] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes emotions from the tone of voice and the content of the text. For example, it recognizes emotions such as when the user is in a hurry or is frustrated. The input is text and audio data, and the output is the user's emotional state.

[0765] Step 5:

[0766] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it generates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database. The input is the analysis results and sentiment recognition results, and the output is the location information.

[0767] Step 6:

[0768] Server: Organizes location information obtained from the database and generates text responses in natural language based on sentiment recognition results. For example, it might generate a text response such as, "We will let you know shortly. The coffee maker is on the second floor in the home appliance section." The input is location information and sentiment recognition results, and the output is a text response.

[0769] Step 7:

[0770] Server: The generated text response is converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This conversion creates speech data that is natural-sounding, corresponding to the user's question. The input is a text response, and the output is speech data.

[0771] Step 8:

[0772] Terminal: The created audio data is sent to the terminal and played back to the user through the speaker. Additionally, supplementary information such as location data is displayed on the screen. Input is audio data and supplementary information, while output is playback of the audio response and display on the screen.

[0773] (Application Example 2)

[0774] 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."

[0775] In modern commercial facilities and stores, enabling users to quickly and accurately obtain information about the products and services they need is crucial for improving customer satisfaction. However, many customers may prefer not to ask store staff directly, and staff are not always available to assist. Therefore, there is a need for unmanned systems that efficiently answer customer questions. Furthermore, systems that can respond in accordance with customer emotions are expected to further enhance the customer experience.

[0776] 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. In this invention, the server includes means for analyzing the user's voice data and text data and recognizing the user's emotions, means for executing queries against the database and obtaining search results, and means for organizing the search results and generating text in natural language according to the emotions. This enables appropriate and rapid responses to user questions, as well as friendly responses that are tailored to the user's emotions.

[0777] "Means for obtaining voice input from the user" refers to devices or programs for collecting voices spoken by the user, and includes microphones and voice input interfaces.

[0778] "Means of converting voice input into text data" refers to technologies and programs that convert acquired voice into string information, and this includes speech recognition engines and their algorithms.

[0779] "Methods for analyzing text data using natural language processing to identify user intent" refers to technologies and programs that linguistically analyze string information to understand the intent behind user requests and questions, and includes natural language processing (NLP) engines.

[0780] "Means of analyzing user voice and text data to recognize user emotions" refers to technologies and programs that estimate psychological states and emotions based on the characteristics of voice and text, and emotion recognition engines are included in this.

[0781] "A means of executing queries against a database and retrieving search results" refers to the technology or program that sends commands to a database to obtain specific information and receives the response, and this includes SQL queries.

[0782] "Means for organizing search results and generating text in natural language according to emotions" refers to technologies and programs that organize acquired information into a certain structure and generate text in a way that aligns with the user's emotions, and response generation engines are included in this.

[0783] "Means of converting text into audio data" refers to technologies and programs that convert textual information into audio information, and this includes text-to-speech (TTS) engines.

[0784] "Means of outputting audio data to the user" refers to devices or programs that play back the generated audio information, and this includes speakers and audio playback functions.

[0785] "Means of displaying supplementary information on a display" refers to devices or programs that visually represent content in addition to information provided by voice, and this includes displays and screen display functions.

[0786] This invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. A detailed embodiment of this system is described below.

[0787] System Configuration

[0788] The system primarily consists of terminals that acquire user voice input and servers that perform data analysis and response generation.

[0789] Hardware and software to be used

[0790] Hardware: Smartphones, microphones, speakers, displays, servers

[0791] software:

[0792] Speech recognition: Google Cloud Speech-to-Text API

[0793] Natural Language Processing (NLP): NLTK (Python's natural language processing library)

[0794] Emotion recognition: IBM Watson Natural Language Understanding

[0795] Database: MySQL

[0796] Text-to-Speech (TTS): Amazon Polly

[0797] System Operation Overview

[0798] 1. Acquisition of voice input

[0799] The user speaks into a smartphone app, and the voice is captured by the microphone.

[0800] Example: The user asks, "Where is the coffee maker?"

[0801] 2. Text conversion of voice input

[0802] The acquired audio is converted into text data using the Google Cloud Speech-to-Text API.

[0803] This means that the audio "Where is the coffee maker?" will be converted to text "Where is the coffee maker?".

[0804] 3. Analysis of text data

[0805] The server analyzes the generated text data using NLTK to identify the user's intent.

[0806] The analysis results identify the user's intention: "I want to know the location of the product (coffee maker)."

[0807] 4. Recognition of emotions

[0808] The server uses IBM Watson to recognize emotions from the user's text data and voice tone.

[0809] Emotion recognition can sometimes reveal that the user is in a hurry.

[0810] 5. Querying the database

[0811] The server queries the MySQL database to retrieve location information for the relevant product or service.

[0812] The query result retrieves the location information of the coffee maker (e.g., the home appliance section on the second floor).

[0813] 6. Organizing search results and generating responses

[0814] The server organizes the acquired location information and generates natural responses based on the user's emotions.

[0815] For example, for a user in a hurry, a response such as "I'll be right there. The coffee maker is on the second floor in the home appliance section" will be generated.

[0816] 7. Text to Speech Conversion

[0817] The server converts the generated text response into audio data using Amazon Polly.

[0818] This converts the text "The coffee maker is in the home appliance section on the second floor" into audio data.

[0819] 8. Output of audio data to the user

[0820] Audio data is played through the smartphone's speaker, providing information to the user.

[0821] Furthermore, supplementary information (e.g., maps and location information) is displayed on the smartphone screen.

[0822] Specific example

[0823] As a concrete example, the following prompt statements could be used:

[0824] Example of a prompt:

[0825] The user is asking "Where is the coffee maker?" via voice. Convert the voice to text, analyze it, and obtain the product's location information. Also, recognize the user's emotions and respond appropriately.

[0826] Based on this prompt, the system can generate appropriate answers to the user's questions and provide information quickly and accurately.

[0827] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0828] Step 1:

[0829] The user speaks into their smartphone, and voice input is obtained. The user states a question or request, and the device's (smartphone's) microphone captures the voice. The input is the user's voice, and the output is the captured voice data.

[0830] Step 2:

[0831] The acquired audio data is converted into text data using the Google Cloud Speech-to-Text API. The device sends the audio data to the speech recognition engine in the cloud and receives the text data in return. The input is audio data, and the output is the converted text data.

[0832] Step 3:

[0833] The server analyzes the received text data using NLTK to identify the user's intent. The server sends the text data to a natural language processing engine, which then identifies the user's request (e.g., "I want to know the location of the product") as a result of the analysis. The input is text data, and the output is the analysis result (user's intent).

[0834] Step 4:

[0835] The server analyzes text and audio data using IBM Watson to recognize the user's emotions. The server estimates emotions from the tone of voice and the content of the text. The input is text and audio data, and the output is recognized emotion data.

[0836] Step 5:

[0837] The server executes queries against the MySQL database and retrieves search results. Based on the user's intent and sentiment, the server executes queries to request information from the database. The input is the query information, and the output is the search results retrieved from the database (e.g., product location information).

[0838] Step 6:

[0839] The server organizes search results and generates text responses in natural language based on sentiment. The server then creates appropriate responses for the user based on the information it retrieves. Inputs are search results and sentiment data, while output is the generated text response.

[0840] Step 7:

[0841] The generated text response is converted into speech data using Amazon Polly. The server sends the text response to the speech synthesis engine and receives the speech data. The input is the text response, and the output is the converted speech data.

[0842] Step 8:

[0843] The device plays the received audio data and displays supplementary information on the display. The device plays audio through the speaker and displays related information on the display. The input is audio data and supplementary information, and the output is the audio and display content provided to the user.

[0844] 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.

[0845] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">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.

[0846] 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.

[0847] [Third Embodiment]

[0848] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0849] 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.

[0850] 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).

[0851] 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.

[0852] 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.

[0853] 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).

[0854] 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.

[0855] 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.

[0856] 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.

[0857] 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.

[0858] 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.

[0859] 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".

[0860] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. The system consists of the following parts:

[0861] (1) Acquisition of voice input

[0862] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is sent to the system through the microphone.

[0863] (2) Text conversion of voice input

[0864] Terminal: The acquired voice data is sent to the internal speech recognition engine. This engine converts the voice into text data. For example, the user's voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[0865] (3) Analysis of text data

[0866] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[0867] (4) Querying the database

[0868] Server: Based on the results of parsing the text data, the server executes queries against the database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the location information of coffee makers from the database.

[0869] (5) Organizing search results and generating responses

[0870] Server: Search results retrieved from the database are organized by the server, and a user-friendly text response is generated. For example, based on the retrieved location information, a text response such as "The coffee maker is on the second floor in the home appliance section" is generated.

[0871] (6) Text to speech conversion

[0872] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[0873] (7) Output of audio data to the user

[0874] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[0875] Specific example

[0876] For example, if a user asks a robot in a large commercial facility, "Please tell me the opening hours of this store," the process would be as follows:

[0877] 1. Terminal: The microphone captures the user's voice and sends the voice data to the server.

[0878] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[0879] 3. Server: A natural language processing engine analyzes the text and identifies the user's intention, such as "I want to know the business hours information."

[0880] 4. Server: Execute an appropriate query on the database (e.g., SELECT hours FROM stores WHERE name='this store';) to retrieve the business hours information.

[0881] 5. Server: Based on the search results, it generates a text response that says, "This store is open every day from 9 am to 9 pm."

[0882] 6. Server: Converts text responses into audio data and sends the audio data to the terminal.

[0883] 7. Terminal: Plays audio data and tells the user through the speaker, "This store is open every day from 9 am to 9 pm." The opening hours are also displayed on the screen.

[0884] This system allows users to quickly and accurately obtain the information they need within commercial facilities.

[0885] The following describes the processing flow.

[0886] Step 1:

[0887] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[0888] Step 2:

[0889] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[0890] Step 3:

[0891] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[0892] Step 4:

[0893] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[0894] Step 5:

[0895] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[0896] Step 6:

[0897] The server identifies the user's intent as "I want to know the location of the product," and then generates a query to the database. For example, it creates an SQL query such as "SELECT location FROM products WHERE name='coffee maker';".

[0898] Step 7:

[0899] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[0900] Step 8:

[0901] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. For example, it might generate text such as, "The coffee maker is located in the home appliance section on the second floor."

[0902] Step 9:

[0903] The server sends the generated text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[0904] Step 10:

[0905] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[0906] Step 11:

[0907] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[0908] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility.

[0909] (Example 1)

[0910] 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."

[0911] Traditional information retrieval systems in commercial facilities and stores had problems that made it difficult for users to obtain information quickly and accurately. In particular, voice-based information retrieval was insufficient, and users often had to perform manual searches. Furthermore, the accuracy of natural language processing was low, and it was sometimes impossible to accurately analyze user intent. In addition, the means of effectively delivering search results to users were limited, resulting in a system that was difficult for users to use.

[0912] 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.

[0913] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for querying a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, and means for displaying supplementary information on a display device. This enables the user to quickly and accurately acquire the necessary information within a commercial facility.

[0914] "Voice input" refers to audio data that a user provides to the system via a microphone.

[0915] "Text data" refers to character information converted from voice input by a speech recognition engine.

[0916] "Natural language processing" is a general term for technologies and methods used to analyze text data and identify user intent.

[0917] A "database" refers to a management system for managing and retrieving information, as well as a collection of structured data used to store that information.

[0918] A "query" is a set of instructions sent to a database to retrieve or manipulate information.

[0919] "Search results" refer to the data obtained when a query is executed against a database.

[0920] "Voice data" refers to digital data used to store the voice responses generated by the system.

[0921] A "display device" is a device used to visually display text data and supplementary information.

[0922] "Supplemental information" refers to additional text, images, or other information displayed on a display device to complement the voice response.

[0923] A "user" refers to a person who provides input for using the system to search for information.

[0924] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. This system automates a series of processes from acquiring voice input to outputting voice responses, and is specifically configured as follows.

[0925] Hardware and software configuration

[0926] Acquiring voice input

[0927] Terminal: The terminal is equipped with a high-sensitivity microphone. When the user provides voice input through this microphone, the voice is stored as digital audio data within the terminal. The terminal has the capability to process the audio data in real time.

[0928] Voice input to text conversion

[0929] Terminal: Voice data is converted into text data using a speech recognition engine such as the Google Cloud Speech-to-Text API. For example, the voice saying "Where is the coffee maker?" is converted into text data and sent to the server.

[0930] Text data analysis

[0931] Server: The server is equipped with the BERT model as its natural language processing engine. Text data is sent to the server and parsed by the BERT model. This identifies the intent behind the user's question.

[0932] Database query

[0933] Server: Based on the results of parsing the text data, the server executes SQL queries against the PostgreSQL database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the necessary information from the database.

[0934] Organizing search results and generating responses

[0935] Server: The server organizes the acquired information and generates a text response in a user-friendly format. For example, it might generate a specific response such as, "The coffee maker is located in the home appliance section on the second floor."

[0936] Text to speech conversion

[0937] Server: The organized text responses are converted into speech data using a speech synthesis engine such as Amazon Polly. This speech data is sent to the terminal in a natural speech format.

[0938] Output of audio data to the user

[0939] Terminal: The generated audio data is played back to the user through the terminal's speaker. Supplementary information (e.g., maps and location information) is also displayed on the display device.

[0940] Specific example

[0941] For example, if a user asks a robot in a large commercial facility, "Could you tell me the opening hours of this store?", the following process will occur.

[0942] 1. Terminal: A high-sensitivity microphone captures the user's voice and stores it as digital audio data within the terminal.

[0943] 2. Terminal: The Google Cloud Speech-to-Text API converts the audio into text data, "Please tell me the opening hours of this store," and sends it to the server.

[0944] 3. Server: The BERT model analyzes the text data and identifies the user's intention, such as "I want to know the business hours information."

[0945] 4. Server: Execute the SQL query "SELECT hours FROM stores WHERE name='this store';" against the PostgreSQL database to retrieve the relevant business hours information.

[0946] 5. Server: Based on the acquired business hours information, it generates a text response that says, "This store is open every day from 9:00 to 21:00."

[0947] 6. Server: Use Amazon Polly to convert text responses into audio data and send it to the terminal.

[0948] 7. Terminal: Plays audio data and provides responses to the user through the speaker. The display shows supplementary information about business hours.

[0949] Example of a prompt

[0950] "Design an information retrieval system for a commercial facility. This system should include seven processing steps: voice input to text conversion, natural language processing analysis, database querying, and voice response generation. Please also specify the names of any relevant hardware and software."

[0951] This system will enable users to quickly and accurately obtain the information they need within commercial facilities.

[0952] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0953] Step 1:

[0954] Terminal: The terminal is equipped with a highly sensitive microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is stored on the terminal as digital audio data. Once this audio data is acquired, the system is ready to proceed to the next processing step. For example, if the user says, "Where is the coffee maker?", their voice will be picked up by the microphone.

[0955] Input: User's voice

[0956] Output: Digital audio data

[0957] Step 2:

[0958] Terminal: Digital voice data is converted to text data using the Google Cloud Speech-to-Text API. This API analyzes the voice input and converts it into text information. In this process, the voice data is transformed into the text "Where is the coffee maker?". After being converted to text data, the data is sent to the server.

[0959] Input: Digital audio data

[0960] Output: Text data

[0961] Step 3:

[0962] Server: The server analyzes the received text data using a BERT model. The server uses a natural language processing engine to identify the user's intent from the text data. For example, the intent "I want to know the location of the coffee maker" is identified. Once the analysis is complete, the analysis results are generated.

[0963] Input: Text data

[0964] Output: Analysis results (user intent)

[0965] Step 4:

[0966] Server: Based on the analysis results, the server executes an SQL query against the PostgreSQL database. During the SQL query generation stage, for example, a query like "SELECT location FROM products WHERE name='coffee maker';" is created. This query is sent to the database, and the necessary information is retrieved. The information returned is the location information of the coffee maker.

[0967] Input: Analysis results (user intent)

[0968] Output: Search results from the database (location information)

[0969] Step 5:

[0970] Server: The server organizes the retrieved search results. Based on the organized data, a user-friendly text response is generated. For example, a response such as "The coffee maker is on the second floor in the home appliance section" is created. Once this response is generated, the process moves on to the next step of speech conversion.

[0971] Input: Search results from the database (location information)

[0972] Output: Text response

[0973] Step 6:

[0974] Server: The server uses a speech synthesis engine such as Amazon Polly to convert the generated text response into speech data. This conversion produces the speech data "The coffee maker is on the second floor in the home appliance section." After this speech data is generated, it is sent to the terminal.

[0975] Input: Text response

[0976] Output: Audio data

[0977] Step 7:

[0978] Terminal: The terminal plays audio data transmitted from the server through its speaker, providing information to the user. Simultaneously, it uses a display device to show a map indicating the coffee maker's location and any necessary supplementary information. The user can hear the audio message, "The coffee maker is in the home appliance section on the second floor," and confirm its specific location on the display.

[0979] Input: Audio data

[0980] Output: Audio output to the user and display of supplementary information.

[0981] (Application Example 1)

[0982] 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."

[0983] In current commercial facilities and stores, customers spend a significant amount of time and effort searching for information such as the location of products and services, opening hours, and promotional campaigns. This compromises customer convenience and prevents an efficient shopping experience within stores. Furthermore, there are limitations to what staff can do to handle searches, and further efficiency improvements are needed, especially in large commercial facilities.

[0984] 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.

[0985] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for executing queries against a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, means for displaying supplementary information on a display, and means for responding to customer questions in a physical store and providing necessary information both audibly and visually. This enables customers to efficiently acquire the necessary information and improve their in-store shopping experience.

[0986] "Means for acquiring voice input" refers to devices or functions for acquiring voice from a user.

[0987] "Means for converting voice input into text data" refers to systems and algorithms for converting acquired voice data into text data.

[0988] "A means of analyzing text data using natural language processing to identify user intent" refers to a technology that analyzes converted text data to identify the information or requests that the user is seeking.

[0989] "A means of executing queries against a database and retrieving search results" refers to a system that queries a database based on a user's request and retrieves the necessary information.

[0990] "Methods for organizing search results and generating text in natural language" refers to technologies for formatting acquired information into an easily understandable form and generating responses in natural language.

[0991] "Means for converting text to audio data" refers to a function for converting generated text data into speech.

[0992] "Means of outputting audio data to the user" refers to devices or functions that provide the converted audio data to the user.

[0993] "Means for displaying supplementary information on a display" refers to technologies for displaying text and image information on a display as an auxiliary means, in addition to voice response.

[0994] "Means of responding to customer questions in physical stores and providing necessary information via voice and visual means" refers to systems or devices that respond to user questions within a physical store and provide necessary information using voice and visual information.

[0995] The embodiments for carrying out this invention will be described in detail below.

[0996] System Overview

[0997] Hardware configuration

[0998] 1. Microphone: Used to obtain user voice input.

[0999] 2. Speaker: Used to provide audio output to the user.

[1000] 3. Display: Used to display auxiliary information.

[1001] 4. Network connection: Used to communicate with the server.

[1002] Software Configuration

[1003] 1. Speech recognition engine: As an example, we will use the Google Cloud Speech-to-Text API.

[1004] 2. Natural Language Processing Engine: OpenAI GPT-4 will be used as an example.

[1005] 3. Database management system: PostgreSQL will be used as an example.

[1006] 4. Text-to-Speech Engine: As an example, we will use the Google Cloud Text-to-Speech API.

[1007] System operation

[1008] 1. Acquisition of voice input

[1009] When a user speaks to the robot, the microphone captures the audio and sends the audio data to the server.

[1010] 2. Text conversion of voice input

[1011] The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API).

[1012] 3. Analysis of text data

[1013] The text data is analyzed by a natural language processing engine (e.g., OpenAI GPT-4) on the server to identify the user's intent.

[1014] 4. Querying the database

[1015] Based on the results of the text data analysis, the server executes queries against the database (e.g., PostgreSQL) to retrieve the necessary information.

[1016] 5. Organizing search results and generating responses

[1017] The server organizes the acquired information and generates a text response in natural language.

[1018] 6. Text to Speech Conversion

[1019] The server converts the text response into speech data using a speech synthesis engine (e.g., Google Cloud Text-to-Speech API).

[1020] 7. Output of audio data to the user

[1021] The converted audio data is played back to the user through the speaker. Supplementary information is also displayed on the screen.

[1022] Specific example

[1023] For example, if a user asks a supermarket robot, "Where can I find the sale milk?", the robot would process it as follows:

[1024] Example of a prompt

[1025] User: "Where can I find milk on sale?"

[1026] System: "Recognizes speech data and converts it to text."

[1027] GPT-4: "Users want to know where the discounted milk is located."

[1028] Server: "SELECT location FROM products WHERE name='Special Sale Milk';"

[1029] Database: "It's located at the far left of the refrigerated section."

[1030] Natural language generation: "The discounted milk is on the far left of the refrigerated section."

[1031] Google Cloud Text-to-Speech API: "The discounted milk is on the far left of the refrigerated section."

[1032] Speaker: (Voice) "The sale milk is located at the far left of the refrigerated section."

[1033] Display: (Shows the location of the refrigerated section on a map)

[1034] In this way, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[1035] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1036] Step 1:

[1037] The user speaks into the device. The device's microphone picks up the user's voice input. For example, if the user says, "Where can I find the sale milk?", that voice data is acquired. Input: User's voice. Output: Voice data.

[1038] Step 2:

[1039] The device sends the acquired audio data to the server. The server uses a speech recognition engine (Google Cloud Speech-to-Text API) to convert the audio data into text data. Specifically, the server processes the audio data and generates the text "Where can I find the discounted milk?". Input: Audio data. Output: Text data.

[1040] Step 3:

[1041] The server passes the acquired text data to a natural language processing engine (OpenAI GPT-4) for analysis. Here, the user's intent is identified as "I want to know the location of the discounted milk." Input: Text data. Output: User's intent.

[1042] Step 4:

[1043] The server generates and executes a query against the database (PostgreSQL) based on the identified user's intent. In this case, the query executed is "SELECT location FROM products WHERE name='Special Offer Milk';". Input: User intent. Output: Query result.

[1044] Step 5:

[1045] The server organizes the search results retrieved from the database and generates a text response in natural language. Specifically, it generates a text response such as, "The discounted milk is on the far left of the refrigerated section." Input: Query results. Output: Text response.

[1046] Step 6:

[1047] The server converts the generated text response into speech data using a text-to-speech engine (Google Cloud Text-to-Speech API). Specifically, the server generates the speech data "The discounted milk is on the far left of the refrigerated section." Input: Text response. Output: Speech data.

[1048] Step 7:

[1049] The terminal plays audio data received from the server through its speaker and provides a response to the user. It also displays the location of the refrigerated section on a map as supplementary information. Input: Audio data and supplementary information. Output: Audio response and visual information.

[1050] Through these steps, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[1051] 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.

[1052] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. The system consists of the following parts:

[1053] (1) Acquisition of voice input

[1054] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone.

[1055] (2) Text conversion of voice input

[1056] Terminal: The acquired voice data is sent to the speech recognition engine. This engine converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[1057] (3) Analysis of text data

[1058] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[1059] (4) Recognition of emotions

[1060] Server: Text and audio data are sent to the emotion engine. The emotion engine analyzes the user's emotions from their voice tone and text content. For example, it recognizes emotions such as when the user is in a hurry or is frustrated.

[1061] (5) Querying the database

[1062] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[1063] (6) Organizing search results and generating responses

[1064] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[1065] (7) Text to speech conversion

[1066] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[1067] (8) Output of audio data to the user

[1068] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[1069] Specific example

[1070] Example 1: Product Search

[1071] When a user asks, "Where is the coffee maker?", the following process takes place:

[1072] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[1073] 2. Server: Converts audio data into text data and analyzes it.

[1074] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[1075] 4. Server: Executes queries on the database to obtain location information.

[1076] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[1077] 6. Terminal: Plays audio data and displays supplementary information on the display.

[1078] Example 2: Inquiry about business hours

[1079] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[1080] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[1081] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[1082] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[1083] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[1084] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[1085] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[1086] This system will not only allow users to obtain quick and accurate information within commercial facilities, but also provide them with friendly, emotionally resonant responses.

[1087] The following describes the processing flow.

[1088] Step 1:

[1089] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[1090] Step 2:

[1091] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[1092] Step 3:

[1093] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[1094] Step 4:

[1095] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[1096] Step 5:

[1097] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[1098] Step 6:

[1099] The server sends text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies the user's emotions from the tone of voice and the content of the text.

[1100] Step 7:

[1101] The server generates queries to the database based on the user's intent and feelings. For example, it might create an SQL query like "SELECT location FROM products WHERE name='coffee maker';".

[1102] Step 8:

[1103] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[1104] Step 9:

[1105] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. It also adjusts the tone and content of the response, taking into account the sentiment recognition results.

[1106] Step 10:

[1107] The server sends the text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[1108] Step 11:

[1109] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[1110] Step 12:

[1111] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[1112] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility. Furthermore, the emotion engine provides friendly responses that are tailored to the user's emotions.

[1113] (Example 2)

[1114] 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."

[1115] Conventional unmanned search tools provide location and business hours information based on user voice input, but they lack the ability to generate responses that take user emotions into account, resulting in a limited user experience. Improving the speed and accuracy of information delivery is also a challenge.

[1116] 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.

[1117] In this invention, the server includes means for recognizing the user's emotions from voice input and text data, means for organizing search results and generating text in natural language based on the emotion recognition results, and means for converting text into voice data. This enables detailed responses tailored to the user's situation and emotions, improving the user experience and providing information quickly and accurately.

[1118] A "user" refers to the end user who attempts to obtain information using this system.

[1119] "Voice input" refers to the process of acquiring the voice a user speaks into a device as digital data.

[1120] "Text data" refers to data in string format converted by speech recognition.

[1121] "Natural language processing" refers to the technology of analyzing text data to identify its meaning and intent.

[1122] "Emotion recognition" refers to the technology that analyzes and identifies a user's emotional state from voice data and text data.

[1123] A "database" refers to a computerized information management system in which various types of information are systematically organized and stored.

[1124] A "query" refers to a set of commands executed on a database for searching or performing operations.

[1125] "Search results" refer to information obtained by executing queries against a database.

[1126] "Speech synthesis" refers to the technology that converts text data into speech data.

[1127] "Audio data" refers to audio information generated by speech synthesis and output to the user.

[1128] "Supplementary information" refers to additional visually displayed information (e.g., maps or location information) that is added to the voice response.

[1129] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to the user's emotions. This system consists of the following parts:

[1130] (1) Acquisition of voice input

[1131] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone. For example, if the user asks, "Where is the coffee maker?", that voice will be acquired by the terminal.

[1132] (2) Text conversion of voice input

[1133] Terminal: The acquired voice data is sent to a speech recognition engine (e.g., a speech recognition API), which converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[1134] (3) Analysis of text data

[1135] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[1136] (4) Recognition of emotions

[1137] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes the user's emotions from the tone of their voice and the content of their text. For example, it can recognize emotions such as when the user is in a hurry or is frustrated.

[1138] (5) Querying the database

[1139] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[1140] (6) Organizing search results and generating responses

[1141] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[1142] (7) Text to speech conversion

[1143] Server: The organized text responses are converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This data is expressed as natural speech corresponding to the user's question.

[1144] (8) Output of audio data to the user

[1145] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[1146] Specific example

[1147] Example 1: Product Search

[1148] When a user asks, "Where is the coffee maker?", the following process takes place:

[1149] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[1150] 2. Server: Converts audio data into text data and analyzes it.

[1151] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[1152] 4. Server: Executes queries on the database to obtain location information.

[1153] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[1154] 6. Terminal: Plays audio data and displays supplementary information on the display.

[1155] Example 2: Inquiry about business hours

[1156] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[1157] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[1158] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[1159] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[1160] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[1161] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[1162] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[1163] Example prompts for use with a generative AI model

[1164] "Please explain the specific processing flow when a user asks, 'Where is the coffee maker?'"

[1165] "Please explain how the system recognizes the user's emotions and how it generates responses based on those emotions."

[1166] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1167] Step 1:

[1168] Terminal: The terminal is equipped with a microphone, and the user provides voice input. For example, the user might ask, "Where is the coffee maker?" This voice data is acquired in digital format and sent to the next processing step. The input is the user's voice data, and the output is voice data in digital format.

[1169] Step 2:

[1170] Terminal: The acquired digital voice data is sent to a speech recognition engine (e.g., a speech recognition API). This engine analyzes the voice data and converts it into text data. For example, the voice data "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?". The input is voice data, and the output is text data.

[1171] Step 3:

[1172] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker. The input is text data, and the output is the intent of the user's question.

[1173] Step 4:

[1174] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes emotions from the tone of voice and the content of the text. For example, it recognizes emotions such as when the user is in a hurry or is frustrated. The input is text and audio data, and the output is the user's emotional state.

[1175] Step 5:

[1176] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it generates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database. The input is the analysis results and sentiment recognition results, and the output is the location information.

[1177] Step 6:

[1178] Server: Organizes location information obtained from the database and generates text responses in natural language based on sentiment recognition results. For example, it might generate a text response such as, "We will let you know shortly. The coffee maker is on the second floor in the home appliance section." The input is location information and sentiment recognition results, and the output is a text response.

[1179] Step 7:

[1180] Server: The generated text response is converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This conversion creates speech data that is natural-sounding, corresponding to the user's question. The input is a text response, and the output is speech data.

[1181] Step 8:

[1182] Terminal: The created audio data is sent to the terminal and played back to the user through the speaker. Additionally, supplementary information such as location data is displayed on the screen. Input is audio data and supplementary information, while output is playback of the audio response and display on the screen.

[1183] (Application Example 2)

[1184] 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."

[1185] In modern commercial facilities and stores, enabling users to quickly and accurately obtain information about the products and services they need is crucial for improving customer satisfaction. However, many customers may prefer not to ask store staff directly, and staff are not always available to assist. Therefore, there is a need for unmanned systems that efficiently answer customer questions. Furthermore, systems that can respond in accordance with customer emotions are expected to further enhance the customer experience.

[1186] 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. In this invention, the server includes means for analyzing the user's voice data and text data and recognizing the user's emotions, means for executing queries against the database and obtaining search results, and means for organizing the search results and generating text in natural language according to the emotions. This enables appropriate and rapid responses to user questions, as well as friendly responses that are tailored to the user's emotions.

[1187] "Means for obtaining voice input from the user" refers to devices or programs for collecting voices spoken by the user, and includes microphones and voice input interfaces.

[1188] "Means of converting voice input into text data" refers to technologies and programs that convert acquired voice into string information, and this includes speech recognition engines and their algorithms.

[1189] "Methods for analyzing text data using natural language processing to identify user intent" refers to technologies and programs that linguistically analyze string information to understand the intent behind user requests and questions, and includes natural language processing (NLP) engines.

[1190] "Means of analyzing user voice and text data to recognize user emotions" refers to technologies and programs that estimate psychological states and emotions based on the characteristics of voice and text, and emotion recognition engines are included in this.

[1191] "A means of executing queries against a database and retrieving search results" refers to the technology or program that sends commands to a database to obtain specific information and receives the response, and this includes SQL queries.

[1192] "Means for organizing search results and generating text in natural language according to emotions" refers to technologies and programs that organize acquired information into a certain structure and generate text in a way that aligns with the user's emotions, and response generation engines are included in this.

[1193] "Means of converting text into audio data" refers to technologies and programs that convert textual information into audio information, and this includes text-to-speech (TTS) engines.

[1194] "Means of outputting audio data to the user" refers to devices or programs that play back the generated audio information, and this includes speakers and audio playback functions.

[1195] "Means of displaying supplementary information on a display" refers to devices or programs that visually represent content in addition to information provided by voice, and this includes displays and screen display functions.

[1196] This invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. A detailed embodiment of this system is described below.

[1197] System Configuration

[1198] The system primarily consists of terminals that acquire user voice input and servers that perform data analysis and response generation.

[1199] Hardware and software to be used

[1200] Hardware: Smartphones, microphones, speakers, displays, servers

[1201] software:

[1202] Speech recognition: Google Cloud Speech-to-Text API

[1203] Natural Language Processing (NLP): NLTK (Python's natural language processing library)

[1204] Emotion recognition: IBM Watson Natural Language Understanding

[1205] Database: MySQL

[1206] Text-to-Speech (TTS): Amazon Polly

[1207] System Operation Overview

[1208] 1. Acquisition of voice input

[1209] The user speaks into a smartphone app, and the voice is captured by the microphone.

[1210] Example: The user asks, "Where is the coffee maker?"

[1211] 2. Text conversion of voice input

[1212] The acquired audio is converted into text data using the Google Cloud Speech-to-Text API.

[1213] This means that the audio "Where is the coffee maker?" will be converted to text "Where is the coffee maker?".

[1214] 3. Analysis of text data

[1215] The server analyzes the generated text data using NLTK to identify the user's intent.

[1216] The analysis results identify the user's intention: "I want to know the location of the product (coffee maker)."

[1217] 4. Recognition of emotions

[1218] The server uses IBM Watson to recognize emotions from the user's text data and voice tone.

[1219] Emotion recognition can sometimes reveal that the user is in a hurry.

[1220] 5. Querying the database

[1221] The server queries the MySQL database to retrieve location information for the relevant product or service.

[1222] The query result retrieves the location information of the coffee maker (e.g., the home appliance section on the second floor).

[1223] 6. Organizing search results and generating responses

[1224] The server organizes the acquired location information and generates natural responses based on the user's emotions.

[1225] For example, for a user in a hurry, a response such as "I'll be right there. The coffee maker is on the second floor in the home appliance section" will be generated.

[1226] 7. Text to Speech Conversion

[1227] The server converts the generated text response into audio data using Amazon Polly.

[1228] This converts the text "The coffee maker is in the home appliance section on the second floor" into audio data.

[1229] 8. Output of audio data to the user

[1230] Audio data is played through the smartphone's speaker, providing information to the user.

[1231] Furthermore, supplementary information (e.g., maps and location information) is displayed on the smartphone screen.

[1232] Specific example

[1233] As a concrete example, the following prompt statements could be used:

[1234] Example of a prompt:

[1235] The user is asking "Where is the coffee maker?" via voice. Convert the voice to text, analyze it, and obtain the product's location information. Also, recognize the user's emotions and respond appropriately.

[1236] Based on this prompt, the system can generate appropriate answers to the user's questions and provide information quickly and accurately.

[1237] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1238] Step 1:

[1239] The user speaks into their smartphone, and voice input is obtained. The user states a question or request, and the device's (smartphone's) microphone captures the voice. The input is the user's voice, and the output is the captured voice data.

[1240] Step 2:

[1241] The acquired audio data is converted into text data using the Google Cloud Speech-to-Text API. The device sends the audio data to the speech recognition engine in the cloud and receives the text data in return. The input is audio data, and the output is the converted text data.

[1242] Step 3:

[1243] The server analyzes the received text data using NLTK to identify the user's intent. The server sends the text data to a natural language processing engine, which then identifies the user's request (e.g., "I want to know the location of the product") as a result of the analysis. The input is text data, and the output is the analysis result (user's intent).

[1244] Step 4:

[1245] The server analyzes text and audio data using IBM Watson to recognize the user's emotions. The server estimates emotions from the tone of voice and the content of the text. The input is text and audio data, and the output is recognized emotion data.

[1246] Step 5:

[1247] The server executes queries against the MySQL database and retrieves search results. Based on the user's intent and sentiment, the server executes queries to request information from the database. The input is the query information, and the output is the search results retrieved from the database (e.g., product location information).

[1248] Step 6:

[1249] The server organizes search results and generates text responses in natural language based on sentiment. The server then creates appropriate responses for the user based on the information it retrieves. Inputs are search results and sentiment data, while output is the generated text response.

[1250] Step 7:

[1251] The generated text response is converted into speech data using Amazon Polly. The server sends the text response to the speech synthesis engine and receives the speech data. The input is the text response, and the output is the converted speech data.

[1252] Step 8:

[1253] The device plays the received audio data and displays supplementary information on the display. The device plays audio through the speaker and displays related information on the display. The input is audio data and supplementary information, and the output is the audio and display content provided to the user.

[1254] 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.

[1255] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">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.

[1256] 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.

[1257] [Fourth Embodiment]

[1258] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1259] 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.

[1260] 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).

[1261] 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.

[1262] 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.

[1263] 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).

[1264] 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.

[1265] 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.

[1266] 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.

[1267] 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.

[1268] 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.

[1269] 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.

[1270] 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".

[1271] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. The system consists of the following parts:

[1272] (1) Acquisition of voice input

[1273] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is sent to the system through the microphone.

[1274] (2) Text conversion of voice input

[1275] Terminal: The acquired voice data is sent to the internal speech recognition engine. This engine converts the voice into text data. For example, the user's voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[1276] (3) Analysis of text data

[1277] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[1278] (4) Querying the database

[1279] Server: Based on the results of parsing the text data, the server executes queries against the database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the location information of coffee makers from the database.

[1280] (5) Organizing search results and generating responses

[1281] Server: Search results retrieved from the database are organized by the server, and a user-friendly text response is generated. For example, based on the retrieved location information, a text response such as "The coffee maker is on the second floor in the home appliance section" is generated.

[1282] (6) Text to speech conversion

[1283] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[1284] (7) Output of audio data to the user

[1285] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[1286] Specific example

[1287] For example, if a user asks a robot in a large commercial facility, "Please tell me the opening hours of this store," the process would be as follows:

[1288] 1. Terminal: The microphone captures the user's voice and sends the voice data to the server.

[1289] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[1290] 3. Server: A natural language processing engine analyzes the text and identifies the user's intention, such as "I want to know the business hours information."

[1291] 4. Server: Execute an appropriate query on the database (e.g., SELECT hours FROM stores WHERE name='this store';) to retrieve the business hours information.

[1292] 5. Server: Based on the search results, it generates a text response that says, "This store is open every day from 9 am to 9 pm."

[1293] 6. Server: Converts text responses into audio data and sends the audio data to the terminal.

[1294] 7. Terminal: Plays audio data and tells the user through the speaker, "This store is open every day from 9 am to 9 pm." The opening hours are also displayed on the screen.

[1295] This system allows users to quickly and accurately obtain the information they need within commercial facilities.

[1296] The following describes the processing flow.

[1297] Step 1:

[1298] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[1299] Step 2:

[1300] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[1301] Step 3:

[1302] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[1303] Step 4:

[1304] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[1305] Step 5:

[1306] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[1307] Step 6:

[1308] The server identifies the user's intent as "I want to know the location of the product," and then generates a query to the database. For example, it creates an SQL query such as "SELECT location FROM products WHERE name='coffee maker';".

[1309] Step 7:

[1310] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[1311] Step 8:

[1312] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. For example, it might generate text such as, "The coffee maker is located in the home appliance section on the second floor."

[1313] Step 9:

[1314] The server sends the generated text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[1315] Step 10:

[1316] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[1317] Step 11:

[1318] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[1319] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility.

[1320] (Example 1)

[1321] 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".

[1322] Traditional information retrieval systems in commercial facilities and stores had problems that made it difficult for users to obtain information quickly and accurately. In particular, voice-based information retrieval was insufficient, and users often had to perform manual searches. Furthermore, the accuracy of natural language processing was low, and it was sometimes impossible to accurately analyze user intent. In addition, the means of effectively delivering search results to users were limited, resulting in a system that was difficult for users to use.

[1323] 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.

[1324] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for querying a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, and means for displaying supplementary information on a display device. This enables the user to quickly and accurately acquire the necessary information within a commercial facility.

[1325] "Voice input" refers to audio data that a user provides to the system via a microphone.

[1326] "Text data" refers to character information converted from voice input by a speech recognition engine.

[1327] "Natural language processing" is a general term for technologies and methods used to analyze text data and identify user intent.

[1328] A "database" refers to a management system for managing and retrieving information, as well as a collection of structured data used to store that information.

[1329] A "query" is a set of instructions sent to a database to retrieve or manipulate information.

[1330] "Search results" refer to the data obtained when a query is executed against a database.

[1331] "Voice data" refers to digital data used to store the voice responses generated by the system.

[1332] A "display device" is a device used to visually display text data and supplementary information.

[1333] "Supplemental information" refers to additional text, images, or other information displayed on a display device to complement the voice response.

[1334] A "user" refers to a person who provides input for using the system to search for information.

[1335] The present invention provides an unmanned search tool that is convenient for users to use when searching for information in commercial facilities and stores. This system automates a series of processes from acquiring voice input to outputting voice responses, and is specifically configured as follows.

[1336] Hardware and software configuration

[1337] Acquiring voice input

[1338] Terminal: The terminal is equipped with a high-sensitivity microphone. When the user provides voice input through this microphone, the voice is stored as digital audio data within the terminal. The terminal has the capability to process the audio data in real time.

[1339] Voice input to text conversion

[1340] Terminal: Voice data is converted into text data using a speech recognition engine such as the Google Cloud Speech-to-Text API. For example, the voice saying "Where is the coffee maker?" is converted into text data and sent to the server.

[1341] Text data analysis

[1342] Server: The server is equipped with the BERT model as its natural language processing engine. Text data is sent to the server and parsed by the BERT model. This identifies the intent behind the user's question.

[1343] Database query

[1344] Server: Based on the results of parsing the text data, the server executes SQL queries against the PostgreSQL database. For example, a query like "SELECT location FROM products WHERE name='coffee maker';" is generated to retrieve the necessary information from the database.

[1345] Organizing search results and generating responses

[1346] Server: The server organizes the acquired information and generates a text response in a user-friendly format. For example, it might generate a specific response such as, "The coffee maker is located in the home appliance section on the second floor."

[1347] Text to speech conversion

[1348] Server: The organized text responses are converted into speech data using a speech synthesis engine such as Amazon Polly. This speech data is sent to the terminal in a natural speech format.

[1349] Output of audio data to the user

[1350] Terminal: The generated audio data is played back to the user through the terminal's speaker. Supplementary information (e.g., maps and location information) is also displayed on the display device.

[1351] Specific example

[1352] For example, if a user asks a robot in a large commercial facility, "Could you tell me the opening hours of this store?", the following process will occur.

[1353] 1. Terminal: A high-sensitivity microphone captures the user's voice and stores it as digital audio data within the terminal.

[1354] 2. Terminal: The Google Cloud Speech-to-Text API converts the audio into text data, "Please tell me the opening hours of this store," and sends it to the server.

[1355] 3. Server: The BERT model analyzes the text data and identifies the user's intention, such as "I want to know the business hours information."

[1356] 4. Server: Execute the SQL query "SELECT hours FROM stores WHERE name='this store';" against the PostgreSQL database to retrieve the relevant business hours information.

[1357] 5. Server: Based on the acquired business hours information, it generates a text response that says, "This store is open every day from 9:00 to 21:00."

[1358] 6. Server: Use Amazon Polly to convert text responses into audio data and send it to the terminal.

[1359] 7. Terminal: Plays audio data and provides responses to the user through the speaker. The display shows supplementary information about business hours.

[1360] Example of a prompt

[1361] "Design an information retrieval system for a commercial facility. This system should include seven processing steps: voice input to text conversion, natural language processing analysis, database querying, and voice response generation. Please also specify the names of any relevant hardware and software."

[1362] This system will enable users to quickly and accurately obtain the information they need within commercial facilities.

[1363] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1364] Step 1:

[1365] Terminal: The terminal is equipped with a highly sensitive microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is stored on the terminal as digital audio data. Once this audio data is acquired, the system is ready to proceed to the next processing step. For example, if the user says, "Where is the coffee maker?", their voice will be picked up by the microphone.

[1366] Input: User's voice

[1367] Output: Digital audio data

[1368] Step 2:

[1369] Terminal: Digital voice data is converted to text data using the Google Cloud Speech-to-Text API. This API analyzes the voice input and converts it into text information. In this process, the voice data is transformed into the text "Where is the coffee maker?". After being converted to text data, the data is sent to the server.

[1370] Input: Digital audio data

[1371] Output: Text data

[1372] Step 3:

[1373] Server: The server analyzes the received text data using a BERT model. The server uses a natural language processing engine to identify the user's intent from the text data. For example, the intent "I want to know the location of the coffee maker" is identified. Once the analysis is complete, the analysis results are generated.

[1374] Input: Text data

[1375] Output: Analysis results (user intent)

[1376] Step 4:

[1377] Server: Based on the analysis results, the server executes an SQL query against the PostgreSQL database. During the SQL query generation stage, for example, a query like "SELECT location FROM products WHERE name='coffee maker';" is created. This query is sent to the database, and the necessary information is retrieved. The information returned is the location information of the coffee maker.

[1378] Input: Analysis results (user intent)

[1379] Output: Search results from the database (location information)

[1380] Step 5:

[1381] Server: The server organizes the retrieved search results. Based on the organized data, a user-friendly text response is generated. For example, a response such as "The coffee maker is on the second floor in the home appliance section" is created. Once this response is generated, the process moves on to the next step of speech conversion.

[1382] Input: Search results from the database (location information)

[1383] Output: Text response

[1384] Step 6:

[1385] Server: The server uses a speech synthesis engine such as Amazon Polly to convert the generated text response into speech data. This conversion produces the speech data "The coffee maker is on the second floor in the home appliance section." After this speech data is generated, it is sent to the terminal.

[1386] Input: Text response

[1387] Output: Audio data

[1388] Step 7:

[1389] Terminal: The terminal plays audio data transmitted from the server through its speaker, providing information to the user. Simultaneously, it uses a display device to show a map indicating the coffee maker's location and any necessary supplementary information. The user can hear the audio message, "The coffee maker is in the home appliance section on the second floor," and confirm its specific location on the display.

[1390] Input: Audio data

[1391] Output: Audio output to the user and display of supplementary information.

[1392] (Application Example 1)

[1393] 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".

[1394] In current commercial facilities and stores, customers spend a significant amount of time and effort searching for information such as the location of products and services, opening hours, and promotional campaigns. This compromises customer convenience and prevents an efficient shopping experience within stores. Furthermore, there are limitations to what staff can do to handle searches, and further efficiency improvements are needed, especially in large commercial facilities.

[1395] 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.

[1396] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data using natural language processing to identify the user's intent, means for executing queries against a database and acquiring search results, means for organizing the search results and generating text in natural language, means for converting the text into voice data, means for outputting the voice data to the user, means for displaying supplementary information on a display, and means for responding to customer questions in a physical store and providing necessary information both audibly and visually. This enables customers to efficiently acquire the necessary information and improve their in-store shopping experience.

[1397] "Means for acquiring voice input" refers to devices or functions for acquiring voice from a user.

[1398] "Means for converting voice input into text data" refers to systems and algorithms for converting acquired voice data into text data.

[1399] "A means of analyzing text data using natural language processing to identify user intent" refers to a technology that analyzes converted text data to identify the information or requests that the user is seeking.

[1400] "A means of executing queries against a database and retrieving search results" refers to a system that queries a database based on a user's request and retrieves the necessary information.

[1401] "Methods for organizing search results and generating text in natural language" refers to technologies for formatting acquired information into an easily understandable form and generating responses in natural language.

[1402] "Means for converting text to audio data" refers to a function for converting generated text data into speech.

[1403] "Means of outputting audio data to the user" refers to devices or functions that provide the converted audio data to the user.

[1404] "Means for displaying supplementary information on a display" refers to technologies for displaying text and image information on a display as an auxiliary means, in addition to voice response.

[1405] "Means of responding to customer questions in physical stores and providing necessary information via voice and visual means" refers to systems or devices that respond to user questions within a physical store and provide necessary information using voice and visual information.

[1406] The embodiments for carrying out this invention will be described in detail below.

[1407] System Overview

[1408] Hardware configuration

[1409] 1. Microphone: Used to obtain user voice input.

[1410] 2. Speaker: Used to provide audio output to the user.

[1411] 3. Display: Used to display auxiliary information.

[1412] 4. Network connection: Used to communicate with the server.

[1413] Software Configuration

[1414] 1. Speech recognition engine: As an example, we will use the Google Cloud Speech-to-Text API.

[1415] 2. Natural Language Processing Engine: OpenAI GPT-4 will be used as an example.

[1416] 3. Database management system: PostgreSQL will be used as an example.

[1417] 4. Text-to-Speech Engine: As an example, we will use the Google Cloud Text-to-Speech API.

[1418] System operation

[1419] 1. Acquisition of voice input

[1420] When a user speaks to the robot, the microphone captures the audio and sends the audio data to the server.

[1421] 2. Text conversion of voice input

[1422] The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API).

[1423] 3. Analysis of text data

[1424] The text data is analyzed by a natural language processing engine (e.g., OpenAI GPT-4) on the server to identify the user's intent.

[1425] 4. Querying the database

[1426] Based on the results of the text data analysis, the server executes queries against the database (e.g., PostgreSQL) to retrieve the necessary information.

[1427] 5. Organizing search results and generating responses

[1428] The server organizes the acquired information and generates a text response in natural language.

[1429] 6. Text to Speech Conversion

[1430] The server converts the text response into speech data using a speech synthesis engine (e.g., Google Cloud Text-to-Speech API).

[1431] 7. Output of audio data to the user

[1432] The converted audio data is played back to the user through the speaker. Supplementary information is also displayed on the screen.

[1433] Specific example

[1434] For example, if a user asks a supermarket robot, "Where can I find the sale milk?", the robot would process it as follows:

[1435] Example of a prompt

[1436] User: "Where can I find milk on sale?"

[1437] System: "Recognizes speech data and converts it to text."

[1438] GPT-4: "Users want to know where the discounted milk is located."

[1439] Server: "SELECT location FROM products WHERE name='Special Sale Milk';"

[1440] Database: "It's located at the far left of the refrigerated section."

[1441] Natural language generation: "The discounted milk is on the far left of the refrigerated section."

[1442] Google Cloud Text-to-Speech API: "The discounted milk is on the far left of the refrigerated section."

[1443] Speaker: (Voice) "The sale milk is located at the far left of the refrigerated section."

[1444] Display: (Shows the location of the refrigerated section on a map)

[1445] In this way, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[1446] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1447] Step 1:

[1448] The user speaks into the device. The device's microphone picks up the user's voice input. For example, if the user says, "Where can I find the sale milk?", that voice data is acquired. Input: User's voice. Output: Voice data.

[1449] Step 2:

[1450] The device sends the acquired audio data to the server. The server uses a speech recognition engine (Google Cloud Speech-to-Text API) to convert the audio data into text data. Specifically, the server processes the audio data and generates the text "Where can I find the discounted milk?". Input: Audio data. Output: Text data.

[1451] Step 3:

[1452] The server passes the acquired text data to a natural language processing engine (OpenAI GPT-4) for analysis. Here, the user's intent is identified as "I want to know the location of the discounted milk." Input: Text data. Output: User's intent.

[1453] Step 4:

[1454] The server generates and executes a query against the database (PostgreSQL) based on the identified user's intent. In this case, the query executed is "SELECT location FROM products WHERE name='Special Offer Milk';". Input: User intent. Output: Query result.

[1455] Step 5:

[1456] The server organizes the search results retrieved from the database and generates a text response in natural language. Specifically, it generates a text response such as, "The discounted milk is on the far left of the refrigerated section." Input: Query results. Output: Text response.

[1457] Step 6:

[1458] The server converts the generated text response into speech data using a text-to-speech engine (Google Cloud Text-to-Speech API). Specifically, the server generates the speech data "The discounted milk is on the far left of the refrigerated section." Input: Text response. Output: Speech data.

[1459] Step 7:

[1460] The terminal plays audio data received from the server through its speaker and provides a response to the user. It also displays the location of the refrigerated section on a map as supplementary information. Input: Audio data and supplementary information. Output: Audio response and visual information.

[1461] Through these steps, users can quickly and accurately obtain the information they need. Implementing this system improves the in-store customer experience and enables more efficient shopping.

[1462] 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.

[1463] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. The system consists of the following parts:

[1464] (1) Acquisition of voice input

[1465] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone.

[1466] (2) Text conversion of voice input

[1467] Terminal: The acquired voice data is sent to the speech recognition engine. This engine converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[1468] (3) Analysis of text data

[1469] Server: Text data is sent to the server and parsed by a natural language processing engine. The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[1470] (4) Recognition of emotions

[1471] Server: Text and audio data are sent to the emotion engine. The emotion engine analyzes the user's emotions from their voice tone and text content. For example, it recognizes emotions such as when the user is in a hurry or is frustrated.

[1472] (5) Querying the database

[1473] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[1474] (6) Organizing search results and generating responses

[1475] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[1476] (7) Text to speech conversion

[1477] Server: The organized text responses are converted into speech data using a speech synthesis engine. This data is expressed as natural speech corresponding to the user's questions.

[1478] (8) Output of audio data to the user

[1479] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[1480] Specific example

[1481] Example 1: Product Search

[1482] When a user asks, "Where is the coffee maker?", the following process takes place:

[1483] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[1484] 2. Server: Converts audio data into text data and analyzes it.

[1485] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[1486] 4. Server: Executes queries on the database to obtain location information.

[1487] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[1488] 6. Terminal: Plays audio data and displays supplementary information on the display.

[1489] Example 2: Inquiry about business hours

[1490] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[1491] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[1492] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[1493] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[1494] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[1495] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[1496] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[1497] This system will not only allow users to obtain quick and accurate information within commercial facilities, but also provide them with friendly, emotionally resonant responses.

[1498] The following describes the processing flow.

[1499] Step 1:

[1500] The user asks a question to the robot. For example, they might use voice input such as, "Where is the coffee maker?"

[1501] Step 2:

[1502] The device uses a microphone to capture the user's voice. The captured voice data is temporarily stored within the device.

[1503] Step 3:

[1504] The device transmits the acquired voice data to the speech recognition engine. At this time, the voice data may be sent via the internet to a speech recognition service in the cloud.

[1505] Step 4:

[1506] The server receives voice data from the speech recognition engine and converts it into text data. Specifically, it generates the text "Where is the coffee maker?".

[1507] Step 5:

[1508] The server sends the generated text data to a natural language processing engine. The natural language processing engine analyzes the context of the text data and identifies the user's intent.

[1509] Step 6:

[1510] The server sends text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies the user's emotions from the tone of voice and the content of the text.

[1511] Step 7:

[1512] The server generates queries to the database based on the user's intent and feelings. For example, it might create an SQL query like "SELECT location FROM products WHERE name='coffee maker';".

[1513] Step 8:

[1514] The server sends the generated SQL query to the database and executes the query. The database returns the location information of the coffee maker.

[1515] Step 9:

[1516] The server organizes location information retrieved from the database and generates a text response in a format that is easy for the user to understand. It also adjusts the tone and content of the response, taking into account the sentiment recognition results.

[1517] Step 10:

[1518] The server sends the text response to the speech synthesis engine, which converts it into speech data. The speech synthesis engine converts the text data back into speech data and sends it back to the server.

[1519] Step 11:

[1520] The device receives audio data from the server. The received audio data is played back to the user through the speaker.

[1521] Step 12:

[1522] In addition to voice responses, the device displays supplementary information on its screen. Specifically, the screen displays maps and product location information.

[1523] Throughout this entire processing step, users can quickly and accurately obtain the information they need within the commercial facility. Furthermore, the emotion engine provides friendly responses that are tailored to the user's emotions.

[1524] (Example 2)

[1525] 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".

[1526] Conventional unmanned search tools provide location and business hours information based on user voice input, but they lack the ability to generate responses that take user emotions into account, resulting in a limited user experience. Improving the speed and accuracy of information delivery is also a challenge.

[1527] 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.

[1528] In this invention, the server includes means for recognizing the user's emotions from voice input and text data, means for organizing search results and generating text in natural language based on the emotion recognition results, and means for converting text into voice data. This enables detailed responses tailored to the user's situation and emotions, improving the user experience and providing information quickly and accurately.

[1529] A "user" refers to the end user who attempts to obtain information using this system.

[1530] "Voice input" refers to the process of acquiring the voice a user speaks into a device as digital data.

[1531] "Text data" refers to data in string format converted by speech recognition.

[1532] "Natural language processing" refers to the technology of analyzing text data to identify its meaning and intent.

[1533] "Emotion recognition" refers to the technology that analyzes and identifies a user's emotional state from voice data and text data.

[1534] A "database" refers to a computerized information management system in which various types of information are systematically organized and stored.

[1535] A "query" refers to a set of commands executed on a database for searching or performing operations.

[1536] "Search results" refer to information obtained by executing queries against a database.

[1537] "Speech synthesis" refers to the technology that converts text data into speech data.

[1538] "Audio data" refers to audio information generated by speech synthesis and output to the user.

[1539] "Supplementary information" refers to additional visually displayed information (e.g., maps or location information) that is added to the voice response.

[1540] The present invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to the user's emotions. This system consists of the following parts:

[1541] (1) Acquisition of voice input

[1542] Terminal: The terminal is equipped with a microphone, which is used to acquire user voice input. When the user makes a question or request, their voice is transmitted to the system through the microphone. For example, if the user asks, "Where is the coffee maker?", that voice will be acquired by the terminal.

[1543] (2) Text conversion of voice input

[1544] Terminal: The acquired voice data is sent to a speech recognition engine (e.g., a speech recognition API), which converts the voice into text data. For example, the voice saying "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?".

[1545] (3) Analysis of text data

[1546] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker.

[1547] (4) Recognition of emotions

[1548] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes the user's emotions from the tone of their voice and the content of their text. For example, it can recognize emotions such as when the user is in a hurry or is frustrated.

[1549] (5) Querying the database

[1550] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it creates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database.

[1551] (6) Organizing search results and generating responses

[1552] Server: Organizes location information retrieved from the database and generates text responses in a user-friendly format. It adjusts the tone and content of the responses based on sentiment recognition results. For example, for a user in a hurry, it might generate a response such as, "We'll let you know shortly. The coffee maker is on the second floor in the home appliance section."

[1553] (7) Text to speech conversion

[1554] Server: The organized text responses are converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This data is expressed as natural speech corresponding to the user's question.

[1555] (8) Output of audio data to the user

[1556] Terminal: The created audio data is played back through the terminal's speaker, providing information to the user. In addition to the voice response, supplementary information (e.g., maps and location information) is displayed on the screen.

[1557] Specific example

[1558] Example 1: Product Search

[1559] When a user asks, "Where is the coffee maker?", the following process takes place:

[1560] 1. Terminal: The microphone acquires audio and sends the audio data to the server.

[1561] 2. Server: Converts audio data into text data and analyzes it.

[1562] 3. Server: A natural language processing engine identifies the intent of the question, and an emotion engine recognizes the user's emotions (e.g., being in a hurry).

[1563] 4. Server: Executes queries on the database to obtain location information.

[1564] 5. Server: Generates a text response saying, "We will inform you shortly. The coffee maker is located in the home appliance section on the second floor," and converts it into audio data.

[1565] 6. Terminal: Plays audio data and displays supplementary information on the display.

[1566] Example 2: Inquiry about business hours

[1567] The process for handling a user's question, "What are the opening hours of this store?" is as follows:

[1568] 1. Terminal: Acquires audio using the microphone and sends the audio data to the server.

[1569] 2. Server: Converts the audio data into text data and generates the text "Please tell me the opening hours of this store."

[1570] 3. Server: The natural language processing engine analyzes the text and identifies the user's intention, "I want to know the opening hours." The sentiment engine identifies the user's question.

[1571] 4. Server: Executes an SQL query on the database to retrieve business hours information.

[1572] 5. Server: Generates a text response, "This store is open every day from 9 am to 9 pm," and converts it into audio data.

[1573] 6. Terminal: Plays audio data and transmits it to the user through the speaker. Business hours are also displayed on the screen.

[1574] Example prompts for use with a generative AI model

[1575] "Please explain the specific processing flow when a user asks, 'Where is the coffee maker?'"

[1576] "Please explain how the system recognizes the user's emotions and how it generates responses based on those emotions."

[1577] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1578] Step 1:

[1579] Terminal: The terminal is equipped with a microphone, and the user provides voice input. For example, the user might ask, "Where is the coffee maker?" This voice data is acquired in digital format and sent to the next processing step. The input is the user's voice data, and the output is voice data in digital format.

[1580] Step 2:

[1581] Terminal: The acquired digital voice data is sent to a speech recognition engine (e.g., a speech recognition API). This engine analyzes the voice data and converts it into text data. For example, the voice data "Where is the coffee maker?" is converted into the text data "Where is the coffee maker?". The input is voice data, and the output is text data.

[1582] Step 3:

[1583] Server: Text data is sent to the server and parsed by a natural language processing engine (e.g., a natural language processing API). The parsing identifies the intent of the user's question. For example, it might identify that the user wants to know the location of the coffee maker. The input is text data, and the output is the intent of the user's question.

[1584] Step 4:

[1585] Server: Text and audio data are sent to an emotion engine (e.g., an emotion analysis API). The emotion engine analyzes emotions from the tone of voice and the content of the text. For example, it recognizes emotions such as when the user is in a hurry or is frustrated. The input is text and audio data, and the output is the user's emotional state.

[1586] Step 5:

[1587] Server: Based on the analysis results of the text data and the sentiment recognition results, the server executes queries against the database. For example, it generates an SQL query "SELECT location FROM products WHERE name='coffee maker';" to retrieve the location information of coffee makers from the database. The input is the analysis results and sentiment recognition results, and the output is the location information.

[1588] Step 6:

[1589] Server: Organizes location information obtained from the database and generates text responses in natural language based on sentiment recognition results. For example, it might generate a text response such as, "We will let you know shortly. The coffee maker is on the second floor in the home appliance section." The input is location information and sentiment recognition results, and the output is a text response.

[1590] Step 7:

[1591] Server: The generated text response is converted into speech data using a speech synthesis engine (e.g., a speech synthesis API). This conversion creates speech data that is natural-sounding, corresponding to the user's question. The input is a text response, and the output is speech data.

[1592] Step 8:

[1593] Terminal: The created audio data is sent to the terminal and played back to the user through the speaker. Additionally, supplementary information such as location data is displayed on the screen. Input is audio data and supplementary information, while output is playback of the audio response and display on the screen.

[1594] (Application Example 2)

[1595] 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".

[1596] In modern commercial facilities and stores, enabling users to quickly and accurately obtain information about the products and services they need is crucial for improving customer satisfaction. However, many customers may prefer not to ask store staff directly, and staff are not always available to assist. Therefore, there is a need for unmanned systems that efficiently answer customer questions. Furthermore, systems that can respond in accordance with customer emotions are expected to further enhance the customer experience.

[1597] 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. In this invention, the server includes means for analyzing the user's voice data and text data and recognizing the user's emotions, means for executing queries against the database and obtaining search results, and means for organizing the search results and generating text in natural language according to the emotions. This enables appropriate and rapid responses to user questions, as well as friendly responses that are tailored to the user's emotions.

[1598] "Means for obtaining voice input from the user" refers to devices or programs for collecting voices spoken by the user, and includes microphones and voice input interfaces.

[1599] "Means of converting voice input into text data" refers to technologies and programs that convert acquired voice into string information, and this includes speech recognition engines and their algorithms.

[1600] "Methods for analyzing text data using natural language processing to identify user intent" refers to technologies and programs that linguistically analyze string information to understand the intent behind user requests and questions, and includes natural language processing (NLP) engines.

[1601] "Means of analyzing user voice and text data to recognize user emotions" refers to technologies and programs that estimate psychological states and emotions based on the characteristics of voice and text, and emotion recognition engines are included in this.

[1602] "A means of executing queries against a database and retrieving search results" refers to the technology or program that sends commands to a database to obtain specific information and receives the response, and this includes SQL queries.

[1603] "Means for organizing search results and generating text in natural language according to emotions" refers to technologies and programs that organize acquired information into a certain structure and generate text in a way that aligns with the user's emotions, and response generation engines are included in this.

[1604] "Means of converting text into audio data" refers to technologies and programs that convert textual information into audio information, and this includes text-to-speech (TTS) engines.

[1605] "Means of outputting audio data to the user" refers to devices or programs that play back the generated audio information, and this includes speakers and audio playback functions.

[1606] "Means of displaying supplementary information on a display" refers to devices or programs that visually represent content in addition to information provided by voice, and this includes displays and screen display functions.

[1607] This invention is an unmanned search tool aimed at improving customer service in commercial facilities and stores, and incorporates an emotion engine that recognizes and responds to user emotions. A detailed embodiment of this system is described below.

[1608] System Configuration

[1609] The system primarily consists of terminals that acquire user voice input and servers that perform data analysis and response generation.

[1610] Hardware and software to be used

[1611] Hardware: Smartphones, microphones, speakers, displays, servers

[1612] software:

[1613] Speech recognition: Google Cloud Speech-to-Text API

[1614] Natural Language Processing (NLP): NLTK (Python's natural language processing library)

[1615] Emotion recognition: IBM Watson Natural Language Understanding

[1616] Database: MySQL

[1617] Text-to-Speech (TTS): Amazon Polly

[1618] System Operation Overview

[1619] 1. Acquisition of voice input

[1620] The user speaks into a smartphone app, and the voice is captured by the microphone.

[1621] Example: The user asks, "Where is the coffee maker?"

[1622] 2. Text conversion of voice input

[1623] The acquired audio is converted into text data using the Google Cloud Speech-to-Text API.

[1624] This means that the audio "Where is the coffee maker?" will be converted to text "Where is the coffee maker?".

[1625] 3. Analysis of text data

[1626] The server analyzes the generated text data using NLTK to identify the user's intent.

[1627] The analysis results identify the user's intention: "I want to know the location of the product (coffee maker)."

[1628] 4. Recognition of emotions

[1629] The server uses IBM Watson to recognize emotions from the user's text data and voice tone.

[1630] Emotion recognition can sometimes reveal that the user is in a hurry.

[1631] 5. Querying the database

[1632] The server queries the MySQL database to retrieve location information for the relevant product or service.

[1633] The query result retrieves the location information of the coffee maker (e.g., the home appliance section on the second floor).

[1634] 6. Organizing search results and generating responses

[1635] The server organizes the acquired location information and generates natural responses based on the user's emotions.

[1636] For example, for a user in a hurry, a response such as "I'll be right there. The coffee maker is on the second floor in the home appliance section" will be generated.

[1637] 7. Text to Speech Conversion

[1638] The server converts the generated text response into audio data using Amazon Polly.

[1639] This converts the text "The coffee maker is in the home appliance section on the second floor" into audio data.

[1640] 8. Output of audio data to the user

[1641] Audio data is played through the smartphone's speaker, providing information to the user.

[1642] Furthermore, supplementary information (e.g., maps and location information) is displayed on the smartphone screen.

[1643] Specific example

[1644] As a concrete example, the following prompt statements could be used:

[1645] Example of a prompt:

[1646] The user is asking "Where is the coffee maker?" via voice. Convert the voice to text, analyze it, and obtain the product's location information. Also, recognize the user's emotions and respond appropriately.

[1647] Based on this prompt, the system can generate appropriate answers to the user's questions and provide information quickly and accurately.

[1648] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1649] Step 1:

[1650] The user speaks into their smartphone, and voice input is obtained. The user states a question or request, and the device's (smartphone's) microphone captures the voice. The input is the user's voice, and the output is the captured voice data.

[1651] Step 2:

[1652] The acquired audio data is converted into text data using the Google Cloud Speech-to-Text API. The device sends the audio data to the speech recognition engine in the cloud and receives the text data in return. The input is audio data, and the output is the converted text data.

[1653] Step 3:

[1654] The server analyzes the received text data using NLTK to identify the user's intent. The server sends the text data to a natural language processing engine, which then identifies the user's request (e.g., "I want to know the location of the product") as a result of the analysis. The input is text data, and the output is the analysis result (user's intent).

[1655] Step 4:

[1656] The server analyzes text and audio data using IBM Watson to recognize the user's emotions. The server estimates emotions from the tone of voice and the content of the text. The input is text and audio data, and the output is recognized emotion data.

[1657] Step 5:

[1658] The server executes queries against the MySQL database and retrieves search results. Based on the user's intent and sentiment, the server executes queries to request information from the database. The input is the query information, and the output is the search results retrieved from the database (e.g., product location information).

[1659] Step 6:

[1660] The server organizes search results and generates text responses in natural language based on sentiment. The server then creates appropriate responses for the user based on the information it retrieves. Inputs are search results and sentiment data, while output is the generated text response.

[1661] Step 7:

[1662] The generated text response is converted into speech data using Amazon Polly. The server sends the text response to the speech synthesis engine and receives the speech data. The input is the text response, and the output is the converted speech data.

[1663] Step 8:

[1664] The device plays the received audio data and displays supplementary information on the display. The device plays audio through the speaker and displays related information on the display. The input is audio data and supplementary information, and the output is the audio and display content provided to the user.

[1665] 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.

[1666] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">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.

[1667] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1668] 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.

[1669] 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.

[1670] 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.

[1671] 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.

[1672] 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.

[1673] 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."

[1674] 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.

[1675] 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.

[1676] 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.

[1677] 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.

[1678] 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.

[1679] 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.

[1680] 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.

[1681] 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.

[1682] 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.

[1683] 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.

[1684] 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.

[1685] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1686] The following is further disclosed regarding the embodiments described above.

[1687] (Claim 1)

[1688] A means of obtaining voice input from the user,

[1689] A means of converting voice input into text data,

[1690] A method for analyzing text data using natural language processing to identify user intent,

[1691] A means of executing queries on a database and retrieving search results,

[1692] A means of organizing search results and generating text in natural language,

[1693] A means of converting text into audio data,

[1694] A means of outputting audio data to the user,

[1695] A means of displaying supplementary information on the display,

[1696] A system that includes this.

[1697] (Claim 2)

[1698] The system according to claim 1, further comprising means for obtaining relevant location information from a database and providing it to the user when the user's question concerns the location information of a product or service.

[1699] (Claim 3)

[1700] The system according to claim 1, further comprising means for obtaining relevant business hours information from a database and providing it to the user when the user's question concerns store business hours information.

[1701] "Example 1"

[1702] (Claim 1)

[1703] A means of obtaining voice input from the user,

[1704] A means of converting voice input into text data,

[1705] A method for analyzing text data using natural language processing to identify user intent,

[1706] A means of executing a query on a database and obtaining search results,

[1707] A means of organizing search results and generating text in natural language,

[1708] A means of converting text into audio data,

[1709] A means of outputting audio data to the user,

[1710] A means for displaying supplementary information on a display device,

[1711] A system that includes this.

[1712] (Claim 2)

[1713] The system according to claim 1, further comprising means for obtaining relevant location information from a database and providing it to the user when the user's question concerns the location information of a product or service.

[1714] (Claim 3)

[1715] The system according to claim 1, further comprising means for obtaining relevant business hours information from a database and providing it to the user when the user's question concerns store business hours information.

[1716] "Application Example 1"

[1717] (Claim 1)

[1718] A means of obtaining voice input from the user,

[1719] A means of converting voice input into text data,

[1720] A method for analyzing text data using natural language processing to identify user intent,

[1721] A means of executing queries on a database and retrieving search results,

[1722] A means of organizing search results and generating text in natural language,

[1723] A means of converting text into audio data,

[1724] A means of outputting audio data to the user,

[1725] A means of displaying supplementary information on the display,

[1726] A means of responding to customer questions in a physical store and providing necessary information via audio and visual means,

[1727] A system that includes this.

[1728] (Claim 2)

[1729] The system according to claim 1, further comprising means for obtaining relevant location information from a database and providing it to the user when the user's question concerns the location information of a product or service.

[1730] (Claim 3)

[1731] The system according to claim 1, further comprising means for obtaining relevant business hours information from a database and providing it to the user when the user's question concerns store business hours information.

[1732] "Example 2 of combining an emotion engine"

[1733] (Claim 1)

[1734] A means of obtaining voice input from the user,

[1735] A means of converting voice input into text data,

[1736] A method for analyzing text data using natural language processing to identify user intent,

[1737] A means of recognizing user emotions from voice input and text data,

[1738] A means of executing queries on a database and retrieving search results,

[1739] A means for organizing search results and generating text in natural language based on sentiment recognition results,

[1740] A means of converting text into audio data,

[1741] A means of outputting audio data to the user,

[1742] A means of displaying supplementary information on the display,

[1743] A system that includes this.

[1744] (Claim 2)

[1745] The system according to claim 1, further comprising means for obtaining relevant location information from a database and providing it to the user when the user's question concerns the location information of a product or service.

[1746] (Claim 3)

[1747] The system according to claim 1, further comprising means for obtaining relevant business hours information from a database and providing it to the user when the user's question concerns store business hours information.

[1748] "Application example 2 when combining with an emotional engine"

[1749] (Claim 1)

[1750] A means of obtaining voice input from the user,

[1751] A means of converting voice input into text data,

[1752] A method for analyzing text data using natural language processing to identify user intent,

[1753] A means of analyzing user voice and text data to recognize user emotions,

[1754] A means of executing queries on a database and retrieving search results,

[1755] A means of organizing search results and generating text in natural language based on sentiment,

[1756] A means of converting text into audio data,

[1757] A means of outputting audio data to the user,

[1758] A means of displaying supplementary information on the display,

[1759] A system that includes this.

[1760] (Claim 2)

[1761] The user's question concerns the location information of a product or service, and the means of obtaining the relevant location information from the database and providing it to the user is as follows:

[1762] The system according to claim 1, further comprising means for adjusting the response according to the user's emotion recognition result.

[1763] (Claim 3)

[1764] The user's question concerns the business hours information of a location, and there is a means to retrieve the relevant business hours information from the database and provide it to the user.

[1765] The system according to claim 1, comprising means for recognizing the user's questions or urgent status and adjusting the content and tone of the response. [Explanation of symbols]

[1766] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of obtaining voice input from the user, A means of converting voice input into text data, A method for analyzing text data using natural language processing to identify user intent, A means of executing queries on a database and retrieving search results, A means of organizing search results and generating text in natural language, A means of converting text into audio data, A means of outputting audio data to the user, A means of displaying supplementary information on the display, A system that includes this.

2. The system according to claim 1, further comprising means for obtaining relevant location information from a database and providing it to the user when the user's question concerns the location information of a product or service.

3. The system according to claim 1, further comprising means for obtaining relevant business hours information from a database and providing it to the user when the user's question concerns store business hours information.

Images