System

Abstract

A system is provided.SOLUTION: This system is provided with a means for responding to a telephone call by using an automatic response means when the telephone call arrives, and for recording call contents, a generation means for analyzing and summarizing the recorded call contents, a means for reporting the summarized call contents to the terminal of a user, and a selection means for allowing the user to respond based on the summarized contents.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] The problem is that a lot of time is spent answering the phone. There is a growing need, especially among businessmen and managers, to be able to focus only on important calls. There is also an increasing number of individuals who want to protect their families from the anxiety of nuisance and fraudulent calls. The objective of this invention is to reduce these time and anxieties and further improve the efficiency of answering the phone. [Means for solving the problem]

[0005] The present invention provides a system in which an automatic answering means answers an incoming call and records the contents of the call. The recorded call contents are analyzed and summarized by a generating means. The summarized call contents are notified to the user's terminal, and the user can use a selection means to decide how to respond based on the summary. Specifically, the automatic answering means responds to the caller based on a pre-set script, and the generating means analyzes and summarizes the call contents using an artificial intelligence model. This significantly reduces the time spent on unimportant phone calls, allowing the user to focus only on important phone calls. This system meets individual needs and achieves efficient phone response.

[0006] The "automatic answering means" is a function that automatically answers an incoming call and records the caller's comments.

[0007] "Recording means" is a function for saving the contents of a call as text data.

[0008] The "generation means" is a function that includes an artificial intelligence model for analyzing stored text data, extracting important points, and summarizing them.

[0009] The "notification means" is a function for transmitting the summarized contents of the call to the user's terminal.

[0010] The "selection means" is a function that allows the user to determine the need for a response based on the summarized content of the call and decide on an action if necessary.

[0011] "Artificial intelligence model" refers to the AI ​​technology used to analyze text data and generate summaries.

[0012] "User device" refers to the device used to receive the summarized call content, such as a smartphone or mobile phone. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0014] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0017] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0018] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0019] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0026] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 2, in the data processing device 12, a specific process 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" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process 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.

[0031] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0033] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0034] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[0035] 1. Automated Response Methods

[0036] When a call comes in, the server activates the automated response means, which responds appropriately to the caller based on a pre-set script. For example, the automated response means might say, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves them as text data.

[0037] 2. Generation means

[0038] The recorded content of the call is analyzed by the generation means. The server sends the recorded text data to an AI model, which summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0039] 3. Means of notification

[0040] Once the summary is generated, the server sends it to the user's device as an SMS. At this time, the summary is conveyed to the user as a short message. Specifically, a message such as "Proposal for the introduction of a new product. Please schedule a date for the next meeting to discuss the details" will be sent to the user's smartphone.

[0041] 4. Selection methods

[0042] Users can check the contents of the SMS they receive and decide whether or not a response is necessary. For example, if they determine that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore it.

[0043] Specific examples

[0044] As an example, consider a scenario in which a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic answering means, responding with "Hello, thank you for calling. How can I help you?" The caller then introduces the new product and requests for a next meeting, and the content is recorded and saved as text data. The server then uses a generation means to analyze the text data and summarizes the content of the call as "Explanation regarding the proposed introduction of the new product. We would like to set a date for the next meeting to discuss the details." This summary is then sent to the user's smartphone as an SMS. The user checks the content of the SMS and, determining that it is important, calls back and sets up a detailed meeting.

[0045] The present invention allows users to concentrate on important phone calls without being bothered by unnecessary calls, and is also effective as a means for avoiding nuisance calls and fraudulent calls.

[0046] The processing flow will be explained below.

[0047] Step 1:

[0048] Incoming calls and starting automatic answering

[0049] Server: When a call comes in, the server detects the call and activates an automated attendant, which greets the caller based on a pre-defined script (e.g., "Hello, thank you for calling. How can I help you?").

[0050] Step 2:

[0051] Call recording

[0052] Server: The automated answering system receives the caller's speech, and the server records the call in real time. The recorded content is converted into text data and stored in data storage.

[0053] Step 3:

[0054] Ending the call and preparing to send data

[0055] Server: When the call ends, prepares to send the recorded text data to the generation means. At this time, the process of converting the recorded file into text format is executed.

[0056] Step 4:

[0057] Generative data analysis and summary generation

[0058] Server: The recorded text data is sent to the AI ​​model, which analyzes the data, extracts key points, and generates a summary.

[0059] Example: A summary such as "Explanation regarding the proposal to introduce a new product. I would like to set a date for the next detailed meeting." is generated.

[0060] Step 5:

[0061] Convert summary content to SMS format and send

[0062] Server: Converts the generated summary into SMS format, then sends the SMS message to the user's device.

[0063] Step 6:

[0064] Receipt and confirmation by the user

[0065] User: The SMS arrives on the user's device and a summary is displayed. The user confirms the SMS content.

[0066] Example: A message is sent to the terminal saying, "Proposal for the introduction of a new product. I would like to set a date for the next detailed meeting."

[0067] Step 7:

[0068] Deciding and implementing a response

[0069] User: After reviewing the summary, the user decides what to do next based on the summary, and if necessary, calls back or takes other action.

[0070] Example: The user decides that the summary is important and calls back to schedule a meeting to discuss the details.

[0071] This system's series of processes allows users to focus on important phone calls without being bothered by unnecessary calls, and also makes it easier to avoid nuisance and fraudulent calls.

[0072] Example 1

[0073] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0074] In conventional telephone answering systems, users must answer the phone in person to check the content of incoming calls, which leads to problems such as wasted call time and increased burden from nuisance calls.In addition, there is a high risk that important call content will be overlooked, so an effective filtering method is necessary.

[0075] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0076] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the calls, a means for analyzing and summarizing the recorded contents of the calls, a means for notifying the user of the summarized contents of the calls to the user's terminal, a means for allowing the user to select and respond based on the summary, a means for saving the contents of the calls as text data, a means for sending the text data to an artificial intelligence model for analysis, and a means for summarizing the contents of the summaries in an easy-to-read format and sending them to the user's terminal. This allows the user to quickly respond to important calls without being bothered by unnecessary calls. It also makes it possible to avoid nuisance calls and fraudulent calls.

[0077] An "automatic answering means" is a means for automatically answering an incoming call and responding based on a pre-set script.

[0078] "Means for recording telephone conversations" refers to a means for recording telephone conversations as audio data and saving it.

[0079] The "generation means" is a means for analyzing the recorded content of a call and summarizing it using an artificial intelligence model.

[0080] "Means of notifying the user's device" refers to a means of sending a summary of the call contents to the user's mobile device or smartphone in the form of SMS or the like.

[0081] The "selection means" is a means for the user to check the received summary content and determine the need for action based on its importance.

[0082] "Means for saving as text data" refers to means for converting recorded voice data into text format and saving it.

[0083] An "artificial intelligence model" is a technology that uses machine learning techniques to analyze text data and perform summarization or other processing. Specific examples include generative AI models.

[0084] "Means for summarizing in an easy-to-read and concise manner" refers to a means for omitting the detailed parts of the generated summary and extracting only the main points to express them concisely.

[0085] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[0086] 1. Automated Response Methods

[0087] When a call comes in, the server activates an automatic answering means. The automatic answering means responds appropriately to the caller based on a pre-set script. For example, it might respond, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves it as audio data.

[0088] 2. Recording Method

[0089] The server stores the recorded conversation as audio data and also as text data, using voice recognition technology.

[0090] 3. Generation means

[0091] The text data of the call is sent by the server to an artificial intelligence model (e.g., OpenAI (registered trademark) GPT-4 (registered trademark)), which analyzes and summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0092] 4. Means of notification

[0093] The server then sends the generated summary to the user's device as an SMS message. The summary is written in a concise and easy-to-read format. For example, the user's smartphone will receive a message such as, "Proposal for the introduction of a new product. We would like to set up a date for the next meeting to discuss the details."

[0094] 5. Selection methods

[0095] Users can check the contents of the SMS they receive and decide whether the call is important or not. If it is, they can call back and schedule a detailed meeting. If it is not important, they can ignore the call.

[0096] Hardware and software used

[0097] Server: Cloud computing service (e.g., AWS (registered trademark), Google (registered trademark) Cloud)

[0098] Speech recognition technology: Technology that converts voice data into text data

[0099] Artificial intelligence models: Generative AI models such as OpenAI GPT-4

[0100] SMS sending service: Use a service such as Twilio

[0101] Specific examples

[0102] Consider the case where a businessman receives a sales call on his mobile phone. The server detects the incoming call and quickly activates an automated answering system, responding with, "Hello, thank you for calling. How can I help you?" The caller introduces the new product and requests a next meeting, and the conversation is recorded and saved as voice data by the server. The server then saves this voice data as text data using speech recognition technology, and then analyzes the text data using a generation system to summarize the conversation as, "An explanation of the proposed introduction of the new product. I would like to set a date for the next meeting to discuss the details." This summary is sent to the businessman's smartphone as an SMS. The businessman checks the contents of the SMS and, determining that it is important, calls back himself to set up a detailed meeting.

[0103] Prompt Sentence Examples

[0104] "Please tell me the procedure for the system to automatically answer incoming calls and analyze and summarize the contents of the calls."

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Step 1:

[0107] When the server detects an incoming call, it activates the auto-answering means. Specifically, it receives an incoming call signal and executes the auto-answering program. The input is the incoming call signal, and the output is a trigger signal to start the response. Based on this trigger signal, the server executes a pre-set script and plays an auto-answering message such as "Hello, thank you for calling. How can I help you?"

[0108] Step 2:

[0109] The server records the contents of telephone conversations and saves them as audio data. From the moment a call begins, the server activates the recording function and captures all of the call content. The input is the audio signal during the call, and the output is the recorded audio data. This audio data is saved in a database after the call ends.

[0110] Step 3:

[0111] The server converts the stored voice data into text data using voice recognition technology. This process uses voice recognition software. The input is the recorded voice data, and the output is text data generated by the voice recognition software. The server stores this text data and prepares it for the next analysis step.

[0112] Step 4:

[0113] The server analyzes the text data using a generation method and generates a summary. Specifically, the text data is sent to a generative AI model (e.g., OpenAI GPT-4), which summarizes the content. The input is text data, and the output is summarized text data. For example, the content of a long sales call can be summarized as "An explanation of the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0114] Step 5:

[0115] The server sends the generated summary to the user's device as an SMS message. Specifically, it uses an SMS sending service (e.g., Twilio) to compose and send the message. The input is the summarized text data, and the output is an SMS message sent to the user's device. For example, a message such as "Proposal for the introduction of a new product. Please schedule the next meeting to discuss the details" is sent to the user's smartphone.

[0116] Step 6:

[0117] Users can check the contents of the SMS they receive and decide how to respond based on its importance. The input is the SMS message, and the output is the action the user takes (e.g., call back or ignore). If the user determines that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore the message.

[0118] (Application example 1)

[0119] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0120] In recent years, the number of nuisance and fraudulent phone calls has been increasing, causing users to be bothered by unnecessary calls and potentially miss important calls. In addition, businesses need to handle a large number of phone calls, creating a need for efficient call management. Furthermore, there is a need for a method to reduce the time and effort required to understand the content of calls that require specific responses. To solve these issues, there is a need for technology that can automatically analyze and summarize the content of phone calls and communicate it to users.

[0121] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0122] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the call, a means for analyzing the recorded contents of the call and generating a summary, a means for notifying the user of the summarized contents of the call to the user's terminal, and a notification means for providing the summarized contents via multiple notification means (e.g., SMS and email). This allows the user to quickly grasp important summarized information without having to listen to the entire contents of the call. Furthermore, the diversification of notification means allows the user to receive the summary in the most optimal way depending on their situation and priority.

[0123] An "automatic answering means" is a device or software that automatically answers an incoming call and records the content of the call.

[0124] "Means for recording telephone conversations" refers to a device or software for saving the contents of incoming telephone conversations as voice data or text data.

[0125] The "generator" is a device or software that analyzes the recorded conversation content, extracts important information, and generates a summary.

[0126] The "notification means" is a device or software for transmitting the generated summary to the user's terminal.

[0127] The "selection means" is a device or software that allows the user to determine the importance based on the summary received and select an appropriate response.

[0128] "Multiple notification methods" is a general term for methods of conveying information to users in different ways, such as SMS and email.

[0129] An "artificial intelligence model" is an algorithm or software used to analyze call content and generate summaries using machine learning and deep learning techniques.

[0130] This invention is a system that automatically answers an incoming call, analyzes and summarizes the contents of the call, and notifies the user. Next, we will explain how to specifically implement this system.

[0131] This system mainly uses a server, user terminals, and a network environment. The server is the main hardware for running a program that integrates speech recognition, text analysis, summary generation, and notification functions, and uses the following software to realize each function. For speech recognition, the "speech_recognition" Python library is used, and for summary generation, the "t5-small" model from Hugging Face, a generative AI model from "transformers," is used. Notification methods include sending SMS using the "Twilio" API and email using Python's "smtplib."

[0132] When the server detects an incoming call, it activates an automatic answering system and answers the call based on a pre-defined script. During this process, the incoming voice is recorded through a microphone and converted into text data using the "speech_recognition" library. The converted text data is then stored on the server.

[0133] The server then sends the stored text data to a generative AI model in the "transformers" library to generate a summary, which is a short piece of text containing important information.

[0134] This summary is sent to the user's device via SMS using the Twilio API or via email using smtplib. The user can then review the message and decide how to respond based on its importance.

[0135] Specific examples

[0136] For example, consider a scenario in which this system is implemented in a company's customer support. When a new customer support inquiry is received by phone, the server uses an automated response function to respond with, "Hello, this is Company Name Support. How can I help you?" The call is recorded and then converted into text using the "speech_recognition" library. A generative AI model in the "transformers" library then summarizes the call as, "There is a problem with Product A. Please contact the user for details." This summary is then sent to the representative's device via SMS and email.

[0137] Prompt Sentence Examples

[0138] Example prompt sentence:

[0139] Summarize the customer support call.

[0140] Voice input: "Hello, thank you for contacting us about an issue with Product A. Please tell us the specific problem you're experiencing and how it's progressing."

[0141] Voice output: "There is a problem with product A. Please check with the user for details."

[0142] In this way, the server automatically analyzes and summarizes the contents of the call and quickly conveys only the important information to the user, enabling efficient and effective telephone response.

[0143] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0144] Step 1:

[0145] When a call is received, the server activates an automatic answering means.

[0146] Input: Incoming phone signal.

[0147] Data processing and calculation: Detects incoming calls and loads pre-configured scripts to prepare an automatic response.

[0148] Output: Generates a response message based on the auto-response script.

[0149] Step 2:

[0150] The server answers the call using an automated answering means and records the contents of the call.

[0151] Input: Greeting message and call audio based on the auto-attendant script.

[0152] Data processing and calculation: Capture the contents of the call as audio data through the microphone and generate a recording file.

[0153] Output: A recording file of the call saved as audio data.

[0154] Step 3:

[0155] The server converts the recorded conversation into text data using a voice recognition library.

[0156] Input: Recording file (audio data).

[0157] Data processing and calculation: Analyzes voice data using the speech recognition library (speech_recognition) and generates text data.

[0158] Output: The call contents converted into text data.

[0159] Step 4:

[0160] The server sends the text data to a generative AI model to generate a summary.

[0161] Input: Text data.

[0162] Data processing and calculation: Summarize text data using a generative AI model (the t5-small model from the transformers library).

[0163] Output: Summarized text data.

[0164] Step 5:

[0165] The server sends the summarized content to the user's device via multiple notification methods (SMS and email).

[0166] Input: Abstracted text data.

[0167] Data processing and calculation: Using notification methods (Twilio API and smtplib), the summary data is sent to the user's terminal as SMS and email.

[0168] Output: Summary notification via SMS and email received on user device.

[0169] Step 6:

[0170] The user reviews the received summary and decides on the appropriate response, if necessary.

[0171] Input: Summary data sent via SMS or email.

[0172] Data processing and calculation: Determine importance and urgency based on summary data.

[0173] Output: The subsequent action resulting from the user's decision (e.g., calling back, replying by email, etc.).

[0174] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0175] This invention combines a system that uses an automatic answering means to respond to incoming calls, records and analyzes the content of the call, and an emotion engine that recognizes the user's emotions. This enables flexible responses that match the user's emotional state, resulting in more effective and personalized notifications.

[0176] System configuration

[0177] This system is configured as follows:

[0178] 1. Automated Response Methods

[0179] When a call comes in, the server activates an automated answering machine, which responds to the call based on a pre-defined script.

[0180] 2. Recording Method

[0181] The contents of the call are recorded in real time by the server and saved as text data.

[0182] 3. Generation means

[0183] The recorded text data is sent to a generation means, specifically an artificial intelligence model, where it is analyzed and summarized.

[0184] 4. Emotion Engine

[0185] The emotion engine analyzes emotions from the user's voice, facial expressions, and text input. For example, it analyzes facial expressions and voice when the user opens an SMS to determine the user's emotional state.

[0186] 5. Means of notification

[0187] Based on the analysis results of the emotion engine, the server sends a summary of the call to the user's device in a user-friendly format as an SMS. If the user's emotional state indicates stress, the server can summarize the content briefly or adjust the notification method.

[0188] 6. Selection methods

[0189] Users can check the contents of the SMS they receive and choose what to read based on their emotional state, for example, they can skip over messages they deem unimportant and check the details of messages they deem important.

[0190] Specific examples

[0191] Consider a scenario where a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic response means. The automatic response means responds with "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server uses the generation means to analyze the text data and summarize it as "An explanation of the proposed introduction of a new product. We would like to set a date for the next meeting to discuss the details."

[0192] The emotion engine analyzes data from the user's smartphone camera and microphone to determine the user's emotional state. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary of the message. The user can check the received SMS, check the details if it is important, or skip if it is not necessary.

[0193] Specific processing of the program

[0194] 1. Detecting incoming calls

[0195] When the server detects an incoming call, it activates an automatic answering means and starts a call. The automatic answering means answers the call according to a pre-set script.

[0196] 2. Recording of call content

[0197] The server records the call in real time, converts it into text data, and saves it.

[0198] 3. Summary Generation

[0199] A generating means is used to analyze the recorded text data and generate a summary, which is then converted into an SMS format.

[0200] 4. Sentiment Analysis and Notifications

[0201] The emotion engine analyzes the user's emotional state and adjusts the content and method of notifications based on the results. For example, if the user is feeling stressed, the summary content will be shortened or the notifications will be suppressed. The adjusted content will be sent to the user's device as an SMS.

[0202] 5. User Choice

[0203] The user can check the summary of the SMS they received and decide whether or not a response is necessary. If necessary, they can take action such as calling back.

[0204] This system allows users to quickly obtain only the important information without wasting time on unnecessary calls. Utilizing an emotion engine also reduces the user's mental burden and enables more personalized notification responses.

[0205] The processing flow will be explained below.

[0206] Step 1:

[0207] Incoming calls and starting automatic answering

[0208] Server: When a call comes in, the server detects the call and activates an automated response system, which responds to the caller based on a pre-defined script, saying, "Hello, thank you for calling. How can I help you?"

[0209] Step 2:

[0210] Call recording

[0211] Server: When the automated answering system receives the call, the server records the call in real time. The recorded content is converted into text data and stored in a database.

[0212] Step 3:

[0213] Ending the call and preparing to send data

[0214] Server: When the call ends, prepares the recorded text data to be sent to the generation means. During this data conversion process, the text data is formatted.

[0215] Step 4:

[0216] Generative data analysis and summary generation

[0217] Server: The recorded text data is sent to the AI ​​model, which analyzes the text data, extracts key points, and creates a summary.

[0218] Example: If the call content is "Pitching a new product and requesting a next meeting," the generated summary will be "Introducing a new product. Requesting a detailed meeting."

[0219] Step 5:

[0220] Analyzing user emotions with an emotion engine

[0221] Device: Uses the camera and microphone on the user's smartphone to capture the user's facial expressions and voice data.

[0222] Server: The acquired facial expression and voice data is sent to the emotion engine, which analyzes the user's emotional state (e.g., stress level, excitement level, relaxation level).

[0223] Step 6:

[0224] Tailoring notification content based on emotional state

[0225] Server: Based on the user's emotional state analyzed by the emotion engine, the notification method and content of the summarized call details are adjusted. For example, if the user is feeling stressed, the notification content is made shorter.

[0226] Step 7:

[0227] Convert summary content to SMS format and send

[0228] Server: Converts the generated summary into SMS format and sends the adjusted notification content to the user's device as an SMS.

[0229] Example: The summary "Introducing a new product. Requesting a meeting to discuss details" can be converted into an SMS message as "Important: Introduces a new product. Requesting a meeting to discuss details."

[0230] Step 8:

[0231] Receipt and confirmation by the user

[0232] User: An SMS is delivered to the user's device. The user checks the contents of the SMS.

[0233] Example: The device displays the message "Important: Introducing a new product, please arrange a meeting."

[0234] Step 9:

[0235] Deciding and implementing a response

[0236] User: Based on the summary of the SMS received, the user decides whether to take action and, if necessary, take action such as calling back to confirm the details.

[0237] Example: The user decides it's important, calls the sales rep back, and schedules a meeting.

[0238] This system's series of processes allows users to quickly obtain only the important information without being bothered by unnecessary calls. In addition, the use of an emotion engine reduces the user's mental burden and enables more personalized responses.

[0239] Example 2

[0240] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0241] While automated answering and recording of incoming phone calls are commonplace today, this alone is not sufficient for user convenience. In particular, without call content summarization and emotional analysis, users are often overwhelmed with unnecessary information, which can increase their mental burden. Furthermore, special consideration is required for users under stress, and flexible responses tailored to individual emotional states are required. Therefore, there is a need for a system that not only analyzes and summarizes call content, but also recognizes the user's emotions and adjusts notification content accordingly.

[0242] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0243] In this invention, the server includes means for recording the contents of a call when a call is received, means for analyzing the recorded contents of the call and generating a summary, means for analyzing the user's emotions, and means for adjusting the summarized contents of the call based on the user's emotional state and notifying the user's terminal. This makes it possible to quickly provide important information while taking the user's emotional state into consideration.

[0244] An "incoming call" refers to receiving an external call to a communication device.

[0245] "Automatic answering means" refers to a function that responds to incoming calls with a pre-set script or message without manual operation.

[0246] "Call content" refers to all audio information exchanged over the telephone.

[0247] "Recording means" refers to a device or software for storing the contents of a call.

[0248] "Generation means" refers to a processing device or software for analyzing the recorded content of a call and generating a summary.

[0249] "Emotion analysis means" refers to the function of analyzing the user's emotional state from audio and video data.

[0250] "Abridged call content" refers to information that briefly summarizes the content of a call.

[0251] "User's device" refers to a communication device used by a User that can receive System notifications.

[0252] "Notification means" refers to the function of delivering the generated summary content to the user's terminal.

[0253] "Selection tool" refers to a function that allows a user to select an appropriate response based on the summary content received.

[0254] An "artificial intelligence model" refers to an algorithm or system that learns from data and performs analysis and predictions.

[0255] This invention combines an emotion analysis system that recognizes the user's emotions with a system that uses an automatic answering means to respond to incoming calls and records and analyzes the content of the call, thereby enabling notifications to be delivered in line with the user's emotional state, enabling more effective and personalized information to be provided.

[0256] This system is configured as follows:

[0257] Hardware and Software Use

[0258] Server: VoIP system, Python script, Twilio API, Google Cloud Speech-to-Text API, OpenAI GPT-4, Azure® Cognitive Services

[0259] User's device: Smartphone camera, microphone, SMS function

[0260] Specific processing

[0261] Incoming calls and auto-answering

[0262] The server detects an incoming call. When an incoming call is detected, it activates an automated response system that responds to the caller based on a pre-defined script. For example, it might send a message like this: "Hello, thank you for calling. How can I help you?"

[0263] Call recording

[0264] The server records the call in real time and converts the audio data into text data for storage. This process uses the Google Cloud Speech-to-Text API.

[0265] Generate a summary of the call

[0266] The server then uses a generation method to analyze the recorded text data and generate a summary. Using the OpenAI GPT-4 generative AI model, the server inputs the following prompt: "Please summarize the following call content: [Text data]."

[0267] Emotion analysis

[0268] The emotion analyzer uses data obtained from the smartphone camera and microphone to analyze the user's emotional state. This data is analyzed through Microsoft® Azure Cognitive Services to determine the user's emotional state.

[0269] Call summary notifications

[0270] Based on the results of the emotion analysis, the server sends a summary of the call to the user's device as an SMS. If the user's emotional state indicates stress, the summary will be made more concise or the notification method will be adjusted.

[0271] Specific examples

[0272] When a businessman receives a sales call on his mobile phone, the server detects the incoming call and activates the automatic response means. The automatic response means responds with, "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server then uses the generation means to summarize it as, "This is an explanation of the proposal to introduce a new product. I would like to set a date for the next meeting to discuss the details."

[0273] The emotion analysis function analyzes the user's emotions based on data collected from the smartphone camera and microphone. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary. The user can check the received SMS and, if the content is deemed important, review the details, or skip if it is not necessary.

[0274] This invention allows users to quickly obtain only the information they need without wasting time on unnecessary calls. In addition, emotion analysis reduces the user's mental burden and enables more personalized notification responses.

[0275] Prompt Sentence Examples

[0276] Please summarize the call below.

[0277] [Text data]

[0278] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0279] Step 1:

[0280] The server detects incoming calls. When a call comes in, it activates an automated answering system that responds to the caller based on a pre-defined script. For example, it might play a message like, "Hello, thank you for calling. How can I help you?"

[0281] Input: Incoming signal

[0282] Output: Start auto-reply, play greeting

[0283] What happens: The VoIP system detects an incoming call, and a Python script uses the Twilio API to play a pre-recorded message.

[0284] Step 2:

[0285] The server records the call in real time and converts the audio data into text data for storage, using the Google Cloud Speech-to-Text API.

[0286] Input: Call audio data

[0287] Output: Text data

[0288] How it works: The recorded audio data is saved in WAV format, and a Python script calls the Google Cloud Speech-to-Text API to convert the audio data into text data.

[0289] Step 3:

[0290] The server uses a generative method to analyze the text data and generate a summary. At this stage, a generative AI model (OpenAI GPT-4) is used.

[0291] Input: Text data

[0292] Output: Summary text

[0293] How it works: The Python script inputs the following prompt to the generative AI model: "Please summarize the following call content: [Text data]." The generative AI model outputs the summary text.

[0294] Step 4:

[0295] The server analyzes data obtained from the user's camera and microphone using emotion analysis tools, which are powered by Microsoft Azure Cognitive Services.

[0296] Input: Audio and video data from cameras and microphones

[0297] Output: Emotion analysis results

[0298] How it works: A smartphone app collects data from the camera and microphone and sends it to a server, which then sends it to Azure Cognitive Services for sentiment analysis.

[0299] Step 5:

[0300] The server adjusts the notification content based on the results of the sentiment analysis and sends a summary text to the user's device. The notification method uses the Twilio API.

[0301] Input: Summary text, sentiment analysis results

[0302] Output: SMS notification

[0303] Specific operation: The server compiles a concise summary text based on the sentiment analysis results and sends it as an SMS to the user's smartphone using the Twilio API.

[0304] Step 6:

[0305] Users can check the received SMS and respond as needed, for example, by checking the details if the message is important and calling back if necessary.

[0306] Input: SMS notification

[0307] Output: User action

[0308] Specific action: The user opens the SMS on their smartphone, checks the contents, and if necessary calls back or takes other action.

[0309] (Application example 2)

[0310] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0311] Current automated answering systems are effective in answering incoming calls and summarizing the content of calls, but they do not provide flexible responses that take into account the user's emotional state. This can leave users feeling stressed and dissatisfied, resulting in problems such as inefficiencies such as an increase in complaints and inquiries. Customer service in brick-and-mortar stores also needs to be able to respond in a way that takes into account the emotional state of customers, but this is difficult to achieve with conventional systems.

[0312] The specific processing 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 answering an incoming call using an automatic answering means and recording the content of the call, means for analyzing and summarizing the recorded content of the call, means for analyzing the emotional state of the user, and means for adjusting the content of the notification based on the analysis result. This enables flexible and personalized responses according to the emotional state of the user.

[0313] "When a call comes in" refers to a situation where an external connection is attempted over a telephone line and the user's device or system receives the connection.

[0314] "Automatic answering means" refers to a device or program that has the function of automatically answering an incoming telephone call without human intervention.

[0315] "Call recording means" refers to a device or program that automatically saves conversations and audio data that occur during a telephone call.

[0316] "Recorded call content" refers to information saved as audio or text data exchanged during a call.

[0317] "Analysis and summary generation means" refers to a device or program that analyzes the recorded content of a call, extracts only the important information, and provides it in a shortened form.

[0318] "Emotion analysis means for analyzing a user's emotional state" refers to technologies and algorithms for measuring or inferring emotions from a user's voice, facial expressions, text input, etc.

[0319] "Notification means for adjusting notification content" refers to a device or program for automatically changing the method or content of notifications based on the user's emotional state.

[0320] "Means for notifying the user of the summarized call content" refers to a device or program that transmits the information summarized by the generating means to the user's device such as a smartphone or PC.

[0321] "A means for the user to decide what to do based on the summary content" refers to a device or program that allows the user to look at the summary information notified to them and decide what to do next based on its importance and necessity.

[0322] This invention provides a system that uses an automatic answering means to answer incoming calls, records the contents of the call, analyzes the record to generate a summary, notifies the user of the summary, and further analyzes the user's emotional state to optimize the content of the notification. The following describes an example of application of this system to customer support functions in brick-and-mortar stores.

[0323] System Program

[0324] The server has a speech recognition engine, a text generation engine, an emotion analysis engine, a notification engine, and the ability to send emails using the SMTP protocol. It is recommended that the main software configuration uses the speech_recognition library, the TextBlob library, the pipeline function of the transformers library, and the smtplib library.

[0325] Explanation of program processing

[0326] 1. Incoming call detection and automatic answering

[0327] When a call comes in, the server's automatic answering means is activated, which responds to the call and gives an initial response to the customer based on a pre-set script.

[0328] 2. Recording of call content

[0329] The server uses a speech recognition engine (e.g., speech_recognition library) to record the call and save it as text data.

[0330] 3. Summary Generation

[0331] The recorded text data is analyzed by a text generation engine (e.g., the TextBlob library), which extracts key phrases and concepts and generates summaries based on them.

[0332] 4. Emotional state analysis

[0333] The summarized text and audio data are analyzed for the user's emotional state using a sentiment analysis engine (e.g., the pipeline function in the transformers library).

[0334] 5. Coordination and delivery of notifications

[0335] Based on the results of the sentiment analysis, the notification engine will tailor the notification content. For example, if it detects that the user is feeling stressed, a brief summary of the content will be sent. Notifications are implemented via the SMTP protocol (e.g., the smtplib library).

[0336] Specific examples

[0337] This technology is applied to customer service in brick-and-mortar stores. When a customer makes an inquiry, the in-store customer support robot automatically responds, records the content of the inquiry, and saves it as text data. The saved text data is analyzed in real time, and a summary is generated. At this time, the emotional state of the customer is also analyzed along with the analyzed summary data. For example, if a customer expresses anger, this information is notified to staff, and a flexible response is required.

[0338] Prompt Sentence Examples

[0339] "Analyze the following speech input and output a result about the emotional state: 'Hello, customer. Can you tell me more about this weekend's sale? I'm happy.'"

[0340] In this way, the system can provide a more personalized customer experience and improve the user experience.

[0341] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0342] Step 1:

[0343] When a call comes in, the server detects the call. It receives the incoming call signal and activates the automatic answering means. The automatic answering means answers the call and starts the call based on a pre-set script. The input is the incoming call signal and the output is the automatic answering message.

[0344] Step 2:

[0345] The server uses a speech recognition engine (e.g., the speech_recognition library) to record the call contents. The input is the voice data during the call, and the voice data is converted into text data in real time and saved. The output is the converted text data.

[0346] Step 3:

[0347] The server uses a generator (e.g., TextBlob library) to analyze the recorded text data, extract important key phrases, and generate a summary. The text data is analyzed and key phrases and key content parts are extracted. The input is the text data of the call, and the output is the summarized text data.

[0348] Step 4:

[0349] The server uses sentiment analysis tools (e.g., the pipeline function in the transformers library) to analyze the user's emotional state from the summarized text and audio data. It analyzes the audio and text data and determines the emotional state (e.g., positive, negative, neutral). The input is the summarized text and audio data, and the output is the sentiment analysis result.

[0350] Step 5:

[0351] The server uses a notification mechanism to adjust the notification content based on the emotion analysis results. For example, if the user is feeling stressed, the server adjusts the notification content by making the summary more concise. The input is the emotion analysis results and the summarized text data, and the output is the adjusted notification content.

[0352] Step 6:

[0353] The server sends the tailored notification content to the user's device. The notification is sent in the form of email or other format using the SMTP protocol. A notification engine (e.g., the smtplib library) is used to send emails with detailed information to staff or other interested parties. The input is the tailored notification content, and the output is the notification sent to the user's device.

[0354] Step 7:

[0355] The user checks the received notification and decides how to respond if necessary. Based on the content of the notification, the next action (e.g., direct response by staff, request for additional information) is taken. The input is the notification sent to the user's device, and the output is the user's decision on action.

[0356] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0357] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0358] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0359] [Second embodiment]

[0360] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0361] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0362] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0363] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0364] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0365] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0366] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0367] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0368] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0369] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0370] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0371] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0372] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[0373] 1. Automated Response Methods

[0374] When a call comes in, the server activates the automated response means, which responds appropriately to the caller based on a pre-set script. For example, the automated response means might say, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves them as text data.

[0375] 2. Generation means

[0376] The recorded content of the call is analyzed by the generation means. The server sends the recorded text data to an AI model, which summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0377] 3. Means of notification

[0378] Once the summary is generated, the server sends it to the user's device as an SMS. At this time, the summary is conveyed to the user as a short message. Specifically, a message such as "Proposal for the introduction of a new product. Please schedule a date for the next meeting to discuss the details" will be sent to the user's smartphone.

[0379] 4. Selection methods

[0380] Users can check the contents of the SMS they receive and decide whether or not a response is necessary. For example, if they determine that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore it.

[0381] Specific examples

[0382] As an example, consider a scenario in which a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic answering means, responding with "Hello, thank you for calling. How can I help you?" The caller then introduces the new product and requests for a next meeting, and the content is recorded and saved as text data. The server then uses a generation means to analyze the text data and summarizes the content of the call as "Explanation regarding the proposed introduction of the new product. We would like to set a date for the next meeting to discuss the details." This summary is then sent to the user's smartphone as an SMS. The user checks the content of the SMS and, determining that it is important, calls back and sets up a detailed meeting.

[0383] The present invention allows users to concentrate on important phone calls without being bothered by unnecessary calls, and is also effective as a means for avoiding nuisance calls and fraudulent calls.

[0384] The processing flow will be explained below.

[0385] Step 1:

[0386] Incoming calls and starting automatic answering

[0387] Server: When a call comes in, the server detects the call and activates an automated attendant, which greets the caller based on a pre-defined script (e.g., "Hello, thank you for calling. How can I help you?").

[0388] Step 2:

[0389] Call recording

[0390] Server: The automated answering system receives the caller's speech, and the server records the call in real time. The recorded content is converted into text data and stored in data storage.

[0391] Step 3:

[0392] Ending the call and preparing to send data

[0393] Server: When the call ends, prepares to send the recorded text data to the generation means. At this time, the process of converting the recorded file into text format is executed.

[0394] Step 4:

[0395] Generative data analysis and summary generation

[0396] Server: The recorded text data is sent to the AI ​​model, which analyzes the data, extracts key points, and generates a summary.

[0397] Example: A summary such as "Explanation regarding the proposal to introduce a new product. I would like to set a date for the next detailed meeting." is generated.

[0398] Step 5:

[0399] Convert summary content to SMS format and send

[0400] Server: Converts the generated summary into SMS format, then sends the SMS message to the user's device.

[0401] Step 6:

[0402] Receipt and confirmation by the user

[0403] User: The SMS arrives on the user's device and a summary is displayed. The user confirms the SMS content.

[0404] Example: A message is sent to the terminal saying, "Proposal for the introduction of a new product. I would like to set a date for the next detailed meeting."

[0405] Step 7:

[0406] Deciding and implementing a response

[0407] User: After reviewing the summary, the user decides what to do next based on the summary, and if necessary, calls back or takes other action.

[0408] Example: The user decides that the summary is important and calls back to schedule a meeting to discuss the details.

[0409] This system's series of processes allows users to focus on important phone calls without being bothered by unnecessary calls, and also makes it easier to avoid nuisance and fraudulent calls.

[0410] Example 1

[0411] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0412] In conventional telephone answering systems, users must answer the phone in person to check the content of incoming calls, which leads to problems such as wasted call time and increased burden from nuisance calls.In addition, there is a high risk that important call content will be overlooked, so an effective filtering method is necessary.

[0413] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0414] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the calls, a means for analyzing and summarizing the recorded contents of the calls, a means for notifying the user of the summarized contents of the calls to the user's terminal, a means for allowing the user to select and respond based on the summary, a means for saving the contents of the calls as text data, a means for sending the text data to an artificial intelligence model for analysis, and a means for summarizing the contents of the summaries in an easy-to-read format and sending them to the user's terminal. This allows the user to quickly respond to important calls without being bothered by unnecessary calls. It also makes it possible to avoid nuisance calls and fraudulent calls.

[0415] An "automatic answering means" is a means for automatically answering an incoming call and responding based on a pre-set script.

[0416] "Means for recording telephone conversations" refers to a means for recording telephone conversations as audio data and saving it.

[0417] The "generation means" is a means for analyzing the recorded content of a call and summarizing it using an artificial intelligence model.

[0418] "Means of notifying the user's device" refers to a means of sending a summary of the call contents to the user's mobile device or smartphone in the form of SMS or the like.

[0419] The "selection means" is a means for the user to check the received summary content and determine the need for action based on its importance.

[0420] "Means for saving as text data" refers to means for converting recorded voice data into text format and saving it.

[0421] An "artificial intelligence model" is a technology that uses machine learning techniques to analyze text data and perform summarization or other processing. Specific examples include generative AI models.

[0422] "Means for summarizing in an easy-to-read and concise manner" refers to a means for omitting the detailed parts of the generated summary and extracting only the main points to express them concisely.

[0423] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[0424] 1. Automated Response Methods

[0425] When a call comes in, the server activates an automatic answering means. The automatic answering means responds appropriately to the caller based on a pre-set script. For example, it might respond, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves it as audio data.

[0426] 2. Recording Method

[0427] The server stores the recorded conversation as audio data and also as text data, using voice recognition technology.

[0428] 3. Generation means

[0429] The text data of the call is sent by the server to an artificial intelligence model (e.g., OpenAI GPT-4), which analyzes and summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0430] 4. Means of notification

[0431] The server then sends the generated summary to the user's device as an SMS message. The summary is written in a concise and easy-to-read format. For example, the user's smartphone will receive a message such as, "Proposal for the introduction of a new product. We would like to set up a date for the next meeting to discuss the details."

[0432] 5. Selection methods

[0433] Users can check the contents of the SMS they receive and decide whether the call is important or not. If it is, they can call back and schedule a detailed meeting. If it is not important, they can ignore the call.

[0434] Hardware and software used

[0435] Server: Cloud computing service (e.g. AWS, Google Cloud)

[0436] Speech recognition technology: Technology that converts voice data into text data

[0437] Artificial intelligence models: Generative AI models such as OpenAI GPT-4

[0438] SMS sending service: Use a service such as Twilio

[0439] Specific examples

[0440] Consider the case where a businessman receives a sales call on his mobile phone. The server detects the incoming call and quickly activates an automated answering system, responding with, "Hello, thank you for calling. How can I help you?" The caller introduces the new product and requests a next meeting, and the conversation is recorded and saved as voice data by the server. The server then saves this voice data as text data using speech recognition technology, and then analyzes the text data using a generation system to summarize the conversation as, "An explanation of the proposed introduction of the new product. I would like to set a date for the next meeting to discuss the details." This summary is sent to the businessman's smartphone as an SMS. The businessman checks the contents of the SMS and, determining that it is important, calls back himself to set up a detailed meeting.

[0441] Prompt Sentence Examples

[0442] "Please tell me the procedure for the system to automatically answer incoming calls and analyze and summarize the contents of the calls."

[0443] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0444] Step 1:

[0445] When the server detects an incoming call, it activates the auto-answering means. Specifically, it receives an incoming call signal and executes the auto-answering program. The input is the incoming call signal, and the output is a trigger signal to start the response. Based on this trigger signal, the server executes a pre-set script and plays an auto-answering message such as "Hello, thank you for calling. How can I help you?"

[0446] Step 2:

[0447] The server records the contents of telephone conversations and saves them as audio data. From the moment a call begins, the server activates the recording function and captures all of the call content. The input is the audio signal during the call, and the output is the recorded audio data. This audio data is saved in a database after the call ends.

[0448] Step 3:

[0449] The server converts the stored voice data into text data using voice recognition technology. This process uses voice recognition software. The input is the recorded voice data, and the output is text data generated by the voice recognition software. The server stores this text data and prepares it for the next analysis step.

[0450] Step 4:

[0451] The server analyzes the text data using a generation method and generates a summary. Specifically, the text data is sent to a generative AI model (e.g., OpenAI GPT-4), which summarizes the content. The input is text data, and the output is summarized text data. For example, the content of a long sales call can be summarized as "An explanation of the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0452] Step 5:

[0453] The server sends the generated summary to the user's device as an SMS message. Specifically, it uses an SMS sending service (e.g., Twilio) to compose and send the message. The input is the summarized text data, and the output is an SMS message sent to the user's device. For example, a message such as "Proposal for the introduction of a new product. Please schedule the next meeting to discuss the details" is sent to the user's smartphone.

[0454] Step 6:

[0455] Users can check the contents of the SMS they receive and decide how to respond based on its importance. The input is the SMS message, and the output is the action the user takes (e.g., call back or ignore). If the user determines that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore the message.

[0456] (Application example 1)

[0457] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0458] In recent years, the number of nuisance and fraudulent phone calls has been increasing, causing users to be bothered by unnecessary calls and potentially miss important calls. In addition, businesses need to handle a large number of phone calls, creating a need for efficient call management. Furthermore, there is a need for a method to reduce the time and effort required to understand the content of calls that require specific responses. To solve these issues, there is a need for technology that can automatically analyze and summarize the content of phone calls and communicate it to users.

[0459] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0460] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the call, a means for analyzing the recorded contents of the call and generating a summary, a means for notifying the user of the summarized contents of the call to the user's terminal, and a notification means for providing the summarized contents via multiple notification means (e.g., SMS and email). This allows the user to quickly grasp important summarized information without having to listen to the entire contents of the call. Furthermore, the diversification of notification means allows the user to receive the summary in the most optimal way depending on their situation and priority.

[0461] An "automatic answering means" is a device or software that automatically answers an incoming call and records the content of the call.

[0462] "Means for recording telephone conversations" refers to a device or software for saving the contents of incoming telephone conversations as voice data or text data.

[0463] The "generator" is a device or software that analyzes the recorded conversation content, extracts important information, and generates a summary.

[0464] The "notification means" is a device or software for transmitting the generated summary to the user's terminal.

[0465] The "selection means" is a device or software that allows the user to determine the importance based on the summary received and select an appropriate response.

[0466] "Multiple notification methods" is a general term for methods of conveying information to users in different ways, such as SMS and email.

[0467] An "artificial intelligence model" is an algorithm or software used to analyze call content and generate summaries using machine learning and deep learning techniques.

[0468] This invention is a system that automatically answers an incoming call, analyzes and summarizes the contents of the call, and notifies the user. Next, we will explain how to specifically implement this system.

[0469] This system mainly uses a server, user terminals, and a network environment. The server is the main hardware for running a program that integrates speech recognition, text analysis, summary generation, and notification functions, and uses the following software to realize each function. For speech recognition, the "speech_recognition" Python library is used, and for summary generation, the "t5-small" model from Hugging Face, a generative AI model from "transformers," is used. Notification methods include sending SMS using the "Twilio" API and email using Python's "smtplib."

[0470] When the server detects an incoming call, it activates an automatic answering system and answers the call based on a pre-defined script. During this process, the incoming voice is recorded through a microphone and converted into text data using the "speech_recognition" library. The converted text data is then stored on the server.

[0471] The server then sends the stored text data to a generative AI model in the "transformers" library to generate a summary, which is a short piece of text containing important information.

[0472] This summary is sent to the user's device via SMS using the Twilio API or via email using smtplib. The user can then review the message and decide how to respond based on its importance.

[0473] Specific examples

[0474] For example, consider a scenario in which this system is implemented in a company's customer support. When a new customer support inquiry is received by phone, the server uses an automated response function to respond with, "Hello, this is Company Name Support. How can I help you?" The call is recorded and then converted into text using the "speech_recognition" library. A generative AI model in the "transformers" library then summarizes the call as, "There is a problem with Product A. Please contact the user for details." This summary is then sent to the representative's device via SMS and email.

[0475] Prompt Sentence Examples

[0476] Example prompt sentence:

[0477] Summarize the customer support call.

[0478] Voice input: "Hello, thank you for contacting us about an issue with Product A. Please tell us the specific problem you're experiencing and how it's progressing."

[0479] Voice output: "There is a problem with product A. Please check with the user for details."

[0480] In this way, the server automatically analyzes and summarizes the contents of the call and quickly conveys only the important information to the user, enabling efficient and effective telephone response.

[0481] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0482] Step 1:

[0483] When a call is received, the server activates an automatic answering means.

[0484] Input: Incoming phone signal.

[0485] Data processing and calculation: Detects incoming calls and loads pre-configured scripts to prepare an automatic response.

[0486] Output: Generates a response message based on the auto-response script.

[0487] Step 2:

[0488] The server answers the call using an automated answering means and records the contents of the call.

[0489] Input: Greeting message and call audio based on the auto-attendant script.

[0490] Data processing and calculation: Capture the contents of the call as audio data through the microphone and generate a recording file.

[0491] Output: A recording file of the call saved as audio data.

[0492] Step 3:

[0493] The server converts the recorded conversation into text data using a voice recognition library.

[0494] Input: Recording file (audio data).

[0495] Data processing and calculation: Analyzes voice data using the speech recognition library (speech_recognition) and generates text data.

[0496] Output: The call contents converted into text data.

[0497] Step 4:

[0498] The server sends the text data to a generative AI model to generate a summary.

[0499] Input: Text data.

[0500] Data processing and calculation: Summarize text data using a generative AI model (the t5-small model from the transformers library).

[0501] Output: Summarized text data.

[0502] Step 5:

[0503] The server sends the summarized content to the user's device via multiple notification methods (SMS and email).

[0504] Input: Abstracted text data.

[0505] Data processing and calculation: Using notification methods (Twilio API and smtplib), the summary data is sent to the user's terminal as SMS and email.

[0506] Output: Summary notification via SMS and email received on user device.

[0507] Step 6:

[0508] The user reviews the received summary and decides on the appropriate response, if necessary.

[0509] Input: Summary data sent via SMS or email.

[0510] Data processing and calculation: Determine importance and urgency based on summary data.

[0511] Output: The subsequent action resulting from the user's decision (e.g., calling back, replying by email, etc.).

[0512] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0513] This invention combines a system that uses an automatic answering means to respond to incoming calls, records and analyzes the content of the call, and an emotion engine that recognizes the user's emotions. This enables flexible responses that match the user's emotional state, resulting in more effective and personalized notifications.

[0514] System configuration

[0515] This system is configured as follows:

[0516] 1. Automated Response Methods

[0517] When a call comes in, the server activates an automated answering machine, which responds to the call based on a pre-defined script.

[0518] 2. Recording Method

[0519] The contents of the call are recorded in real time by the server and saved as text data.

[0520] 3. Generation means

[0521] The recorded text data is sent to a generation means, specifically an artificial intelligence model, where it is analyzed and summarized.

[0522] 4. Emotion Engine

[0523] The emotion engine analyzes emotions from the user's voice, facial expressions, and text input. For example, it analyzes facial expressions and voice when the user opens an SMS to determine the user's emotional state.

[0524] 5. Means of notification

[0525] Based on the analysis results of the emotion engine, the server sends a summary of the call to the user's device in a user-friendly format as an SMS. If the user's emotional state indicates stress, the server can summarize the content briefly or adjust the notification method.

[0526] 6. Selection methods

[0527] Users can check the contents of the SMS they receive and choose what to read based on their emotional state, for example, they can skip over messages they deem unimportant and check the details of messages they deem important.

[0528] Specific examples

[0529] Consider a scenario where a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic response means. The automatic response means responds with "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server uses the generation means to analyze the text data and summarize it as "An explanation of the proposed introduction of a new product. We would like to set a date for the next meeting to discuss the details."

[0530] The emotion engine analyzes data from the user's smartphone camera and microphone to determine the user's emotional state. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary of the message. The user can check the received SMS, check the details if it is important, or skip if it is not necessary.

[0531] Specific processing of the program

[0532] 1. Detecting incoming calls

[0533] When the server detects an incoming call, it activates an automatic answering means and starts a call. The automatic answering means answers the call according to a pre-set script.

[0534] 2. Recording of call content

[0535] The server records the call in real time, converts it into text data, and saves it.

[0536] 3. Summary Generation

[0537] A generating means is used to analyze the recorded text data and generate a summary, which is then converted into an SMS format.

[0538] 4. Sentiment Analysis and Notifications

[0539] The emotion engine analyzes the user's emotional state and adjusts the content and method of notifications based on the results. For example, if the user is feeling stressed, the summary content will be shortened or the notifications will be suppressed. The adjusted content will be sent to the user's device as an SMS.

[0540] 5. User Choice

[0541] The user can check the summary of the SMS they received and decide whether or not a response is necessary. If necessary, they can take action such as calling back.

[0542] This system allows users to quickly obtain only the important information without wasting time on unnecessary calls. Utilizing an emotion engine also reduces the user's mental burden and enables more personalized notification responses.

[0543] The processing flow will be explained below.

[0544] Step 1:

[0545] Incoming calls and starting automatic answering

[0546] Server: When a call comes in, the server detects the call and activates an automated response system, which responds to the caller based on a pre-defined script, saying, "Hello, thank you for calling. How can I help you?"

[0547] Step 2:

[0548] Call recording

[0549] Server: When the automated answering system receives the call, the server records the call in real time. The recorded content is converted into text data and stored in a database.

[0550] Step 3:

[0551] Ending the call and preparing to send data

[0552] Server: When the call ends, prepares the recorded text data to be sent to the generation means. During this data conversion process, the text data is formatted.

[0553] Step 4:

[0554] Generative data analysis and summary generation

[0555] Server: The recorded text data is sent to the AI ​​model, which analyzes the text data, extracts key points, and creates a summary.

[0556] Example: If the call content is "Pitching a new product and requesting a next meeting," the generated summary will be "Introducing a new product. Requesting a detailed meeting."

[0557] Step 5:

[0558] Analyzing user emotions with an emotion engine

[0559] Device: Uses the camera and microphone on the user's smartphone to capture the user's facial expressions and voice data.

[0560] Server: The acquired facial expression and voice data is sent to the emotion engine, which analyzes the user's emotional state (e.g., stress level, excitement level, relaxation level).

[0561] Step 6:

[0562] Tailoring notification content based on emotional state

[0563] Server: Based on the user's emotional state analyzed by the emotion engine, the notification method and content of the summarized call details are adjusted. For example, if the user is feeling stressed, the notification content is made shorter.

[0564] Step 7:

[0565] Convert summary content to SMS format and send

[0566] Server: Converts the generated summary into SMS format and sends the adjusted notification content to the user's device as an SMS.

[0567] Example: The summary "Introducing a new product. Requesting a meeting to discuss details" can be converted into an SMS message as "Important: Introduces a new product. Requesting a meeting to discuss details."

[0568] Step 8:

[0569] Receipt and confirmation by the user

[0570] User: An SMS is delivered to the user's device. The user checks the contents of the SMS.

[0571] Example: The device displays the message "Important: Introducing a new product, please arrange a meeting."

[0572] Step 9:

[0573] Deciding and implementing a response

[0574] User: Based on the summary of the SMS received, the user decides whether to take action and, if necessary, take action such as calling back to confirm the details.

[0575] Example: The user decides it's important, calls the sales rep back, and schedules a meeting.

[0576] This system's series of processes allows users to quickly obtain only the important information without being bothered by unnecessary calls. In addition, the use of an emotion engine reduces the user's mental burden and enables more personalized responses.

[0577] Example 2

[0578] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0579] While automated answering and recording of incoming phone calls are commonplace today, this alone is not sufficient for user convenience. In particular, without call content summarization and emotional analysis, users are often overwhelmed with unnecessary information, which can increase their mental burden. Furthermore, special consideration is required for users under stress, and flexible responses tailored to individual emotional states are required. Therefore, there is a need for a system that not only analyzes and summarizes call content, but also recognizes the user's emotions and adjusts notification content accordingly.

[0580] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0581] In this invention, the server includes means for recording the contents of a call when a call is received, means for analyzing the recorded contents of the call and generating a summary, means for analyzing the user's emotions, and means for adjusting the summarized contents of the call based on the user's emotional state and notifying the user's terminal. This makes it possible to quickly provide important information while taking the user's emotional state into consideration.

[0582] An "incoming call" refers to receiving an external call to a communication device.

[0583] "Automatic answering means" refers to a function that responds to incoming calls with a pre-set script or message without manual operation.

[0584] "Call content" refers to all audio information exchanged over the telephone.

[0585] "Recording means" refers to a device or software for storing the contents of a call.

[0586] "Generation means" refers to a processing device or software for analyzing the recorded content of a call and generating a summary.

[0587] "Emotion analysis means" refers to the function of analyzing the user's emotional state from audio and video data.

[0588] "Abridged call content" refers to information that briefly summarizes the content of a call.

[0589] "User's device" refers to a communication device used by a User that can receive System notifications.

[0590] "Notification means" refers to the function of delivering the generated summary content to the user's terminal.

[0591] "Selection tool" refers to a function that allows a user to select an appropriate response based on the summary content received.

[0592] An "artificial intelligence model" refers to an algorithm or system that learns from data and performs analysis and predictions.

[0593] This invention combines an emotion analysis system that recognizes the user's emotions with a system that uses an automatic answering means to respond to incoming calls and records and analyzes the content of the call, thereby enabling notifications to be delivered in line with the user's emotional state, enabling more effective and personalized information to be provided.

[0594] This system is configured as follows:

[0595] Hardware and Software Use

[0596] Server: VoIP system, Python script, Twilio API, Google Cloud Speech-to-Text API, OpenAI GPT-4, Azure Cognitive Services

[0597] User's device: Smartphone camera, microphone, SMS function

[0598] Specific processing

[0599] Incoming calls and auto-answering

[0600] The server detects an incoming call. When an incoming call is detected, it activates an automated response system that responds to the caller based on a pre-defined script. For example, it might send a message like this: "Hello, thank you for calling. How can I help you?"

[0601] Call recording

[0602] The server records the call in real time and converts the audio data into text data for storage. This process uses the Google Cloud Speech-to-Text API.

[0603] Generate a summary of the call

[0604] The server then uses a generation method to analyze the recorded text data and generate a summary. Using the OpenAI GPT-4 generative AI model, the server inputs the following prompt: "Please summarize the following call content: [Text data]."

[0605] Emotion analysis

[0606] The sentiment analysis method uses data obtained from a smartphone camera and microphone to analyze the user's emotional state. This data is analyzed through Microsoft Azure Cognitive Services to determine the user's emotional state.

[0607] Call summary notifications

[0608] Based on the results of the emotion analysis, the server sends a summary of the call to the user's device as an SMS. If the user's emotional state indicates stress, the summary will be made more concise or the notification method will be adjusted.

[0609] Specific examples

[0610] When a businessman receives a sales call on his mobile phone, the server detects the incoming call and activates the automatic response means. The automatic response means responds with, "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server then uses the generation means to summarize it as, "This is an explanation of the proposal to introduce a new product. I would like to set a date for the next meeting to discuss the details."

[0611] The emotion analysis function analyzes the user's emotions based on data collected from the smartphone camera and microphone. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary. The user can check the received SMS and, if the content is deemed important, review the details, or skip if it is not necessary.

[0612] This invention allows users to quickly obtain only the information they need without wasting time on unnecessary calls. In addition, emotion analysis reduces the user's mental burden and enables more personalized notification responses.

[0613] Prompt Sentence Examples

[0614] Please summarize the call below.

[0615] [Text data]

[0616] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0617] Step 1:

[0618] The server detects incoming calls. When a call comes in, it activates an automated answering system that responds to the caller based on a pre-defined script. For example, it might play a message like, "Hello, thank you for calling. How can I help you?"

[0619] Input: Incoming signal

[0620] Output: Start auto-reply, play greeting

[0621] What happens: The VoIP system detects an incoming call, and a Python script uses the Twilio API to play a pre-recorded message.

[0622] Step 2:

[0623] The server records the call in real time and converts the audio data into text data for storage, using the Google Cloud Speech-to-Text API.

[0624] Input: Call audio data

[0625] Output: Text data

[0626] How it works: The recorded audio data is saved in WAV format, and a Python script calls the Google Cloud Speech-to-Text API to convert the audio data into text data.

[0627] Step 3:

[0628] The server uses a generative method to analyze the text data and generate a summary. At this stage, a generative AI model (OpenAI GPT-4) is used.

[0629] Input: Text data

[0630] Output: Summary text

[0631] How it works: The Python script inputs the following prompt to the generative AI model: "Please summarize the following call content: [Text data]." The generative AI model outputs the summary text.

[0632] Step 4:

[0633] The server analyzes data obtained from the user's camera and microphone using emotion analysis tools, which are powered by Microsoft Azure Cognitive Services.

[0634] Input: Audio and video data from cameras and microphones

[0635] Output: Emotion analysis results

[0636] How it works: A smartphone app collects data from the camera and microphone and sends it to a server, which then sends it to Azure Cognitive Services for sentiment analysis.

[0637] Step 5:

[0638] The server adjusts the notification content based on the results of the sentiment analysis and sends a summary text to the user's device. The notification method uses the Twilio API.

[0639] Input: Summary text, sentiment analysis results

[0640] Output: SMS notification

[0641] Specific operation: The server compiles a concise summary text based on the sentiment analysis results and sends it as an SMS to the user's smartphone using the Twilio API.

[0642] Step 6:

[0643] Users can check the received SMS and respond as needed, for example, by checking the details if the message is important and calling back if necessary.

[0644] Input: SMS notification

[0645] Output: User action

[0646] Specific action: The user opens the SMS on their smartphone, checks the contents, and if necessary calls back or takes other action.

[0647] (Application example 2)

[0648] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0649] Current automated answering systems are effective in answering incoming calls and summarizing the content of calls, but they do not provide flexible responses that take into account the user's emotional state. This can leave users feeling stressed and dissatisfied, resulting in problems such as inefficiencies such as an increase in complaints and inquiries. Customer service in brick-and-mortar stores also needs to be able to respond in a way that takes into account the emotional state of customers, but this is difficult to achieve with conventional systems.

[0650] The specific processing 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 answering an incoming call using an automatic answering means and recording the content of the call, means for analyzing and summarizing the recorded content of the call, means for analyzing the emotional state of the user, and means for adjusting the content of the notification based on the analysis result. This enables flexible and personalized responses according to the emotional state of the user.

[0651] "When a call comes in" refers to a situation where an external connection is attempted over a telephone line and the user's device or system receives the connection.

[0652] "Automatic answering means" refers to a device or program that has the function of automatically answering an incoming telephone call without human intervention.

[0653] "Call recording means" refers to a device or program that automatically saves conversations and audio data that occur during a telephone call.

[0654] "Recorded call content" refers to information saved as audio or text data exchanged during a call.

[0655] "Analysis and summary generation means" refers to a device or program that analyzes the recorded content of a call, extracts only the important information, and provides it in a shortened form.

[0656] "Emotion analysis means for analyzing a user's emotional state" refers to technologies and algorithms for measuring or inferring emotions from a user's voice, facial expressions, text input, etc.

[0657] "Notification means for adjusting notification content" refers to a device or program for automatically changing the method or content of notifications based on the user's emotional state.

[0658] "Means for notifying the user of the summarized call content" refers to a device or program that transmits the information summarized by the generating means to the user's device such as a smartphone or PC.

[0659] "A means for the user to decide what to do based on the summary content" refers to a device or program that allows the user to look at the summary information notified to them and decide what to do next based on its importance and necessity.

[0660] This invention provides a system that uses an automatic answering means to answer incoming calls, records the contents of the call, analyzes the record to generate a summary, notifies the user of the summary, and further analyzes the user's emotional state to optimize the content of the notification. The following describes an example of application of this system to customer support functions in brick-and-mortar stores.

[0661] System Program

[0662] The server has a speech recognition engine, a text generation engine, an emotion analysis engine, a notification engine, and the ability to send emails using the SMTP protocol. It is recommended that the main software configuration uses the speech_recognition library, the TextBlob library, the pipeline function of the transformers library, and the smtplib library.

[0663] Explanation of program processing

[0664] 1. Incoming call detection and automatic answering

[0665] When a call comes in, the server's automatic answering means is activated, which responds to the call and gives an initial response to the customer based on a pre-set script.

[0666] 2. Recording of call content

[0667] The server uses a speech recognition engine (e.g., speech_recognition library) to record the call and save it as text data.

[0668] 3. Summary Generation

[0669] The recorded text data is analyzed by a text generation engine (e.g., the TextBlob library), which extracts key phrases and concepts and generates summaries based on them.

[0670] 4. Emotional state analysis

[0671] The summarized text and audio data are analyzed for the user's emotional state using a sentiment analysis engine (e.g., the pipeline function in the transformers library).

[0672] 5. Coordination and delivery of notifications

[0673] Based on the results of the sentiment analysis, the notification engine will tailor the notification content. For example, if it detects that the user is feeling stressed, a brief summary of the content will be sent. Notifications are implemented via the SMTP protocol (e.g., the smtplib library).

[0674] Specific examples

[0675] This technology is applied to customer service in brick-and-mortar stores. When a customer makes an inquiry, the in-store customer support robot automatically responds, records the content of the inquiry, and saves it as text data. The saved text data is analyzed in real time, and a summary is generated. At this time, the emotional state of the customer is also analyzed along with the analyzed summary data. For example, if a customer expresses anger, this information is notified to staff, and a flexible response is required.

[0676] Prompt Sentence Examples

[0677] "Analyze the following speech input and output a result about the emotional state: 'Hello, customer. Can you tell me more about this weekend's sale? I'm happy.'"

[0678] In this way, the system can provide a more personalized customer experience and improve the user experience.

[0679] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0680] Step 1:

[0681] When a call comes in, the server detects the call. It receives the incoming call signal and activates the automatic answering means. The automatic answering means answers the call and starts the call based on a pre-set script. The input is the incoming call signal and the output is the automatic answering message.

[0682] Step 2:

[0683] The server uses a speech recognition engine (e.g., the speech_recognition library) to record the call contents. The input is the voice data during the call, and the voice data is converted into text data in real time and saved. The output is the converted text data.

[0684] Step 3:

[0685] The server uses a generator (e.g., TextBlob library) to analyze the recorded text data, extract important key phrases, and generate a summary. The text data is analyzed and key phrases and key content parts are extracted. The input is the text data of the call, and the output is the summarized text data.

[0686] Step 4:

[0687] The server uses sentiment analysis tools (e.g., the pipeline function in the transformers library) to analyze the user's emotional state from the summarized text and audio data. It analyzes the audio and text data and determines the emotional state (e.g., positive, negative, neutral). The input is the summarized text and audio data, and the output is the sentiment analysis result.

[0688] Step 5:

[0689] The server uses a notification mechanism to adjust the notification content based on the emotion analysis results. For example, if the user is feeling stressed, the server adjusts the notification content by making the summary more concise. The input is the emotion analysis results and the summarized text data, and the output is the adjusted notification content.

[0690] Step 6:

[0691] The server sends the tailored notification content to the user's device. The notification is sent in the form of email or other format using the SMTP protocol. A notification engine (e.g., the smtplib library) is used to send emails with detailed information to staff or other interested parties. The input is the tailored notification content, and the output is the notification sent to the user's device.

[0692] Step 7:

[0693] The user checks the received notification and decides how to respond if necessary. Based on the content of the notification, the next action (e.g., direct response by staff, request for additional information) is taken. The input is the notification sent to the user's device, and the output is the user's decision on action.

[0694] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0695] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0696] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0697] [Third embodiment]

[0698] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0699] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0700] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0701] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0702] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0703] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0704] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0705] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0706] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0707] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0708] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0709] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0710] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[0711] 1. Automated Response Methods

[0712] When a call comes in, the server activates the automated response means, which responds appropriately to the caller based on a pre-set script. For example, the automated response means might say, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves them as text data.

[0713] 2. Generation means

[0714] The recorded content of the call is analyzed by the generation means. The server sends the recorded text data to an AI model, which summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0715] 3. Means of notification

[0716] Once the summary is generated, the server sends it to the user's device as an SMS. At this time, the summary is conveyed to the user as a short message. Specifically, a message such as "Proposal for the introduction of a new product. Please schedule a date for the next meeting to discuss the details" will be sent to the user's smartphone.

[0717] 4. Selection methods

[0718] Users can check the contents of the SMS they receive and decide whether or not a response is necessary. For example, if they determine that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore it.

[0719] Specific examples

[0720] As an example, consider a scenario in which a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic answering means, responding with "Hello, thank you for calling. How can I help you?" The caller then introduces the new product and requests for a next meeting, and the content is recorded and saved as text data. The server then uses a generation means to analyze the text data and summarizes the content of the call as "Explanation regarding the proposed introduction of the new product. We would like to set a date for the next meeting to discuss the details." This summary is then sent to the user's smartphone as an SMS. The user checks the content of the SMS and, determining that it is important, calls back and sets up a detailed meeting.

[0721] The present invention allows users to concentrate on important phone calls without being bothered by unnecessary calls, and is also effective as a means for avoiding nuisance calls and fraudulent calls.

[0722] The processing flow will be explained below.

[0723] Step 1:

[0724] Incoming calls and starting automatic answering

[0725] Server: When a call comes in, the server detects the call and activates an automated attendant, which greets the caller based on a pre-defined script (e.g., "Hello, thank you for calling. How can I help you?").

[0726] Step 2:

[0727] Call recording

[0728] Server: The automated answering system receives the caller's speech, and the server records the call in real time. The recorded content is converted into text data and stored in data storage.

[0729] Step 3:

[0730] Ending the call and preparing to send data

[0731] Server: When the call ends, prepares to send the recorded text data to the generation means. At this time, the process of converting the recorded file into text format is executed.

[0732] Step 4:

[0733] Generative data analysis and summary generation

[0734] Server: The recorded text data is sent to the AI ​​model, which analyzes the data, extracts key points, and generates a summary.

[0735] Example: A summary such as "Explanation regarding the proposal to introduce a new product. I would like to set a date for the next detailed meeting." is generated.

[0736] Step 5:

[0737] Convert summary content to SMS format and send

[0738] Server: Converts the generated summary into SMS format, then sends the SMS message to the user's device.

[0739] Step 6:

[0740] Receipt and confirmation by the user

[0741] User: The SMS arrives on the user's device and a summary is displayed. The user confirms the SMS content.

[0742] Example: A message is sent to the terminal saying, "Proposal for the introduction of a new product. I would like to set a date for the next detailed meeting."

[0743] Step 7:

[0744] Deciding and implementing a response

[0745] User: After reviewing the summary, the user decides what to do next based on the summary, and if necessary, calls back or takes other action.

[0746] Example: The user decides that the summary is important and calls back to schedule a meeting to discuss the details.

[0747] This system's series of processes allows users to focus on important phone calls without being bothered by unnecessary calls, and also makes it easier to avoid nuisance and fraudulent calls.

[0748] Example 1

[0749] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0750] In conventional telephone answering systems, users must answer the phone in person to check the content of incoming calls, which leads to problems such as wasted call time and increased burden from nuisance calls.In addition, there is a high risk that important call content will be overlooked, so an effective filtering method is necessary.

[0751] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0752] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the calls, a means for analyzing and summarizing the recorded contents of the calls, a means for notifying the user of the summarized contents of the calls to the user's terminal, a means for allowing the user to select and respond based on the summary, a means for saving the contents of the calls as text data, a means for sending the text data to an artificial intelligence model for analysis, and a means for summarizing the contents of the summaries in an easy-to-read format and sending them to the user's terminal. This allows the user to quickly respond to important calls without being bothered by unnecessary calls. It also makes it possible to avoid nuisance calls and fraudulent calls.

[0753] An "automatic answering means" is a means for automatically answering an incoming call and responding based on a pre-set script.

[0754] "Means for recording telephone conversations" refers to a means for recording telephone conversations as audio data and saving it.

[0755] The "generation means" is a means for analyzing the recorded content of a call and summarizing it using an artificial intelligence model.

[0756] "Means of notifying the user's device" refers to a means of sending a summary of the call contents to the user's mobile device or smartphone in the form of SMS or the like.

[0757] The "selection means" is a means for the user to check the received summary content and determine the need for action based on its importance.

[0758] "Means for saving as text data" refers to means for converting recorded voice data into text format and saving it.

[0759] An "artificial intelligence model" is a technology that uses machine learning techniques to analyze text data and perform summarization or other processing. Specific examples include generative AI models.

[0760] "Means for summarizing in an easy-to-read and concise manner" refers to a means for omitting the detailed parts of the generated summary and extracting only the main points to express them concisely.

[0761] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[0762] 1. Automated Response Methods

[0763] When a call comes in, the server activates an automatic answering means. The automatic answering means responds appropriately to the caller based on a pre-set script. For example, it might respond, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves it as audio data.

[0764] 2. Recording Method

[0765] The server stores the recorded conversation as audio data and also as text data, using voice recognition technology.

[0766] 3. Generation means

[0767] The text data of the call is sent by the server to an artificial intelligence model (e.g., OpenAI GPT-4), which analyzes and summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0768] 4. Means of notification

[0769] The server then sends the generated summary to the user's device as an SMS message. The summary is written in a concise and easy-to-read format. For example, the user's smartphone will receive a message such as, "Proposal for the introduction of a new product. We would like to set up a date for the next meeting to discuss the details."

[0770] 5. Selection methods

[0771] Users can check the contents of the SMS they receive and decide whether the call is important or not. If it is, they can call back and schedule a detailed meeting. If it is not important, they can ignore the call.

[0772] Hardware and software used

[0773] Server: Cloud computing service (e.g. AWS, Google Cloud)

[0774] Speech recognition technology: Technology that converts voice data into text data

[0775] Artificial intelligence models: Generative AI models such as OpenAI GPT-4

[0776] SMS sending service: Use a service such as Twilio

[0777] Specific examples

[0778] Consider the case where a businessman receives a sales call on his mobile phone. The server detects the incoming call and quickly activates an automated answering system, responding with, "Hello, thank you for calling. How can I help you?" The caller introduces the new product and requests a next meeting, and the conversation is recorded and saved as voice data by the server. The server then saves this voice data as text data using speech recognition technology, and then analyzes the text data using a generation system to summarize the conversation as, "An explanation of the proposed introduction of the new product. I would like to set a date for the next meeting to discuss the details." This summary is sent to the businessman's smartphone as an SMS. The businessman checks the contents of the SMS and, determining that it is important, calls back himself to set up a detailed meeting.

[0779] Prompt Sentence Examples

[0780] "Please tell me the procedure for the system to automatically answer incoming calls and analyze and summarize the contents of the calls."

[0781] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0782] Step 1:

[0783] When the server detects an incoming call, it activates the auto-answering means. Specifically, it receives an incoming call signal and executes the auto-answering program. The input is the incoming call signal, and the output is a trigger signal to start the response. Based on this trigger signal, the server executes a pre-set script and plays an auto-answering message such as "Hello, thank you for calling. How can I help you?"

[0784] Step 2:

[0785] The server records the contents of telephone conversations and saves them as audio data. From the moment a call begins, the server activates the recording function and captures all of the call content. The input is the audio signal during the call, and the output is the recorded audio data. This audio data is saved in a database after the call ends.

[0786] Step 3:

[0787] The server converts the stored voice data into text data using voice recognition technology. This process uses voice recognition software. The input is the recorded voice data, and the output is text data generated by the voice recognition software. The server stores this text data and prepares it for the next analysis step.

[0788] Step 4:

[0789] The server analyzes the text data using a generation method and generates a summary. Specifically, the text data is sent to a generative AI model (e.g., OpenAI GPT-4), which summarizes the content. The input is text data, and the output is summarized text data. For example, the content of a long sales call can be summarized as "An explanation of the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[0790] Step 5:

[0791] The server sends the generated summary to the user's device as an SMS message. Specifically, it uses an SMS sending service (e.g., Twilio) to compose and send the message. The input is the summarized text data, and the output is an SMS message sent to the user's device. For example, a message such as "Proposal for the introduction of a new product. Please schedule the next meeting to discuss the details" is sent to the user's smartphone.

[0792] Step 6:

[0793] Users can check the contents of the SMS they receive and decide how to respond based on its importance. The input is the SMS message, and the output is the action the user takes (e.g., call back or ignore). If the user determines that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore the message.

[0794] (Application example 1)

[0795] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0796] In recent years, the number of nuisance and fraudulent phone calls has been increasing, causing users to be bothered by unnecessary calls and potentially miss important calls. In addition, businesses need to handle a large number of phone calls, creating a need for efficient call management. Furthermore, there is a need for a method to reduce the time and effort required to understand the content of calls that require specific responses. To solve these issues, there is a need for technology that can automatically analyze and summarize the content of phone calls and communicate it to users.

[0797] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0798] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the call, a means for analyzing the recorded contents of the call and generating a summary, a means for notifying the user of the summarized contents of the call to the user's terminal, and a notification means for providing the summarized contents via multiple notification means (e.g., SMS and email). This allows the user to quickly grasp important summarized information without having to listen to the entire contents of the call. Furthermore, the diversification of notification means allows the user to receive the summary in the most optimal way depending on their situation and priority.

[0799] An "automatic answering means" is a device or software that automatically answers an incoming call and records the content of the call.

[0800] "Means for recording telephone conversations" refers to a device or software for saving the contents of incoming telephone conversations as voice data or text data.

[0801] The "generator" is a device or software that analyzes the recorded conversation content, extracts important information, and generates a summary.

[0802] The "notification means" is a device or software for transmitting the generated summary to the user's terminal.

[0803] The "selection means" is a device or software that allows the user to determine the importance based on the summary received and select an appropriate response.

[0804] "Multiple notification methods" is a general term for methods of conveying information to users in different ways, such as SMS and email.

[0805] An "artificial intelligence model" is an algorithm or software used to analyze call content and generate summaries using machine learning and deep learning techniques.

[0806] This invention is a system that automatically answers an incoming call, analyzes and summarizes the contents of the call, and notifies the user. Next, we will explain how to specifically implement this system.

[0807] This system mainly uses a server, user terminals, and a network environment. The server is the main hardware for running a program that integrates speech recognition, text analysis, summary generation, and notification functions, and uses the following software to realize each function. For speech recognition, the "speech_recognition" Python library is used, and for summary generation, the "t5-small" model from Hugging Face, a generative AI model from "transformers," is used. Notification methods include sending SMS using the "Twilio" API and email using Python's "smtplib."

[0808] When the server detects an incoming call, it activates an automatic answering system and answers the call based on a pre-defined script. During this process, the incoming voice is recorded through a microphone and converted into text data using the "speech_recognition" library. The converted text data is then stored on the server.

[0809] The server then sends the stored text data to a generative AI model in the "transformers" library to generate a summary, which is a short piece of text containing important information.

[0810] This summary is sent to the user's device via SMS using the Twilio API or via email using smtplib. The user can then review the message and decide how to respond based on its importance.

[0811] Specific examples

[0812] For example, consider a scenario in which this system is implemented in a company's customer support. When a new customer support inquiry is received by phone, the server uses an automated response function to respond with, "Hello, this is Company Name Support. How can I help you?" The call is recorded and then converted into text using the "speech_recognition" library. A generative AI model in the "transformers" library then summarizes the call as, "There is a problem with Product A. Please contact the user for details." This summary is then sent to the representative's device via SMS and email.

[0813] Prompt Sentence Examples

[0814] Example prompt sentence:

[0815] Summarize the customer support call.

[0816] Voice input: "Hello, thank you for contacting us about an issue with Product A. Please tell us the specific problem you're experiencing and how it's progressing."

[0817] Voice output: "There is a problem with product A. Please check with the user for details."

[0818] In this way, the server automatically analyzes and summarizes the contents of the call and quickly conveys only the important information to the user, enabling efficient and effective telephone response.

[0819] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0820] Step 1:

[0821] When a call is received, the server activates an automatic answering means.

[0822] Input: Incoming phone signal.

[0823] Data processing and calculation: Detects incoming calls and loads pre-configured scripts to prepare an automatic response.

[0824] Output: Generates a response message based on the auto-response script.

[0825] Step 2:

[0826] The server answers the call using an automated answering means and records the contents of the call.

[0827] Input: Greeting message and call audio based on the auto-attendant script.

[0828] Data processing and calculation: Capture the contents of the call as audio data through the microphone and generate a recording file.

[0829] Output: A recording file of the call saved as audio data.

[0830] Step 3:

[0831] The server converts the recorded conversation into text data using a voice recognition library.

[0832] Input: Recording file (audio data).

[0833] Data processing and calculation: Analyzes voice data using the speech recognition library (speech_recognition) and generates text data.

[0834] Output: The call contents converted into text data.

[0835] Step 4:

[0836] The server sends the text data to a generative AI model to generate a summary.

[0837] Input: Text data.

[0838] Data processing and calculation: Summarize text data using a generative AI model (the t5-small model from the transformers library).

[0839] Output: Summarized text data.

[0840] Step 5:

[0841] The server sends the summarized content to the user's device via multiple notification methods (SMS and email).

[0842] Input: Abstracted text data.

[0843] Data processing and calculation: Using notification methods (Twilio API and smtplib), the summary data is sent to the user's terminal as SMS and email.

[0844] Output: Summary notification via SMS and email received on user device.

[0845] Step 6:

[0846] The user reviews the received summary and decides on the appropriate response, if necessary.

[0847] Input: Summary data sent via SMS or email.

[0848] Data processing and calculation: Determine importance and urgency based on summary data.

[0849] Output: The subsequent action resulting from the user's decision (e.g., calling back, replying by email, etc.).

[0850] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0851] This invention combines a system that uses an automatic answering means to respond to incoming calls, records and analyzes the content of the call, and an emotion engine that recognizes the user's emotions. This enables flexible responses that match the user's emotional state, resulting in more effective and personalized notifications.

[0852] System configuration

[0853] This system is configured as follows:

[0854] 1. Automated Response Methods

[0855] When a call comes in, the server activates an automated answering machine, which responds to the call based on a pre-defined script.

[0856] 2. Recording Method

[0857] The contents of the call are recorded in real time by the server and saved as text data.

[0858] 3. Generation means

[0859] The recorded text data is sent to a generation means, specifically an artificial intelligence model, where it is analyzed and summarized.

[0860] 4. Emotion Engine

[0861] The emotion engine analyzes emotions from the user's voice, facial expressions, and text input. For example, it analyzes facial expressions and voice when the user opens an SMS to determine the user's emotional state.

[0862] 5. Means of notification

[0863] Based on the analysis results of the emotion engine, the server sends a summary of the call to the user's device in a user-friendly format as an SMS. If the user's emotional state indicates stress, the server can summarize the content briefly or adjust the notification method.

[0864] 6. Selection methods

[0865] Users can check the contents of the SMS they receive and choose what to read based on their emotional state, for example, they can skip over messages they deem unimportant and check the details of messages they deem important.

[0866] Specific examples

[0867] Consider a scenario where a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic response means. The automatic response means responds with "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server uses the generation means to analyze the text data and summarize it as "An explanation of the proposed introduction of a new product. We would like to set a date for the next meeting to discuss the details."

[0868] The emotion engine analyzes data from the user's smartphone camera and microphone to determine the user's emotional state. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary of the message. The user can check the received SMS, check the details if it is important, or skip if it is not necessary.

[0869] Specific processing of the program

[0870] 1. Detecting incoming calls

[0871] When the server detects an incoming call, it activates an automatic answering means and starts a call. The automatic answering means answers the call according to a pre-set script.

[0872] 2. Recording of call content

[0873] The server records the call in real time, converts it into text data, and saves it.

[0874] 3. Summary Generation

[0875] A generating means is used to analyze the recorded text data and generate a summary, which is then converted into an SMS format.

[0876] 4. Sentiment Analysis and Notifications

[0877] The emotion engine analyzes the user's emotional state and adjusts the content and method of notifications based on the results. For example, if the user is feeling stressed, the summary content will be shortened or the notifications will be suppressed. The adjusted content will be sent to the user's device as an SMS.

[0878] 5. User Choice

[0879] The user can check the summary of the SMS they received and decide whether or not a response is necessary. If necessary, they can take action such as calling back.

[0880] This system allows users to quickly obtain only the important information without wasting time on unnecessary calls. Utilizing an emotion engine also reduces the user's mental burden and enables more personalized notification responses.

[0881] The processing flow will be explained below.

[0882] Step 1:

[0883] Incoming calls and starting automatic answering

[0884] Server: When a call comes in, the server detects the call and activates an automated response system, which responds to the caller based on a pre-defined script, saying, "Hello, thank you for calling. How can I help you?"

[0885] Step 2:

[0886] Call recording

[0887] Server: When the automated answering system receives the call, the server records the call in real time. The recorded content is converted into text data and stored in a database.

[0888] Step 3:

[0889] Ending the call and preparing to send data

[0890] Server: When the call ends, prepares the recorded text data to be sent to the generation means. During this data conversion process, the text data is formatted.

[0891] Step 4:

[0892] Generative data analysis and summary generation

[0893] Server: The recorded text data is sent to the AI ​​model, which analyzes the text data, extracts key points, and creates a summary.

[0894] Example: If the call content is "Pitching a new product and requesting a next meeting," the generated summary will be "Introducing a new product. Requesting a detailed meeting."

[0895] Step 5:

[0896] Analyzing user emotions with an emotion engine

[0897] Device: Uses the camera and microphone on the user's smartphone to capture the user's facial expressions and voice data.

[0898] Server: The acquired facial expression and voice data is sent to the emotion engine, which analyzes the user's emotional state (e.g., stress level, excitement level, relaxation level).

[0899] Step 6:

[0900] Tailoring notification content based on emotional state

[0901] Server: Based on the user's emotional state analyzed by the emotion engine, the notification method and content of the summarized call details are adjusted. For example, if the user is feeling stressed, the notification content is made shorter.

[0902] Step 7:

[0903] Convert summary content to SMS format and send

[0904] Server: Converts the generated summary into SMS format and sends the adjusted notification content to the user's device as an SMS.

[0905] Example: The summary "Introducing a new product. Requesting a meeting to discuss details" can be converted into an SMS message as "Important: Introduces a new product. Requesting a meeting to discuss details."

[0906] Step 8:

[0907] Receipt and confirmation by the user

[0908] User: An SMS is delivered to the user's device. The user checks the contents of the SMS.

[0909] Example: The device displays the message "Important: Introducing a new product, please arrange a meeting."

[0910] Step 9:

[0911] Deciding and implementing a response

[0912] User: Based on the summary of the SMS received, the user decides whether to take action and, if necessary, take action such as calling back to confirm the details.

[0913] Example: The user decides it's important, calls the sales rep back, and schedules a meeting.

[0914] This system's series of processes allows users to quickly obtain only the important information without being bothered by unnecessary calls. In addition, the use of an emotion engine reduces the user's mental burden and enables more personalized responses.

[0915] Example 2

[0916] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0917] While automated answering and recording of incoming phone calls are commonplace today, this alone is not sufficient for user convenience. In particular, without call content summarization and emotional analysis, users are often overwhelmed with unnecessary information, which can increase their mental burden. Furthermore, special consideration is required for users under stress, and flexible responses tailored to individual emotional states are required. Therefore, there is a need for a system that not only analyzes and summarizes call content, but also recognizes the user's emotions and adjusts notification content accordingly.

[0918] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0919] In this invention, the server includes means for recording the contents of a call when a call is received, means for analyzing the recorded contents of the call and generating a summary, means for analyzing the user's emotions, and means for adjusting the summarized contents of the call based on the user's emotional state and notifying the user's terminal. This makes it possible to quickly provide important information while taking the user's emotional state into consideration.

[0920] An "incoming call" refers to receiving an external call to a communication device.

[0921] "Automatic answering means" refers to a function that responds to incoming calls with a pre-set script or message without manual operation.

[0922] "Call content" refers to all audio information exchanged over the telephone.

[0923] "Recording means" refers to a device or software for storing the contents of a call.

[0924] "Generation means" refers to a processing device or software for analyzing the recorded content of a call and generating a summary.

[0925] "Emotion analysis means" refers to the function of analyzing the user's emotional state from audio and video data.

[0926] "Abridged call content" refers to information that briefly summarizes the content of a call.

[0927] "User's device" refers to a communication device used by a User that can receive System notifications.

[0928] "Notification means" refers to the function of delivering the generated summary content to the user's terminal.

[0929] "Selection tool" refers to a function that allows a user to select an appropriate response based on the summary content received.

[0930] An "artificial intelligence model" refers to an algorithm or system that learns from data and performs analysis and predictions.

[0931] This invention combines an emotion analysis system that recognizes the user's emotions with a system that uses an automatic answering means to respond to incoming calls and records and analyzes the content of the call, thereby enabling notifications to be delivered in line with the user's emotional state, enabling more effective and personalized information to be provided.

[0932] This system is configured as follows:

[0933] Hardware and Software Use

[0934] Server: VoIP system, Python script, Twilio API, Google Cloud Speech-to-Text API, OpenAI GPT-4, Azure Cognitive Services

[0935] User's device: Smartphone camera, microphone, SMS function

[0936] Specific processing

[0937] Incoming calls and auto-answering

[0938] The server detects an incoming call. When an incoming call is detected, it activates an automated response system that responds to the caller based on a pre-defined script. For example, it might send a message like this: "Hello, thank you for calling. How can I help you?"

[0939] Call recording

[0940] The server records the call in real time and converts the audio data into text data for storage. This process uses the Google Cloud Speech-to-Text API.

[0941] Generate a summary of the call

[0942] The server then uses a generation method to analyze the recorded text data and generate a summary. Using the OpenAI GPT-4 generative AI model, the server inputs the following prompt: "Please summarize the following call content: [Text data]."

[0943] Emotion analysis

[0944] The sentiment analysis method uses data obtained from a smartphone camera and microphone to analyze the user's emotional state. This data is analyzed through Microsoft Azure Cognitive Services to determine the user's emotional state.

[0945] Call summary notifications

[0946] Based on the results of the emotion analysis, the server sends a summary of the call to the user's device as an SMS. If the user's emotional state indicates stress, the summary will be made more concise or the notification method will be adjusted.

[0947] Specific examples

[0948] When a businessman receives a sales call on his mobile phone, the server detects the incoming call and activates the automatic response means. The automatic response means responds with, "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server then uses the generation means to summarize it as, "This is an explanation of the proposal to introduce a new product. I would like to set a date for the next meeting to discuss the details."

[0949] The emotion analysis function analyzes the user's emotions based on data collected from the smartphone camera and microphone. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary. The user can check the received SMS and, if the content is deemed important, review the details, or skip if it is not necessary.

[0950] This invention allows users to quickly obtain only the information they need without wasting time on unnecessary calls. In addition, emotion analysis reduces the user's mental burden and enables more personalized notification responses.

[0951] Prompt Sentence Examples

[0952] Please summarize the call below.

[0953] [Text data]

[0954] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0955] Step 1:

[0956] The server detects incoming calls. When a call comes in, it activates an automated answering system that responds to the caller based on a pre-defined script. For example, it might play a message like, "Hello, thank you for calling. How can I help you?"

[0957] Input: Incoming signal

[0958] Output: Start auto-reply, play greeting

[0959] What happens: The VoIP system detects an incoming call, and a Python script uses the Twilio API to play a pre-recorded message.

[0960] Step 2:

[0961] The server records the call in real time and converts the audio data into text data for storage, using the Google Cloud Speech-to-Text API.

[0962] Input: Call audio data

[0963] Output: Text data

[0964] How it works: The recorded audio data is saved in WAV format, and a Python script calls the Google Cloud Speech-to-Text API to convert the audio data into text data.

[0965] Step 3:

[0966] The server uses a generative method to analyze the text data and generate a summary. At this stage, a generative AI model (OpenAI GPT-4) is used.

[0967] Input: Text data

[0968] Output: Summary text

[0969] How it works: The Python script inputs the following prompt to the generative AI model: "Please summarize the following call content: [Text data]." The generative AI model outputs the summary text.

[0970] Step 4:

[0971] The server analyzes data obtained from the user's camera and microphone using emotion analysis tools, which are powered by Microsoft Azure Cognitive Services.

[0972] Input: Audio and video data from cameras and microphones

[0973] Output: Emotion analysis results

[0974] How it works: A smartphone app collects data from the camera and microphone and sends it to a server, which then sends it to Azure Cognitive Services for sentiment analysis.

[0975] Step 5:

[0976] The server adjusts the notification content based on the results of the sentiment analysis and sends a summary text to the user's device. The notification method uses the Twilio API.

[0977] Input: Summary text, sentiment analysis results

[0978] Output: SMS notification

[0979] Specific operation: The server compiles a concise summary text based on the sentiment analysis results and sends it as an SMS to the user's smartphone using the Twilio API.

[0980] Step 6:

[0981] Users can check the received SMS and respond as needed, for example, by checking the details if the message is important and calling back if necessary.

[0982] Input: SMS notification

[0983] Output: User action

[0984] Specific action: The user opens the SMS on their smartphone, checks the contents, and if necessary calls back or takes other action.

[0985] (Application example 2)

[0986] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0987] Current automated answering systems are effective in answering incoming calls and summarizing the content of calls, but they do not provide flexible responses that take into account the user's emotional state. This can leave users feeling stressed and dissatisfied, resulting in problems such as inefficiencies such as an increase in complaints and inquiries. Customer service in brick-and-mortar stores also needs to be able to respond in a way that takes into account the emotional state of customers, but this is difficult to achieve with conventional systems.

[0988] The specific processing 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 answering an incoming call using an automatic answering means and recording the content of the call, means for analyzing and summarizing the recorded content of the call, means for analyzing the emotional state of the user, and means for adjusting the content of the notification based on the analysis result. This enables flexible and personalized responses according to the emotional state of the user.

[0989] "When a call comes in" refers to a situation where an external connection is attempted over a telephone line and the user's device or system receives the connection.

[0990] "Automatic answering means" refers to a device or program that has the function of automatically answering an incoming telephone call without human intervention.

[0991] "Call recording means" refers to a device or program that automatically saves conversations and audio data that occur during a telephone call.

[0992] "Recorded call content" refers to information saved as audio or text data exchanged during a call.

[0993] "Analysis and summary generation means" refers to a device or program that analyzes the recorded content of a call, extracts only the important information, and provides it in a shortened form.

[0994] "Emotion analysis means for analyzing a user's emotional state" refers to technologies and algorithms for measuring or inferring emotions from a user's voice, facial expressions, text input, etc.

[0995] "Notification means for adjusting notification content" refers to a device or program for automatically changing the method or content of notifications based on the user's emotional state.

[0996] "Means for notifying the user of the summarized call content" refers to a device or program that transmits the information summarized by the generating means to the user's device such as a smartphone or PC.

[0997] "A means for the user to decide what to do based on the summary content" refers to a device or program that allows the user to look at the summary information notified to them and decide what to do next based on its importance and necessity.

[0998] This invention provides a system that uses an automatic answering means to answer incoming calls, records the contents of the call, analyzes the record to generate a summary, notifies the user of the summary, and further analyzes the user's emotional state to optimize the content of the notification. The following describes an example of application of this system to customer support functions in brick-and-mortar stores.

[0999] System Program

[1000] The server has a speech recognition engine, a text generation engine, an emotion analysis engine, a notification engine, and the ability to send emails using the SMTP protocol. It is recommended that the main software configuration uses the speech_recognition library, the TextBlob library, the pipeline function of the transformers library, and the smtplib library.

[1001] Explanation of program processing

[1002] 1. Incoming call detection and automatic answering

[1003] When a call comes in, the server's automatic answering means is activated, which responds to the call and gives an initial response to the customer based on a pre-set script.

[1004] 2. Recording of call content

[1005] The server uses a speech recognition engine (e.g., speech_recognition library) to record the call and save it as text data.

[1006] 3. Summary Generation

[1007] The recorded text data is analyzed by a text generation engine (e.g., the TextBlob library), which extracts key phrases and concepts and generates summaries based on them.

[1008] 4. Emotional state analysis

[1009] The summarized text and audio data are analyzed for the user's emotional state using a sentiment analysis engine (e.g., the pipeline function in the transformers library).

[1010] 5. Coordination and delivery of notifications

[1011] Based on the results of the sentiment analysis, the notification engine will tailor the notification content. For example, if it detects that the user is feeling stressed, a brief summary of the content will be sent. Notifications are implemented via the SMTP protocol (e.g., the smtplib library).

[1012] Specific examples

[1013] This technology is applied to customer service in brick-and-mortar stores. When a customer makes an inquiry, the in-store customer support robot automatically responds, records the content of the inquiry, and saves it as text data. The saved text data is analyzed in real time, and a summary is generated. At this time, the emotional state of the customer is also analyzed along with the analyzed summary data. For example, if a customer expresses anger, this information is notified to staff, and a flexible response is required.

[1014] Prompt Sentence Examples

[1015] "Analyze the following speech input and output a result about the emotional state: 'Hello, customer. Can you tell me more about this weekend's sale? I'm happy.'"

[1016] In this way, the system can provide a more personalized customer experience and improve the user experience.

[1017] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1018] Step 1:

[1019] When a call comes in, the server detects the call. It receives the incoming call signal and activates the automatic answering means. The automatic answering means answers the call and starts the call based on a pre-set script. The input is the incoming call signal and the output is the automatic answering message.

[1020] Step 2:

[1021] The server uses a speech recognition engine (e.g., the speech_recognition library) to record the call contents. The input is the voice data during the call, and the voice data is converted into text data in real time and saved. The output is the converted text data.

[1022] Step 3:

[1023] The server uses a generator (e.g., TextBlob library) to analyze the recorded text data, extract important key phrases, and generate a summary. The text data is analyzed and key phrases and key content parts are extracted. The input is the text data of the call, and the output is the summarized text data.

[1024] Step 4:

[1025] The server uses sentiment analysis tools (e.g., the pipeline function in the transformers library) to analyze the user's emotional state from the summarized text and audio data. It analyzes the audio and text data and determines the emotional state (e.g., positive, negative, neutral). The input is the summarized text and audio data, and the output is the sentiment analysis result.

[1026] Step 5:

[1027] The server uses a notification mechanism to adjust the notification content based on the emotion analysis results. For example, if the user is feeling stressed, the server adjusts the notification content by making the summary more concise. The input is the emotion analysis results and the summarized text data, and the output is the adjusted notification content.

[1028] Step 6:

[1029] The server sends the tailored notification content to the user's device. The notification is sent in the form of email or other format using the SMTP protocol. A notification engine (e.g., the smtplib library) is used to send emails with detailed information to staff or other interested parties. The input is the tailored notification content, and the output is the notification sent to the user's device.

[1030] Step 7:

[1031] The user checks the received notification and decides how to respond if necessary. Based on the content of the notification, the next action (e.g., direct response by staff, request for additional information) is taken. The input is the notification sent to the user's device, and the output is the user's decision on action.

[1032] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1033] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1034] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1035] [Fourth embodiment]

[1036] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1037] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1038] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1039] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1040] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1041] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1042] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1043] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1044] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1045] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1046] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1047] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1048] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1049] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[1050] 1. Automated Response Methods

[1051] When a call comes in, the server activates the automated response means, which responds appropriately to the caller based on a pre-set script. For example, the automated response means might say, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves them as text data.

[1052] 2. Generation means

[1053] The recorded content of the call is analyzed by the generation means. The server sends the recorded text data to an AI model, which summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[1054] 3. Means of notification

[1055] Once the summary is generated, the server sends it to the user's device as an SMS. At this time, the summary is conveyed to the user as a short message. Specifically, a message such as "Proposal for the introduction of a new product. Please schedule a date for the next meeting to discuss the details" will be sent to the user's smartphone.

[1056] 4. Selection methods

[1057] Users can check the contents of the SMS they receive and decide whether or not a response is necessary. For example, if they determine that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore it.

[1058] Specific examples

[1059] As an example, consider a scenario in which a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic answering means, responding with "Hello, thank you for calling. How can I help you?" The caller then introduces the new product and requests for a next meeting, and the content is recorded and saved as text data. The server then uses a generation means to analyze the text data and summarizes the content of the call as "Explanation regarding the proposed introduction of the new product. We would like to set a date for the next meeting to discuss the details." This summary is then sent to the user's smartphone as an SMS. The user checks the content of the SMS and, determining that it is important, calls back and sets up a detailed meeting.

[1060] The present invention allows users to concentrate on important phone calls without being bothered by unnecessary calls, and is also effective as a means for avoiding nuisance calls and fraudulent calls.

[1061] The processing flow will be explained below.

[1062] Step 1:

[1063] Incoming calls and starting automatic answering

[1064] Server: When a call comes in, the server detects the call and activates an automated attendant, which greets the caller based on a pre-defined script (e.g., "Hello, thank you for calling. How can I help you?").

[1065] Step 2:

[1066] Call recording

[1067] Server: The automated answering system receives the caller's speech, and the server records the call in real time. The recorded content is converted into text data and stored in data storage.

[1068] Step 3:

[1069] Ending the call and preparing to send data

[1070] Server: When the call ends, prepares to send the recorded text data to the generation means. At this time, the process of converting the recorded file into text format is executed.

[1071] Step 4:

[1072] Generative data analysis and summary generation

[1073] Server: The recorded text data is sent to the AI ​​model, which analyzes the data, extracts key points, and generates a summary.

[1074] Example: A summary such as "Explanation regarding the proposal to introduce a new product. I would like to set a date for the next detailed meeting." is generated.

[1075] Step 5:

[1076] Convert summary content to SMS format and send

[1077] Server: Converts the generated summary into SMS format, then sends the SMS message to the user's device.

[1078] Step 6:

[1079] Receipt and confirmation by the user

[1080] User: The SMS arrives on the user's device and a summary is displayed. The user confirms the SMS content.

[1081] Example: A message is sent to the terminal saying, "Proposal for the introduction of a new product. I would like to set a date for the next detailed meeting."

[1082] Step 7:

[1083] Deciding and implementing a response

[1084] User: After reviewing the summary, the user decides what to do next based on the summary, and if necessary, calls back or takes other action.

[1085] Example: The user decides that the summary is important and calls back to schedule a meeting to discuss the details.

[1086] This system's series of processes allows users to focus on important phone calls without being bothered by unnecessary calls, and also makes it easier to avoid nuisance and fraudulent calls.

[1087] Example 1

[1088] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1089] In conventional telephone answering systems, users must answer the phone in person to check the content of incoming calls, which leads to problems such as wasted call time and increased burden from nuisance calls.In addition, there is a high risk that important call content will be overlooked, so an effective filtering method is necessary.

[1090] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1091] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the calls, a means for analyzing and summarizing the recorded contents of the calls, a means for notifying the user of the summarized contents of the calls to the user's terminal, a means for allowing the user to select and respond based on the summary, a means for saving the contents of the calls as text data, a means for sending the text data to an artificial intelligence model for analysis, and a means for summarizing the contents of the summaries in an easy-to-read format and sending them to the user's terminal. This allows the user to quickly respond to important calls without being bothered by unnecessary calls. It also makes it possible to avoid nuisance calls and fraudulent calls.

[1092] An "automatic answering means" is a means for automatically answering an incoming call and responding based on a pre-set script.

[1093] "Means for recording telephone conversations" refers to a means for recording telephone conversations as audio data and saving it.

[1094] The "generation means" is a means for analyzing the recorded content of a call and summarizing it using an artificial intelligence model.

[1095] "Means of notifying the user's device" refers to a means of sending a summary of the call contents to the user's mobile device or smartphone in the form of SMS or the like.

[1096] The "selection means" is a means for the user to check the received summary content and determine the need for action based on its importance.

[1097] "Means for saving as text data" refers to means for converting recorded voice data into text format and saving it.

[1098] An "artificial intelligence model" is a technology that uses machine learning techniques to analyze text data and perform summarization or other processing. Specific examples include generative AI models.

[1099] "Means for summarizing in an easy-to-read and concise manner" refers to a means for omitting the detailed parts of the generated summary and extracting only the main points to express them concisely.

[1100] The present invention is a system that automatically answers incoming calls and analyzes and summarizes the content of the call. This system is configured as follows:

[1101] 1. Automated Response Methods

[1102] When a call comes in, the server activates an automatic answering means. The automatic answering means responds appropriately to the caller based on a pre-set script. For example, it might respond, "Hello, thank you for calling. How can I help you?" At this time, the server records the contents of the call and saves it as audio data.

[1103] 2. Recording Method

[1104] The server stores the recorded conversation as audio data and also as text data, using voice recognition technology.

[1105] 3. Generation means

[1106] The text data of the call is sent by the server to an artificial intelligence model (e.g., OpenAI GPT-4), which analyzes and summarizes the content. For example, the content of a sales call may be summarized as "Explanation regarding the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[1107] 4. Means of notification

[1108] The server then sends the generated summary to the user's device as an SMS message. The summary is written in a concise and easy-to-read format. For example, the user's smartphone will receive a message such as, "Proposal for the introduction of a new product. We would like to set up a date for the next meeting to discuss the details."

[1109] 5. Selection methods

[1110] Users can check the contents of the SMS they receive and decide whether the call is important or not. If it is, they can call back and schedule a detailed meeting. If it is not important, they can ignore the call.

[1111] Hardware and software used

[1112] Server: Cloud computing service (e.g. AWS, Google Cloud)

[1113] Speech recognition technology: Technology that converts voice data into text data

[1114] Artificial intelligence models: Generative AI models such as OpenAI GPT-4

[1115] SMS sending service: Use a service such as Twilio

[1116] Specific examples

[1117] Consider the case where a businessman receives a sales call on his mobile phone. The server detects the incoming call and quickly activates an automated answering system, responding with, "Hello, thank you for calling. How can I help you?" The caller introduces the new product and requests a next meeting, and the conversation is recorded and saved as voice data by the server. The server then saves this voice data as text data using speech recognition technology, and then analyzes the text data using a generation system to summarize the conversation as, "An explanation of the proposed introduction of the new product. I would like to set a date for the next meeting to discuss the details." This summary is sent to the businessman's smartphone as an SMS. The businessman checks the contents of the SMS and, determining that it is important, calls back himself to set up a detailed meeting.

[1118] Prompt Sentence Examples

[1119] "Please tell me the procedure for the system to automatically answer incoming calls and analyze and summarize the contents of the calls."

[1120] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1121] Step 1:

[1122] When the server detects an incoming call, it activates the auto-answering means. Specifically, it receives an incoming call signal and executes the auto-answering program. The input is the incoming call signal, and the output is a trigger signal to start the response. Based on this trigger signal, the server executes a pre-set script and plays an auto-answering message such as "Hello, thank you for calling. How can I help you?"

[1123] Step 2:

[1124] The server records the contents of telephone conversations and saves them as audio data. From the moment a call begins, the server activates the recording function and captures all of the call content. The input is the audio signal during the call, and the output is the recorded audio data. This audio data is saved in a database after the call ends.

[1125] Step 3:

[1126] The server converts the stored voice data into text data using voice recognition technology. This process uses voice recognition software. The input is the recorded voice data, and the output is text data generated by the voice recognition software. The server stores this text data and prepares it for the next analysis step.

[1127] Step 4:

[1128] The server analyzes the text data using a generation method and generates a summary. Specifically, the text data is sent to a generative AI model (e.g., OpenAI GPT-4), which summarizes the content. The input is text data, and the output is summarized text data. For example, the content of a long sales call can be summarized as "An explanation of the proposal to introduce a new product. We would like to set a date for the next meeting to discuss the details."

[1129] Step 5:

[1130] The server sends the generated summary to the user's device as an SMS message. Specifically, it uses an SMS sending service (e.g., Twilio) to compose and send the message. The input is the summarized text data, and the output is an SMS message sent to the user's device. For example, a message such as "Proposal for the introduction of a new product. Please schedule the next meeting to discuss the details" is sent to the user's smartphone.

[1131] Step 6:

[1132] Users can check the contents of the SMS they receive and decide how to respond based on its importance. The input is the SMS message, and the output is the action the user takes (e.g., call back or ignore). If the user determines that the summary is important, they can call back and schedule a detailed meeting. Conversely, if they determine that the message is not important, they can ignore the message.

[1133] (Application example 1)

[1134] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1135] In recent years, the number of nuisance and fraudulent phone calls has been increasing, causing users to be bothered by unnecessary calls and potentially miss important calls. In addition, businesses need to handle a large number of phone calls, creating a need for efficient call management. Furthermore, there is a need for a method to reduce the time and effort required to understand the content of calls that require specific responses. To solve these issues, there is a need for technology that can automatically analyze and summarize the content of phone calls and communicate it to users.

[1136] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1137] In this invention, the server includes a means for answering calls using an automatic answering means and recording the contents of the call, a means for analyzing the recorded contents of the call and generating a summary, a means for notifying the user of the summarized contents of the call to the user's terminal, and a notification means for providing the summarized contents via multiple notification means (e.g., SMS and email). This allows the user to quickly grasp important summarized information without having to listen to the entire contents of the call. Furthermore, the diversification of notification means allows the user to receive the summary in the most optimal way depending on their situation and priority.

[1138] An "automatic answering means" is a device or software that automatically answers an incoming call and records the content of the call.

[1139] "Means for recording telephone conversations" refers to a device or software for saving the contents of incoming telephone conversations as voice data or text data.

[1140] The "generator" is a device or software that analyzes the recorded conversation content, extracts important information, and generates a summary.

[1141] The "notification means" is a device or software for transmitting the generated summary to the user's terminal.

[1142] The "selection means" is a device or software that allows the user to determine the importance based on the summary received and select an appropriate response.

[1143] "Multiple notification methods" is a general term for methods of conveying information to users in different ways, such as SMS and email.

[1144] An "artificial intelligence model" is an algorithm or software used to analyze call content and generate summaries using machine learning and deep learning techniques.

[1145] This invention is a system that automatically answers an incoming call, analyzes and summarizes the contents of the call, and notifies the user. Next, we will explain how to specifically implement this system.

[1146] This system mainly uses a server, user terminals, and a network environment. The server is the main hardware for running a program that integrates speech recognition, text analysis, summary generation, and notification functions, and uses the following software to realize each function. For speech recognition, the "speech_recognition" Python library is used, and for summary generation, the "t5-small" model from Hugging Face, a generative AI model from "transformers," is used. Notification methods include sending SMS using the "Twilio" API and email using Python's "smtplib."

[1147] When the server detects an incoming call, it activates an automatic answering system and answers the call based on a pre-defined script. During this process, the incoming voice is recorded through a microphone and converted into text data using the "speech_recognition" library. The converted text data is then stored on the server.

[1148] The server then sends the stored text data to a generative AI model in the "transformers" library to generate a summary, which is a short piece of text containing important information.

[1149] This summary is sent to the user's device via SMS using the Twilio API or via email using smtplib. The user can then review the message and decide how to respond based on its importance.

[1150] Specific examples

[1151] For example, consider a scenario in which this system is implemented in a company's customer support. When a new customer support inquiry is received by phone, the server uses an automated response function to respond with, "Hello, this is Company Name Support. How can I help you?" The call is recorded and then converted into text using the "speech_recognition" library. A generative AI model in the "transformers" library then summarizes the call as, "There is a problem with Product A. Please contact the user for details." This summary is then sent to the representative's device via SMS and email.

[1152] Prompt Sentence Examples

[1153] Example prompt sentence:

[1154] Summarize the customer support call.

[1155] Voice input: "Hello, thank you for contacting us about an issue with Product A. Please tell us the specific problem you're experiencing and how it's progressing."

[1156] Voice output: "There is a problem with product A. Please check with the user for details."

[1157] In this way, the server automatically analyzes and summarizes the contents of the call and quickly conveys only the important information to the user, enabling efficient and effective telephone response.

[1158] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1159] Step 1:

[1160] When a call is received, the server activates an automatic answering means.

[1161] Input: Incoming phone signal.

[1162] Data processing and calculation: Detects incoming calls and loads pre-configured scripts to prepare an automatic response.

[1163] Output: Generates a response message based on the auto-response script.

[1164] Step 2:

[1165] The server answers the call using an automated answering means and records the contents of the call.

[1166] Input: Greeting message and call audio based on the auto-attendant script.

[1167] Data processing and calculation: Capture the contents of the call as audio data through the microphone and generate a recording file.

[1168] Output: A recording file of the call saved as audio data.

[1169] Step 3:

[1170] The server converts the recorded conversation into text data using a voice recognition library.

[1171] Input: Recording file (audio data).

[1172] Data processing and calculation: Analyzes voice data using the speech recognition library (speech_recognition) and generates text data.

[1173] Output: The call contents converted into text data.

[1174] Step 4:

[1175] The server sends the text data to a generative AI model to generate a summary.

[1176] Input: Text data.

[1177] Data processing and calculation: Summarize text data using a generative AI model (the t5-small model from the transformers library).

[1178] Output: Summarized text data.

[1179] Step 5:

[1180] The server sends the summarized content to the user's device via multiple notification methods (SMS and email).

[1181] Input: Abstracted text data.

[1182] Data processing and calculation: Using notification methods (Twilio API and smtplib), the summary data is sent to the user's terminal as SMS and email.

[1183] Output: Summary notification via SMS and email received on user device.

[1184] Step 6:

[1185] The user reviews the received summary and decides on the appropriate response, if necessary.

[1186] Input: Summary data sent via SMS or email.

[1187] Data processing and calculation: Determine importance and urgency based on summary data.

[1188] Output: The subsequent action resulting from the user's decision (e.g., calling back, replying by email, etc.).

[1189] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1190] This invention combines a system that uses an automatic answering means to respond to incoming calls, records and analyzes the content of the call, and an emotion engine that recognizes the user's emotions. This enables flexible responses that match the user's emotional state, resulting in more effective and personalized notifications.

[1191] System configuration

[1192] This system is configured as follows:

[1193] 1. Automated Response Methods

[1194] When a call comes in, the server activates an automated answering machine, which responds to the call based on a pre-defined script.

[1195] 2. Recording Method

[1196] The contents of the call are recorded in real time by the server and saved as text data.

[1197] 3. Generation means

[1198] The recorded text data is sent to a generation means, specifically an artificial intelligence model, where it is analyzed and summarized.

[1199] 4. Emotion Engine

[1200] The emotion engine analyzes emotions from the user's voice, facial expressions, and text input. For example, it analyzes facial expressions and voice when the user opens an SMS to determine the user's emotional state.

[1201] 5. Means of notification

[1202] Based on the analysis results of the emotion engine, the server sends a summary of the call to the user's device in a user-friendly format as an SMS. If the user's emotional state indicates stress, the server can summarize the content briefly or adjust the notification method.

[1203] 6. Selection methods

[1204] Users can check the contents of the SMS they receive and choose what to read based on their emotional state, for example, they can skip over messages they deem unimportant and check the details of messages they deem important.

[1205] Specific examples

[1206] Consider a scenario where a businessman receives a sales call on his mobile phone. The server detects the incoming call and activates the automatic response means. The automatic response means responds with "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server uses the generation means to analyze the text data and summarize it as "An explanation of the proposed introduction of a new product. We would like to set a date for the next meeting to discuss the details."

[1207] The emotion engine analyzes data from the user's smartphone camera and microphone to determine the user's emotional state. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary of the message. The user can check the received SMS, check the details if it is important, or skip if it is not necessary.

[1208] Specific processing of the program

[1209] 1. Detecting incoming calls

[1210] When the server detects an incoming call, it activates an automatic answering means and starts a call. The automatic answering means answers the call according to a pre-set script.

[1211] 2. Recording of call content

[1212] The server records the call in real time, converts it into text data, and saves it.

[1213] 3. Summary Generation

[1214] A generating means is used to analyze the recorded text data and generate a summary, which is then converted into an SMS format.

[1215] 4. Sentiment Analysis and Notifications

[1216] The emotion engine analyzes the user's emotional state and adjusts the content and method of notifications based on the results. For example, if the user is feeling stressed, the summary content will be shortened or the notifications will be suppressed. The adjusted content will be sent to the user's device as an SMS.

[1217] 5. User Choice

[1218] The user can check the summary of the SMS they received and decide whether or not a response is necessary. If necessary, they can take action such as calling back.

[1219] This system allows users to quickly obtain only the important information without wasting time on unnecessary calls. Utilizing an emotion engine also reduces the user's mental burden and enables more personalized notification responses.

[1220] The processing flow will be explained below.

[1221] Step 1:

[1222] Incoming calls and starting automatic answering

[1223] Server: When a call comes in, the server detects the call and activates an automated response system, which responds to the caller based on a pre-defined script, saying, "Hello, thank you for calling. How can I help you?"

[1224] Step 2:

[1225] Call recording

[1226] Server: When the automated answering system receives the call, the server records the call in real time. The recorded content is converted into text data and stored in a database.

[1227] Step 3:

[1228] Ending the call and preparing to send data

[1229] Server: When the call ends, prepares the recorded text data to be sent to the generation means. During this data conversion process, the text data is formatted.

[1230] Step 4:

[1231] Generative data analysis and summary generation

[1232] Server: The recorded text data is sent to the AI ​​model, which analyzes the text data, extracts key points, and creates a summary.

[1233] Example: If the call content is "Pitching a new product and requesting a next meeting," the generated summary will be "Introducing a new product. Requesting a detailed meeting."

[1234] Step 5:

[1235] Analyzing user emotions with an emotion engine

[1236] Device: Uses the camera and microphone on the user's smartphone to capture the user's facial expressions and voice data.

[1237] Server: The acquired facial expression and voice data is sent to the emotion engine, which analyzes the user's emotional state (e.g., stress level, excitement level, relaxation level).

[1238] Step 6:

[1239] Tailoring notification content based on emotional state

[1240] Server: Based on the user's emotional state analyzed by the emotion engine, the notification method and content of the summarized call details are adjusted. For example, if the user is feeling stressed, the notification content is made shorter.

[1241] Step 7:

[1242] Convert summary content to SMS format and send

[1243] Server: Converts the generated summary into SMS format and sends the adjusted notification content to the user's device as an SMS.

[1244] Example: The summary "Introducing a new product. Requesting a meeting to discuss details" can be converted into an SMS message as "Important: Introduces a new product. Requesting a meeting to discuss details."

[1245] Step 8:

[1246] Receipt and confirmation by the user

[1247] User: An SMS is delivered to the user's device. The user checks the contents of the SMS.

[1248] Example: The device displays the message "Important: Introducing a new product, please arrange a meeting."

[1249] Step 9:

[1250] Deciding and implementing a response

[1251] User: Based on the summary of the SMS received, the user decides whether to take action and, if necessary, take action such as calling back to confirm the details.

[1252] Example: The user decides it's important, calls the sales rep back, and schedules a meeting.

[1253] This system's series of processes allows users to quickly obtain only the important information without being bothered by unnecessary calls. In addition, the use of an emotion engine reduces the user's mental burden and enables more personalized responses.

[1254] Example 2

[1255] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1256] While automated answering and recording of incoming phone calls are commonplace today, this alone is not sufficient for user convenience. In particular, without call content summarization and emotional analysis, users are often overwhelmed with unnecessary information, which can increase their mental burden. Furthermore, special consideration is required for users under stress, and flexible responses tailored to individual emotional states are required. Therefore, there is a need for a system that not only analyzes and summarizes call content, but also recognizes the user's emotions and adjusts notification content accordingly.

[1257] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1258] In this invention, the server includes means for recording the contents of a call when a call is received, means for analyzing the recorded contents of the call and generating a summary, means for analyzing the user's emotions, and means for adjusting the summarized contents of the call based on the user's emotional state and notifying the user's terminal. This makes it possible to quickly provide important information while taking the user's emotional state into consideration.

[1259] An "incoming call" refers to receiving an external call to a communication device.

[1260] "Automatic answering means" refers to a function that responds to incoming calls with a pre-set script or message without manual operation.

[1261] "Call content" refers to all audio information exchanged over the telephone.

[1262] "Recording means" refers to a device or software for storing the contents of a call.

[1263] "Generation means" refers to a processing device or software for analyzing the recorded content of a call and generating a summary.

[1264] "Emotion analysis means" refers to the function of analyzing the user's emotional state from audio and video data.

[1265] "Abridged call content" refers to information that briefly summarizes the content of a call.

[1266] "User's device" refers to a communication device used by a User that can receive System notifications.

[1267] "Notification means" refers to the function of delivering the generated summary content to the user's terminal.

[1268] "Selection tool" refers to a function that allows a user to select an appropriate response based on the summary content received.

[1269] An "artificial intelligence model" refers to an algorithm or system that learns from data and performs analysis and predictions.

[1270] This invention combines an emotion analysis system that recognizes the user's emotions with a system that uses an automatic answering means to respond to incoming calls and records and analyzes the content of the call, thereby enabling notifications to be delivered in line with the user's emotional state, enabling more effective and personalized information to be provided.

[1271] This system is configured as follows:

[1272] Hardware and Software Use

[1273] Server: VoIP system, Python script, Twilio API, Google Cloud Speech-to-Text API, OpenAI GPT-4, Azure Cognitive Services

[1274] User's device: Smartphone camera, microphone, SMS function

[1275] Specific processing

[1276] Incoming calls and auto-answering

[1277] The server detects an incoming call. When an incoming call is detected, it activates an automated response system that responds to the caller based on a pre-defined script. For example, it might send a message like this: "Hello, thank you for calling. How can I help you?"

[1278] Call recording

[1279] The server records the call in real time and converts the audio data into text data for storage. This process uses the Google Cloud Speech-to-Text API.

[1280] Generate a summary of the call

[1281] The server then uses a generation method to analyze the recorded text data and generate a summary. Using the OpenAI GPT-4 generative AI model, the server inputs the following prompt: "Please summarize the following call content: [Text data]."

[1282] Emotion analysis

[1283] The sentiment analysis method uses data obtained from a smartphone camera and microphone to analyze the user's emotional state. This data is analyzed through Microsoft Azure Cognitive Services to determine the user's emotional state.

[1284] Call summary notifications

[1285] Based on the results of the emotion analysis, the server sends a summary of the call to the user's device as an SMS. If the user's emotional state indicates stress, the summary will be made more concise or the notification method will be adjusted.

[1286] Specific examples

[1287] When a businessman receives a sales call on his mobile phone, the server detects the incoming call and activates the automatic response means. The automatic response means responds with, "Hello, thank you for calling. How can I help you?" The contents of the call are recorded and saved as text data. The server then uses the generation means to summarize it as, "This is an explanation of the proposal to introduce a new product. I would like to set a date for the next meeting to discuss the details."

[1288] The emotion analysis function analyzes the user's emotions based on data collected from the smartphone camera and microphone. For example, if the user is feeling stressed, the server adjusts the notification method and sends a concise summary. The user can check the received SMS and, if the content is deemed important, review the details, or skip if it is not necessary.

[1289] This invention allows users to quickly obtain only the information they need without wasting time on unnecessary calls. In addition, emotion analysis reduces the user's mental burden and enables more personalized notification responses.

[1290] Prompt Sentence Examples

[1291] Please summarize the call below.

[1292] [Text data]

[1293] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1294] Step 1:

[1295] The server detects incoming calls. When a call comes in, it activates an automated answering system that responds to the caller based on a pre-defined script. For example, it might play a message like, "Hello, thank you for calling. How can I help you?"

[1296] Input: Incoming signal

[1297] Output: Start auto-reply, play greeting

[1298] What happens: The VoIP system detects an incoming call, and a Python script uses the Twilio API to play a pre-recorded message.

[1299] Step 2:

[1300] The server records the call in real time and converts the audio data into text data for storage, using the Google Cloud Speech-to-Text API.

[1301] Input: Call audio data

[1302] Output: Text data

[1303] How it works: The recorded audio data is saved in WAV format, and a Python script calls the Google Cloud Speech-to-Text API to convert the audio data into text data.

[1304] Step 3:

[1305] The server uses a generative method to analyze the text data and generate a summary. At this stage, a generative AI model (OpenAI GPT-4) is used.

[1306] Input: Text data

[1307] Output: Summary text

[1308] How it works: The Python script inputs the following prompt to the generative AI model: "Please summarize the following call content: [Text data]." The generative AI model outputs the summary text.

[1309] Step 4:

[1310] The server analyzes data obtained from the user's camera and microphone using emotion analysis tools, which are powered by Microsoft Azure Cognitive Services.

[1311] Input: Audio and video data from cameras and microphones

[1312] Output: Emotion analysis results

[1313] How it works: A smartphone app collects data from the camera and microphone and sends it to a server, which then sends it to Azure Cognitive Services for sentiment analysis.

[1314] Step 5:

[1315] The server adjusts the notification content based on the results of the sentiment analysis and sends a summary text to the user's device. The notification method uses the Twilio API.

[1316] Input: Summary text, sentiment analysis results

[1317] Output: SMS notification

[1318] Specific operation: The server compiles a concise summary text based on the sentiment analysis results and sends it as an SMS to the user's smartphone using the Twilio API.

[1319] Step 6:

[1320] Users can check the received SMS and respond as needed, for example, by checking the details if the message is important and calling back if necessary.

[1321] Input: SMS notification

[1322] Output: User action

[1323] Specific action: The user opens the SMS on their smartphone, checks the contents, and if necessary calls back or takes other action.

[1324] (Application example 2)

[1325] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1326] Current automated answering systems are effective in answering incoming calls and summarizing the content of calls, but they do not provide flexible responses that take into account the user's emotional state. This can leave users feeling stressed and dissatisfied, resulting in problems such as inefficiencies such as an increase in complaints and inquiries. Customer service in brick-and-mortar stores also needs to be able to respond in a way that takes into account the emotional state of customers, but this is difficult to achieve with conventional systems.

[1327] The specific processing 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 answering an incoming call using an automatic answering means and recording the content of the call, means for analyzing and summarizing the recorded content of the call, means for analyzing the emotional state of the user, and means for adjusting the content of the notification based on the analysis result. This enables flexible and personalized responses according to the emotional state of the user.

[1328] "When a call comes in" refers to a situation where an external connection is attempted over a telephone line and the user's device or system receives the connection.

[1329] "Automatic answering means" refers to a device or program that has the function of automatically answering an incoming telephone call without human intervention.

[1330] "Call recording means" refers to a device or program that automatically saves conversations and audio data that occur during a telephone call.

[1331] "Recorded call content" refers to information saved as audio or text data exchanged during a call.

[1332] "Analysis and summary generation means" refers to a device or program that analyzes the recorded content of a call, extracts only the important information, and provides it in a shortened form.

[1333] "Emotion analysis means for analyzing a user's emotional state" refers to technologies and algorithms for measuring or inferring emotions from a user's voice, facial expressions, text input, etc.

[1334] "Notification means for adjusting notification content" refers to a device or program for automatically changing the method or content of notifications based on the user's emotional state.

[1335] "Means for notifying the user of the summarized call content" refers to a device or program that transmits the information summarized by the generating means to the user's device such as a smartphone or PC.

[1336] "A means for the user to decide what to do based on the summary content" refers to a device or program that allows the user to look at the summary information notified to them and decide what to do next based on its importance and necessity.

[1337] This invention provides a system that uses an automatic answering means to answer incoming calls, records the contents of the call, analyzes the record to generate a summary, notifies the user of the summary, and further analyzes the user's emotional state to optimize the content of the notification. The following describes an example of application of this system to customer support functions in brick-and-mortar stores.

[1338] System Program

[1339] The server has a speech recognition engine, a text generation engine, an emotion analysis engine, a notification engine, and the ability to send emails using the SMTP protocol. It is recommended that the main software configuration uses the speech_recognition library, the TextBlob library, the pipeline function of the transformers library, and the smtplib library.

[1340] Explanation of program processing

[1341] 1. Incoming call detection and automatic answering

[1342] When a call comes in, the server's automatic answering means is activated, which responds to the call and gives an initial response to the customer based on a pre-set script.

[1343] 2. Recording of call content

[1344] The server uses a speech recognition engine (e.g., speech_recognition library) to record the call and save it as text data.

[1345] 3. Summary Generation

[1346] The recorded text data is analyzed by a text generation engine (e.g., the TextBlob library), which extracts key phrases and concepts and generates summaries based on them.

[1347] 4. Emotional state analysis

[1348] The summarized text and audio data are analyzed for the user's emotional state using a sentiment analysis engine (e.g., the pipeline function in the transformers library).

[1349] 5. Coordination and delivery of notifications

[1350] Based on the results of the sentiment analysis, the notification engine will tailor the notification content. For example, if it detects that the user is feeling stressed, a brief summary of the content will be sent. Notifications are implemented via the SMTP protocol (e.g., the smtplib library).

[1351] Specific examples

[1352] This technology is applied to customer service in brick-and-mortar stores. When a customer makes an inquiry, the in-store customer support robot automatically responds, records the content of the inquiry, and saves it as text data. The saved text data is analyzed in real time, and a summary is generated. At this time, the emotional state of the customer is also analyzed along with the analyzed summary data. For example, if a customer expresses anger, this information is notified to staff, and a flexible response is required.

[1353] Prompt Sentence Examples

[1354] "Analyze the following speech input and output a result about the emotional state: 'Hello, customer. Can you tell me more about this weekend's sale? I'm happy.'"

[1355] In this way, the system can provide a more personalized customer experience and improve the user experience.

[1356] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1357] Step 1:

[1358] When a call comes in, the server detects the call. It receives the incoming call signal and activates the automatic answering means. The automatic answering means answers the call and starts the call based on a pre-set script. The input is the incoming call signal and the output is the automatic answering message.

[1359] Step 2:

[1360] The server uses a speech recognition engine (e.g., the speech_recognition library) to record the call contents. The input is the voice data during the call, and the voice data is converted into text data in real time and saved. The output is the converted text data.

[1361] Step 3:

[1362] The server uses a generator (e.g., TextBlob library) to analyze the recorded text data, extract important key phrases, and generate a summary. The text data is analyzed and key phrases and key content parts are extracted. The input is the text data of the call, and the output is the summarized text data.

[1363] Step 4:

[1364] The server uses sentiment analysis tools (e.g., the pipeline function in the transformers library) to analyze the user's emotional state from the summarized text and audio data. It analyzes the audio and text data and determines the emotional state (e.g., positive, negative, neutral). The input is the summarized text and audio data, and the output is the sentiment analysis result.

[1365] Step 5:

[1366] The server uses a notification mechanism to adjust the notification content based on the emotion analysis results. For example, if the user is feeling stressed, the server adjusts the notification content by making the summary more concise. The input is the emotion analysis results and the summarized text data, and the output is the adjusted notification content.

[1367] Step 6:

[1368] The server sends the tailored notification content to the user's device. The notification is sent in the form of email or other format using the SMTP protocol. A notification engine (e.g., the smtplib library) is used to send emails with detailed information to staff or other interested parties. The input is the tailored notification content, and the output is the notification sent to the user's device.

[1369] Step 7:

[1370] The user checks the received notification and decides how to respond if necessary. Based on the content of the notification, the next action (e.g., direct response by staff, request for additional information) is taken. The input is the notification sent to the user's device, and the output is the user's decision on action.

[1371] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1372] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1373] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1374] The emotion identification model 59 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 an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1375] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1376] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1377] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1378] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1379] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs 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 a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1380] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1381] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1382] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1383] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1384] 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.

[1385] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1386] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1387] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1388] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1389] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1390] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1391] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1392] The following is further disclosed regarding the above embodiment.

[1393] (Claim 1)

[1394] a means for answering a call using an automatic answering means when a call is received and recording the contents of the call;

[1395] A generating means for analyzing and summarizing the recorded call content;

[1396] a means for notifying a user of the summarized call content on the user's terminal;

[1397] The system includes a means for users to choose and act on the summary content.

[1398] (Claim 2)

[1399] 2. The system according to claim 1, wherein the automatic response means responds to the other party based on a pre-set script.

[1400] (Claim 3)

[1401] 2. The system of claim 1, wherein the generating means uses an artificial intelligence model to analyze and summarize the call content.

[1402] "Example 1"

[1403] (Claim 1)

[1404] a means for answering a call using an automatic answering means when a call is received and recording the contents of the call;

[1405] A generating means for analyzing and summarizing the recorded call content;

[1406] a means for notifying a user of the summarized call content on the user's terminal;

[1407] A means for users to choose and act on the summary content;

[1408] A means for saving the contents of the call as text data;

[1409] means for transmitting the text data to an artificial intelligence model for analysis;

[1410] A system that includes a means for summarizing the content in a readable and concise format and transmitting it to a user terminal.

[1411] (Claim 2)

[1412] 2. The system according to claim 1, wherein the automatic response means responds to the other party based on a pre-set script.

[1413] (Claim 3)

[1414] 2. The system of claim 1, wherein the generating means uses an artificial intelligence model to analyze and summarize the call content.

[1415] "Application Example 1"

[1416] (Claim 1)

[1417] a means for answering a call using an automatic answering means when a call is received and recording the contents of the call;

[1418] A generating means for analyzing and summarizing the recorded call content;

[1419] a means for notifying a user of the summarized call content on the user's terminal;

[1420] A means for users to choose and act on the summary content;

[1421] A notification method that provides summarized content through multiple notification methods (e.g., SMS and email);

[1422] A system including:

[1423] (Claim 2)

[1424] 2. The system according to claim 1, wherein the automatic response means responds to the other party based on a pre-set script.

[1425] (Claim 3)

[1426] 2. The system of claim 1, wherein the generating means uses an artificial intelligence model to analyze and summarize the call content.

[1427] "Example 2: Combining Emotion Engines"

[1428] (Claim 1)

[1429] a means for answering a call using an automatic answering means when a call is received and recording the contents of the call;

[1430] A generating means for analyzing and summarizing the recorded call content;

[1431] An emotion analysis means for analyzing the emotion of a user;

[1432] means for adjusting the summarized call content based on the emotional state of the user and notifying the user's terminal;

[1433] The system includes a means for users to choose and act on the summary content.

[1434] (Claim 2)

[1435] 2. The system according to claim 1, wherein the automatic response means responds to the other party based on a pre-set script.

[1436] (Claim 3)

[1437] 2. The system of claim 1, wherein the generating means uses an artificial intelligence model to analyze and summarize the call content.

[1438] "Application example 2 when combining emotion engines"

[1439] (Claim 1)

[1440] a means for answering a call using an automatic answering means when a call is received and recording the contents of the call;

[1441] A generating means for analyzing and summarizing the recorded call content;

[1442] emotion analysis means for analyzing the emotional state of a user;

[1443] a notification means for adjusting the notification content based on the analysis result;

[1444] a means for notifying a user of the summarized call content on the user's terminal;

[1445] The system includes a means for users to choose and act on the summary content.

[1446] (Claim 2)

[1447] 2. The system according to claim 1, wherein the automatic response means responds to the other party based on a pre-set script.

[1448] (Claim 3)

[1449] 2. The system of claim 1, wherein the generating means uses an artificial intelligence model to analyze and summarize the call content. [Explanation of symbols]

[1450] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for answering a call using an automatic answering means when a call is received and recording the contents of the call; A generating means for analyzing and summarizing the recorded call content; a means for notifying a user of the summarized call content on the user's terminal; The system includes a means for users to choose and act on the summary content.

2. 2. The system according to claim 1, wherein the automatic response means responds to the other party based on a preset script.

3. 2. The system of claim 1, wherein the generating means uses an artificial intelligence model to analyze and summarize the call content.

Images