system

Abstract

We provide the system. [Solution] A means for receiving a document file and analyzing it as text data, A means for detecting specific keywords or phrases from analyzed text data, A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules, A means of reconstructing the replaced text data as a document file, A means of outputting a reconstructed document file, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The precautions and texts in the documents provided by manufacturers often contain the expression "our company", but it is inappropriate to directly guide the resellers with this, and it is necessary to manually change it to "manufacturer", which is very time-consuming and laborious. Moreover, such manual work is prone to human errors and may damage the accuracy of the document. In response to such a situation, an automatic system for efficiently and accurately converting the expressions in the document is required.

Means for Solving the Problems

[0005] The system of the present invention includes means for receiving a document file and analyzing it as text data, means for detecting specific keywords or phrases from the analyzed text data, means for replacing the detected keywords or phrases with new expressions according to predefined conversion rules, means for reconstructing the replaced text data as a document file, and means for outputting the reconstructed document file. Furthermore, it also includes means for extracting the uploaded document file as text data in an appropriate manner depending on the format of the document file, and means for utilizing natural language processing technology for replacing the detected keywords or phrases. With such a configuration, it is possible to eliminate manual errors and perform document conversion efficiently and accurately.

[0006] A "document file" refers to an electronic file containing text data or image data, and specifically includes formats such as PDF, Word, and TXT.

[0007] "Text data" refers to data composed of characters and sentences, and is a data format that can be analyzed and edited by programs.

[0008] "Analyzing" refers to the process of breaking down input data into an easily understandable form and identifying each element.

[0009] A "keyword" refers to a specific word or phrase that appears within text data and is an important element that is subject to document transformation and processing.

[0010] A "phrase" refers to a set or group of words that have meaning within a document, and includes specific expressions or phrasing.

[0011] "Detecting" refers to the process of finding and identifying target keywords or phrases within text data.

[0012] A "conversion rule" refers to a predefined set of rules for replacing specific keywords or phrases with other expressions.

[0013] "Replace" refers to the process of changing detected keywords or phrases into other specific expressions.

[0014] "Reconstructing" refers to the process of formatting the converted text data into a new document file and returning it to its original document format.

[0015] "Outputting" refers to the process of saving the converted document in a format that can be provided to the user and making it accessible.

[0016] "Natural language processing technology" refers to the technology that enables computers to understand, generate, and manipulate human language, and is also known as NLP (Natural Language Processing). [Brief explanation of the drawing]

[0017] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9]Shows an emotion map where multiple emotions are mapped. [Figure 10] Shows an emotion map where multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Modes for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), etc.

[0021] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0022] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0037] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0038] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions for use by distributors. The specific procedures and processes for implementing this system are detailed below.

[0039] This system begins with the user uploading document files provided by the manufacturer. The user selects the document files using a terminal and uploads them to the system interface. The terminal then sends the uploaded document files to the server.

[0040] The server receives document files and extracts text data using the appropriate method depending on the file format. If the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0041] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. These keywords and phrases are generally expressions such as "our company" and "standard warranty period."

[0042] The server applies predefined translation rules to detected keywords and phrases. Specifically, it performs processes such as replacing "our company" with "manufacturer." Furthermore, the server uses a custom translation function to make other necessary changes to the wording. This custom translation function utilizes natural language processing technology to perform appropriate translations according to the context.

[0043] The converted text data is reconstructed and returned to its original document format. For example, if the original format was PDF, the converted text data is converted back into a PDF file. The server saves the reconstructed document file and prepares it for later download by the user.

[0044] Finally, a download link appears on the device, and the user downloads the converted document file. In this way, the user can avoid manual errors and perform document conversion efficiently and accurately.

[0045] Specific example

[0046] For example, suppose the manufacturer provides the following document:

[0047] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0048] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." It also changes the phrase "inquiry" to "contact," resulting in the following document:

[0049] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0050] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a distributor.

[0051] The following describes the processing flow.

[0052] Step 1:

[0053] The user selects a document file provided by the manufacturer using a terminal and uploads it to the system interface.

[0054] Step 2:

[0055] The device sends the uploaded document file to the server.

[0056] Step 3:

[0057] The server reads the received document file as text data. If the document is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0058] Step 4:

[0059] The server tokenizes the extracted text data. Tokenization is the process of breaking down text data into words and phrases and identifying each one.

[0060] Step 5:

[0061] The server detects specific keywords or phrases from tokenized text data. This may involve using natural language processing techniques.

[0062] Step 6:

[0063] The server applies predefined conversion rules to the keywords and phrases it detects. Specifically, it performs processes such as replacing "our company" with "manufacturer."

[0064] Step 7:

[0065] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact".

[0066] Step 8:

[0067] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated as a PDF file again.

[0068] Step 9:

[0069] The server saves the reconstructed document files and prepares them for users to download.

[0070] Step 10:

[0071] A download link will appear on the device, and the user will download the converted document file. This file can then be used by the distributor.

[0072] (Example 1)

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

[0074] Traditional manual document conversion processes are time-consuming, labor-intensive, and prone to errors. Furthermore, consistent and improper replacement of specific keywords and phrases is often lacking, making them inefficient, especially when processing large volumes of documents. To address these challenges, a system is needed that automatically and accurately converts specific keywords and phrases within document files.

[0075] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0076] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data into word or phrase units; means for detecting specific keywords or phrases from the tokenized text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to automatically detect specific keywords or phrases and replace them accurately and efficiently.

[0077] A "document file" is an electronic file that contains text data, and includes formats such as PDF and Word files.

[0078] "Text data" refers to string data extracted or parsed from document files, and includes words and phrases.

[0079] "Tokenization" is the process of dividing text data into words or phrases and identifying each one individually.

[0080] A "keyword or phrase" is a word or phrase that has a specific meaning and is an important element within a document.

[0081] "Conversion rules" refer to predefined rules for replacing specific keywords or phrases with new expressions.

[0082] "Reconstruction" is the process of restoring the replaced text data to its original document format.

[0083] "Output" refers to providing the user with the completed document file.

[0084] This invention relates to a system for automatically converting specific keywords or phrases within document files. This significantly reduces manual data entry and allows users to efficiently and accurately convert documents. The specific procedures and processes for implementing this system are detailed below.

[0085] This system includes the user's terminal, a server, appropriate analysis modules, and natural language processing technology. The user selects a document file provided by the manufacturer from their terminal and uploads it to the system. The terminal then sends the uploaded document file to the server.

[0086] The server receives document files in formats such as PDF and Word, selects the appropriate analysis module, and extracts the text data. For example, it uses the PDFLib or pdfminer.six library for PDF files and the python-docx library for Word files.

[0087] The extracted text data is tokenized by the server. Natural language processing libraries (such as NLTK or spaCy) are used for tokenization, dividing the text data into words and phrases. The server then detects specific keywords and phrases from the tokenized text. A predefined keyword list is used for this detection.

[0088] The server applies pre-configured translation rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." Furthermore, it uses custom translation functions as needed to perform contextually appropriate translations using natural language processing techniques (such as BERT or GPT).

[0089] The converted text data is then reconstructed by the server into its original document format. For PDFs, the PDFLib library is used to generate a new PDF file, and for Word files, the python-docx library is used to generate a new Word file.

[0090] Finally, the server saves the reconstructed document file and provides a download link to the user's terminal. The user can then use their terminal to download the converted document file via this link.

[0091] As a concrete example, consider a case where the manufacturer provides the following document:

[0092] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0093] When a user uploads this document to the system, the server extracts the text data and converts the keyword "our company" to "manufacturer" and the phrase "inquiry" to "contact". As a result, a document like the following is generated:

[0094] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0095] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can download the converted document using this link.

[0096] Examples of prompts to input into a generative AI model are as follows:

[0097] Please translate the phrases "our company," "standard warranty period," and "inquiries" into language suitable for sales companies.

[0098] In this way, the system enables efficient and accurate conversion of specific keywords and phrases within a document.

[0099] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0100] Step 1:

[0101] The user selects a document file provided by the manufacturer and uploads it from their device using the system interface. The device processes this upload request and sends the document file to the server.

[0102] Input: Document file selected from the user's terminal (PDF or Word format)

[0103] Output: Document file sent to the server

[0104] Step 2:

[0105] The server receives document files sent from the terminal. It recognizes the format of the document file and selects the appropriate parsing module (PDF parsing module or Word parsing module). For PDF files, it uses the PDFLib or pdfminer.six library to extract text data; for Word files, it uses the python-docx library.

[0106] Input: Document file sent to the server

[0107] Output: Extracted text data

[0108] Specific operation: The PDF parsing module uses PDFLib, and the Word parsing module uses python-docx to extract text data from documents.

[0109] Step 3:

[0110] The server tokenizes the extracted text data. This tokenization uses natural language processing libraries (such as NLTK or spaCy) to divide the text into words and phrases.

[0111] Input: Extracted text data

[0112] Output: Tokenized text data

[0113] Specific operation: The text is split into words and phrases using NLTK's word_tokenize function or spaCy's nlp function.

[0114] Step 4:

[0115] The server detects specific keywords or phrases from tokenized text data. It uses a predefined keyword list to check whether each token is included in the list.

[0116] Input: Tokenized text data

[0117] Output: Detected keywords and phrases

[0118] Specific operation: Loop through the tokenized text data and verify whether each token is present in the keyword list.

[0119] Step 5:

[0120] The server applies predefined conversion rules to detected keywords and phrases. For example, it converts "our company" to "manufacturer." For more complex conversions, it utilizes natural language processing techniques (such as BERT and GPT).

[0121] Input: Detected keywords or phrases

[0122] Output: Converted text data

[0123] Specific operation: Basic substitutions are performed using Python's str.replace function, and context-aware transformations are performed using a transformer model.

[0124] Step 6:

[0125] The server reconstructs the converted text data into the original document format. For PDFs, it uses the PDFLib library to generate a new PDF file, and for Word files, it uses the python-docx library to generate a new Word file.

[0126] Input: Converted text data

[0127] Output: Reconstructed document file (PDF or Word format)

[0128] Specific operation: Insert text data into a new document object and format the entire document.

[0129] Step 7:

[0130] The server saves the reconstructed document file and displays a download link on the user's terminal. The user can then download the converted document file from their terminal via this link.

[0131] Input: Reconstructed document file

[0132] Output: Download link displayed on the user's terminal

[0133] Specific operation: The server saves the document file to the storage directory, generates a download link containing the path, and notifies the user.

[0134] (Application Example 1)

[0135] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0136] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions suitable for sales companies, and further outputs the converted content as audio. Conventional systems require manual modification of the wording in document files, which is inefficient and prone to errors. Furthermore, because they are limited to visual information, it is difficult for visually impaired individuals to obtain the information.

[0137] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0138] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined conversion rules; means for reconstructing the replaced text data as a document file; means for outputting the reconstructed document file; means for converting the converted text data into audio data; and means for outputting the converted audio data. This makes it possible to automatically and efficiently convert the expression of a document file for sales companies and further output its contents as audio.

[0139] A "document file" is a file containing text or image data stored in an electronic format.

[0140] "Text data" refers to information expressed as a string of characters, specifically the content extracted from a document file.

[0141] "Analyzing" means breaking down given data and processing it in order to understand its meaning and structure.

[0142] A "keyword or phrase" refers to a word or short sentence with a specific meaning, representing an important part of a document.

[0143] "Detecting" means finding data that has specific conditions or characteristics.

[0144] "Conversion rules" are pre-defined guidelines or rules for replacing specific keywords or phrases with other expressions.

[0145] "Substitution" means replacing one element with another.

[0146] "Reconstructing" means putting data back together to return it to its original state or format.

[0147] "Outputting" means providing processed data in a format that can be used by the user.

[0148] "Audio data" refers to digital data containing audio information, in a format that can be played back by an audio playback device.

[0149] This invention is a system for efficiently converting the expression of business documents for specific purposes, particularly for generating appropriate expressions for customers and outputting them as audio. The following describes specific embodiments for carrying out the invention.

[0150] Program Overview

[0151] The server system and terminals operate in coordination. The system according to the present invention includes the following main means:

[0152] A method for receiving document files and analyzing them as text data.

[0153] Means for detecting specific keywords or phrases from analyzed text data

[0154] A means of replacing detected keywords or phrases with new expressions according to predefined transformation rules.

[0155] A method for reconstructing the replaced text data as a document file.

[0156] Means for outputting a reconstructed document file

[0157] A means of converting converted text data into audio data.

[0158] Means for outputting converted audio data

[0159] Hardware and software to be used

[0160] The server is hosted using Google Cloud Platform, and utilizes SpaCy for natural language processing and the Google Text-to-Speech API for speech synthesis. The device is a smartphone or tablet with a document file upload application installed.

[0161] Data processing and data calculation

[0162] 1. Upload document files:

[0163] The user uploads document files (such as PDFs or Word documents) provided by the manufacturer to the system interface using their device. The device then sends the uploaded document files to the server.

[0164] 2. Extraction and analysis of text data:

[0165] The server receives document files and extracts text data using pdfminer (for PDF files) or python-docx (for Word files). The extracted text data is tokenized using the SpaCy library, and specific keywords or phrases are detected and analyzed.

[0166] 3. Text data conversion:

[0167] The server uses natural language processing technology to replace detected keywords and phrases with new expressions according to predefined conversion rules. For example, it might replace "our company" with "manufacturer" and "inquiry" with "contact".

[0168] 4. Reconstruction and output of document files:

[0169] The replaced text data is then reconstructed back into its original format and saved as a PDF or Word file. The user downloads the reconstructed document file from the server via their device.

[0170] 5. Generation and output of audio data:

[0171] The converted text data is then converted into audio data using the Google Text-to-Speech API. This audio data can then be played back by the user on their device.

[0172] Specific example

[0173] For example, if a staff member uploads a document like this:

[0174] We offer a standard warranty period. If you encounter any problems during this warranty period, please contact us.

[0175] This document will be transformed as follows:

[0176] The manufacturer provides a standard warranty period. If any problems occur during this warranty period, please contact us.

[0177] Furthermore, this converted text is also output as audio data.

[0178] Example of a prompt

[0179] The user is a store employee. They uploaded a document file received from the manufacturer. Please translate the content of this document into language that is easy for customers to understand, and also output the content as audio.

[0180] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0181] Step 1:

[0182] The user uploads a document file provided by the manufacturer using their device. The device sends the file to the server regardless of its format (PDF, Word, etc.). The input is the document file, and the output is the file uploaded to the server.

[0183] Step 2:

[0184] The server parses the received document files. Specifically, it uses pdfminer to extract text data from PDF files and python-docx to extract text data from Word files. The input is document files, and the output is text data.

[0185] Step 3:

[0186] The server analyzes the extracted text data and performs tokenization. The SpaCy library is used for this process. Tokenization divides the text data into words and phrases. The input is the text data, and the output is the tokenized data.

[0187] Step 4:

[0188] The server detects specific keywords and phrases from tokenized data. This is done using predefined lists and natural language processing techniques. The input is tokenized data, and the output is the detected keywords and phrases.

[0189] Step 5:

[0190] The server replaces detected keywords and phrases with new expressions according to predefined conversion rules. For example, it replaces "our company" with "manufacturer." The input is the detected keywords and phrases, and the output is the text data after the replacement.

[0191] Step 6:

[0192] The server reconstructs the replaced text data and returns it to the original document format (PDF or Word file). This reconstruction uses the corresponding formatting library. The input is the replaced text data, and the output is the reconstructed document file.

[0193] Step 7:

[0194] The server provides a download link to the terminal for the reconstructed document file, allowing the user to retrieve the file. The input is the reconstructed document file, and the output is the download link.

[0195] Step 8:

[0196] The server converts the converted text data into audio data. It uses the Google Text-to-Speech API to convert the text data into audio data. The input is the replaced text data, and the output is the audio data.

[0197] Step 9:

[0198] The device provides the user with the ability to play audio data. The audio data is streamed and can be listened to by the user. The input is audio data, and the output is audio playback.

[0199] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0200] This invention relates to a system for automatically replacing specific keywords or phrases within a document file, and further incorporates an emotion engine to recognize user emotions and optimize the conversion process. The following describes specific embodiments of the system of this invention.

[0201] The process begins with the user using a terminal to select a document file provided by the manufacturer and uploading it to the system interface. The terminal then sends the uploaded document file to the server.

[0202] The server receives document files and extracts text data using the appropriate method according to the file format. For example, if the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0203] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. Natural language processing techniques may be used for this detection.

[0204] Next, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's emotional state from documents uploaded by the user and their activity history. For example, if the user is in a hurry, the emotion engine will detect this and adjust the settings to expedite the conversion process.

[0205] The server applies predefined conversion rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." It can also dynamically adjust conversion rules based on the user's emotions, as analyzed by the emotion engine. For instance, if the user is calm, it will select more polite language; if they are in a hurry, it will convert to more concise language.

[0206] Next, the server uses a custom translation function to make any other necessary changes to the wording. Here too, the results of the sentiment engine's analysis are reflected. The converted text data is then reconstructed and returned to its original document format. If it was in PDF format, it will be generated again as a PDF file.

[0207] The server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, and the user downloads the converted document file. This file can be used by the reseller.

[0208] As a concrete example, consider a case where the manufacturer provides the following document.

[0209] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0210] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." Additionally, by changing "inquiry" to "contact," the following document is generated:

[0211] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0212] Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is streamlined and supported to ensure a smooth user experience. The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a reseller.

[0213] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can also respond flexibly to the user's emotional state.

[0214] The following describes the processing flow.

[0215] Step 1:

[0216] The user selects a document file provided by the manufacturer using their device and uploads it to the system interface. The uploaded document files can be in formats such as PDF or Word.

[0217] Step 2:

[0218] The terminal sends the uploaded document file to the server. The server receives this file and begins processing it.

[0219] Step 3:

[0220] The server determines the format of the received document file and performs text extraction processing according to that format. For example, if the file is in PDF format, the PDF parsing module is used to extract the text data; if it is in Word format, the Word parsing module is used.

[0221] Step 4:

[0222] The server analyzes the extracted text data and performs tokenization. Tokenization is the process of breaking down text data into words and phrases and identifying each token.

[0223] Step 5:

[0224] The server detects specific keywords and phrases from the tokenized text data. Examples of keywords and phrases detected include "our company" and "contact us."

[0225] Step 6:

[0226] The server uses an emotion engine to recognize the user's emotional state. The emotion engine analyzes the user's facial expressions, voice, and operation history as they interact with the system to estimate their emotions.

[0227] Step 7:

[0228] The server applies predefined conversion rules to the keywords and phrases it detects. For example, it might replace "our company" with "manufacturer." In this process, the sentiment engine may dynamically adjust the conversion rules based on the user's emotions as analyzed. For instance, if the user is in a hurry, the conversion will prioritize speed.

[0229] Step 8:

[0230] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact." The sentiment engine's analysis results are reflected here as well.

[0231] Step 9:

[0232] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated again as a PDF file.

[0233] Step 10:

[0234] The server saves the reconstructed document files and prepares them for users to download. The generated files are provided as download links for easy access by users.

[0235] Step 11:

[0236] A download link appears on the terminal, and the user downloads the converted document file. This file can then be used by the reseller. Providing documents quickly converted to the format the user needs significantly improves operational efficiency.

[0237] As a concrete example, consider the following document provided by the manufacturer:

[0238] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0239] When a user uploads this document to the system, the server extracts the text data and converts "our company" to "manufacturer" and "inquiry" to "contact". Furthermore, if the emotion engine detects that the user is in a hurry, it speeds up the conversion process and generates a document like this:

[0240] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0241] This document file will be saved again in PDF format, and a download link will be provided to the user's device. The user can click the link to download the converted document and use it as a distributor.

[0242] (Example 2)

[0243] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0244] Conventional document conversion systems can replace specific keywords and phrases, but they have a weakness in that they cannot perform dynamic conversions that take into account the user's emotional state. In particular, appropriate text conversion is needed depending on whether the user is in a hurry or calm, and there is a need for a way to do this automatically.

[0245] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0246] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data; means for detecting specific keywords or phrases from the tokenized text data; means for analyzing the user's emotional state; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules, and further means for dynamically adjusting the transformation rules based on the user's emotional state; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This enables flexible and efficient document transformation in response to the user's emotional state.

[0247] A "document file" is a general term for electronic file formats that contain text data. Examples include PDF and Word files.

[0248] "Text data" refers to data that represents textual information as a string of characters.

[0249] "Analysis" is the process of extracting information from document files and identifying and understanding their contents.

[0250] "Tokenization" is the process of breaking down text data into units of words or phrases.

[0251] "Keywords" are words or phrases that are considered particularly important within a document.

[0252] A "phrase" is a unit of language in which two or more words combine to form a single meaning.

[0253] "Emotional state" refers to the user's current emotional and psychological state.

[0254] A "conversion rule" is a predefined set of rules for substituting specific keywords or phrases with other expressions.

[0255] "Dynamic adjustment" means changing settings and behavior in real time according to the situation and conditions.

[0256] "Reconstruction" is the process of returning converted text data to its original document format.

[0257] "Output" refers to the act of providing the user with the processing results.

[0258] This invention is a system that automatically performs the process from uploading a document file to analyzing, converting, and downloading its contents. The system consists of a terminal and a server; the terminal provides an interface for receiving user input, and the server performs the analysis, conversion, and output of the document file.

[0259] First, the user accesses the system interface using a terminal and uploads a document file. Document file formats include PDF and Word. The uploaded file is sent to the server, which temporarily stores it.

[0260] Next, the server analyzes the received document file. The appropriate analysis module is selected depending on the format of the document file. Specifically, PyPDF2 is used for PDF files, and python-docx is used for Word files to extract text data. The extracted text data is tokenized and broken down into words and phrases. Natural language processing libraries such as NLTK and SpaCy are used for this tokenization process.

[0261] The server detects specific keywords and phrases from tokenized text data. Regular expressions and predefined keyword lists are used for detection. At this stage, an emotion engine is utilized to analyze the user's emotional state. The emotion engine uses Hugging Face's Transformers model to identify the emotional state from the user's interaction history and uploaded documents.

[0262] Detected keywords and phrases are replaced with new expressions according to predefined conversion rules. These rules are dynamically adjusted based on the user's emotional state. For example, if the user is in a hurry, the emotion engine detects this and adjusts the settings to expedite the conversion process.

[0263] The replaced text data is then subjected to a custom translation function to modify any other necessary phrasing. This converted text data is then reconstructed and returned to its original document format (such as PDF or Word). Libraries such as ReportLab are used for this reconstruction.

[0264] The reconstructed document file is saved on the server and prepared for user download. A download link is displayed on the terminal, and the user downloads the converted document file. The downloaded file can then be used by the user as a reseller.

[0265] As a concrete example, consider the case where the following document is uploaded by a user.

[0266] "We offer a standard warranty period. To ensure your peace of mind, please direct any inquiries to our support team."

[0267] The server analyzes this document and replaces "our company" with "manufacturer" and "inquiry" with "contact". Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is simplified, and a document like the following is generated.

[0268] "The manufacturer provides a standard warranty period. To ensure your peace of mind, please contact the manufacturer's support team."

[0269] Here are some examples of input prompts for a generative AI model.

[0270] Prompt example:

[0271] "Read the following document and replace specific keywords and phrases (for example, replace "our company" with "manufacturer"). Also, change the wording to be concise if the user is in a hurry, and more polite if they are calm. Document: We offer a standard warranty period. For your peace of mind, please contact our support team. User's emotional state: Urgent Result Document: The manufacturer offers a standard warranty period. For your peace of mind, please contact the manufacturer's support team."

[0272] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can respond flexibly according to the user's emotional state.

[0273] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0274] Step 1:

[0275] The user uploads a document file.

[0276] Input: A document file (such as a PDF or Word document) selected by the user using the terminal interface.

[0277] Specific operation: The user clicks the upload button and uploads the selected document file to the system interface. This operation sends the file data to the server as an HTTP request.

[0278] Output: The document file received by the server.

[0279] Step 2:

[0280] The server receives the document file.

[0281] Input: Uploaded document file.

[0282] Specific operation: The server temporarily stores the received document file in storage. A process to automatically identify the file format (such as PDF or Word) is performed.

[0283] Output: Document file stored in the server.

[0284] Step 3:

[0285] The server extracts the text data of the document file.

[0286] Input: Stored document file.

[0287] Specific operation: The server uses an analysis module (such as PyPDF2 or python-docx) according to the format of the document file to extract the text data. In the case of PDF, PyPDF2 is used to extract the text page by page, and in the case of a Word file, python-docx is used to extract the text.

[0288] Output: Extracted text data.

[0289] Step 4:

[0290] The server tokenizes the text data.

[0291] Input: Extracted text data.

[0292] Specific operation: The server uses natural language processing libraries such as NLTK or SpaCy to tokenize the text data. By tokenization, the text is decomposed into units of words or phrases. Specifically, the SpaCy's nlp function is used to tokenize the text data.

[0293] Output: Tokenized text data.

[0294] Step 5:

[0295] The server detects specific keywords or phrases.

[0296] Input: Tokenized text data.

[0297] Specific operation: The server uses regular expressions or predefined keyword lists to detect specific keywords or phrases from text data. For example, to detect the keyword "our company," it uses the regular expression findall function.

[0298] Output: Detected keywords and phrases.

[0299] Step 6:

[0300] The server analyzes the user's emotional state.

[0301] Input: User activity history and text data of uploaded documents.

[0302] Specific operation: The server uses the Hugging Face Transformers model to analyze emotional states. Specifically, it uses the Transformers pipeline function to perform sentiment analysis.

[0303] Output: Analyzed user emotional state.

[0304] Step 7:

[0305] The server replaces keywords and phrases.

[0306] Input: Detected keywords and phrases, and the analyzed user's emotional state.

[0307] Specific operation: The server replaces keywords and phrases with new expressions according to predefined conversion rules. Also, it dynamically adjusts the replacement rules based on the user's emotional state. For example, it may replace "our company" with "the manufacturer" or "inquiry" with "contact".

[0308] Output: The text data after replacement.

[0309] Step 8:

[0310] The server applies custom translation.

[0311] Input: The text data after replacement.

[0312] Specific operation: The server further adjusts the text data appropriately using a custom translation function. For example, if the user is in a hurry, it changes "inquiry" to "contact" to make the text more concise.

[0313] Output: The text data after custom translation is applied.

[0314] Step 9:

[0315] The server reconstructs the converted document.

[0316] Input: The text data after custom translation is applied.

[0317] Specific operation: The server reconstructs the text data into the original document format (such as PDF or Word). In the case of PDF, it uses the ReportLab library for reconstruction.

[0318] Output: The reconstructed document file.

[0319] Step 10:

[0320] The server generates a download link.

[0321] Input: Reconstructed document file.

[0322] Specific operation: The server saves the reconstructed document file and generates a URL so that the user can download the file. This URL is sent to the user's device.

[0323] Output: The generated download link.

[0324] Step 11:

[0325] The user downloads the converted document.

[0326] Input: Download link provided by the server.

[0327] Specific action: The user clicks the download link displayed on the terminal and downloads the converted document file.

[0328] Output: Document files downloaded by the user.

[0329] (Application Example 2)

[0330] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0331] Conventional document conversion systems lacked flexibility and usability because they replaced keywords and phrases based on static rules without considering user sentiment. Furthermore, especially in virtual stores, the accurate and effective conversion of product descriptions is crucial, but current systems struggled to meet this requirement.

[0332] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0333] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for analyzing the user's emotions and dynamically adjusting the document transformation rules based on those emotions; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to optimize the document transformation process according to the user's emotional state, enabling it to be performed more flexibly and efficiently.

[0334] A "document file" is an electronically stored document containing text data.

[0335] "Text data" refers to data that is stored as character information.

[0336] "Analysis" is the process of breaking down data and analyzing its individual elements.

[0337] "Keywords" are particularly important words or phrases within a document.

[0338] A "phrase" is a combination of multiple words that are grammatically or semantically coherent.

[0339] "Conversion rules" are a set of predefined rules for substituting specific keywords or phrases with other expressions.

[0340] "User emotions" refers to the emotional state a user exhibits while using the system, including states such as being in a hurry or being calm.

[0341] "Emotional analysis" refers to the process of determining a user's emotional state based on their actions and input data.

[0342] "Substitution" is the process of replacing one word or phrase with another.

[0343] "Dynamic adjustment" refers to the process of changing conversion rules in real time according to conditions such as the user's emotional state.

[0344] "Reconstruction" refers to the process of converting replaced text data back into a new document file format.

[0345] A "generative AI model" is an artificial intelligence built using deep learning that performs text generation and analysis according to various tasks.

[0346] A "prompt statement" is a sentence or phrase used as input in natural language processing.

[0347] To realize this invention, the following system configuration and processing are necessary. The system is designed so that users can operate it using a terminal such as a smartphone or PC.

[0348] A user accesses the system interface and uploads a document file. This document file is sent to the server. The server extracts the text data using the appropriate module depending on the format of the uploaded document file. For example, a PDF file uses the PDF parsing module, and a Word file uses the Word parsing module.

[0349] The analyzed text data is tokenized using natural language processing techniques. Specifically, it is broken down into words and phrases, and each is processed to identify it. Next, the server detects specific keywords and phrases from the extracted tokenized data. Transformation rules are predefined, and the server uses these rules to detect keywords and phrases.

[0350] Next, the server uses an emotion engine to analyze the user's emotions. The emotion engine uses a generative AI model to analyze the user's emotional state. Based on this emotion analysis, the conversion rules are dynamically adjusted. For example, if the user is in a hurry, it generates concise sentences; if they are calm, it generates polite sentences.

[0351] The server replaces detected keywords and phrases with new expressions according to pre-defined conversion rules. The sentiment engine's analysis results are also reflected in this process. The converted text data is then reconstructed and returned to its original format. For example, if it was in PDF format, it will be generated again as a PDF file.

[0352] Finally, the server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, allowing the user to download the converted document file.

[0353] The hardware and software used include the following:

[0354] Hardware: Smartphones (iOS, Android®), PC

[0355] Software: Python, Hugging Face transformers library, TextBlob library

[0356] As a concrete example, consider the case where a user uploads the following document:

[0357] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0358] The server analyzes this document and transforms specific keywords and phrases as follows:

[0359] The virtual store offers a standard warranty period. For your peace of mind, please contact the virtual store's support team.

[0360] Furthermore, prompts are used to analyze user sentiment. Examples of specific prompts are shown below:

[0361] Text for when the user is in a hurry

[0362] By inputting this prompt into the AI ​​model, the user's emotional state of being in a hurry is analyzed. Based on this information, the server performs appropriate text conversion.

[0363] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0364] Step 1:

[0365] The user uploads a document file using their terminal. The input is the document file itself, which is sent to the server. The output is the document file stored on the server.

[0366] Step 2:

[0367] The server extracts text data in an appropriate manner depending on the format of the uploaded document file. The input is the document file format (PDF, Word, etc.), and the data processing involves using PDF parsing modules or Word parsing modules to obtain the text data. The output is the extracted text data.

[0368] Step 3:

[0369] The server analyzes and tokenizes the extracted text data. The input is the extracted text data, and the data processing utilizes natural language processing techniques to break down the data into words and phrases. The output is the tokenized text data.

[0370] Step 4:

[0371] The server detects specific keywords and phrases from tokenized text data. The input is tokenized text data, and the data operation involves detecting keywords and phrases based on predefined transformation rules. The output is the detected keywords and phrases.

[0372] Step 5:

[0373] The server uses an emotion engine (generative AI model) to analyze the user's emotions. The input consists of the user's operation history and uploaded documents. As part of the data processing, prompt sentences are generated and input into the generative AI model to analyze the user's emotional state. The output is the result of the user's emotion analysis.

[0374] Step 6:

[0375] The server replaces detected keywords and phrases with new expressions based on predefined transformation rules. The input consists of detected keywords and phrases, as well as the user sentiment analysis results. The server performs the replacement process based on predefined transformation rules as a data calculation. The output is the replaced text data.

[0376] Step 7:

[0377] The server reconstructs the replaced text data and converts it back to the original document format. The input is the replaced text data, and the data processing involves converting it back to the original document format (PDF, Word, etc.). The output is the reconstructed document file.

[0378] Step 8:

[0379] The server saves the reconstructed document file and prepares it for the user to download. The input is the reconstructed document file, and a download link to the user's terminal is generated. The output is the download link provided to the user.

[0380] Specific operations include: users dragging and dropping files to upload them; extracting and analyzing text data on the server; analyzing responses to prompts using AI models generated by the sentiment engine; dynamic text replacement processing according to transformation rules; and providing reconstructed documents via generated download links.

[0381] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0382] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0383] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0384] [Second Embodiment]

[0385] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0386] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0387] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0388] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0389] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0390] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0391] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0392] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0393] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0394] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0395] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0396] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0397] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions for use by distributors. The specific procedures and processes for implementing this system are detailed below.

[0398] This system begins with the user uploading document files provided by the manufacturer. The user selects the document files using a terminal and uploads them to the system interface. The terminal then sends the uploaded document files to the server.

[0399] The server receives document files and extracts text data using the appropriate method depending on the file format. If the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0400] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. These keywords and phrases are generally expressions such as "our company" and "standard warranty period."

[0401] The server applies predefined translation rules to detected keywords and phrases. Specifically, it performs processes such as replacing "our company" with "manufacturer." Furthermore, the server uses a custom translation function to make other necessary changes to the wording. This custom translation function utilizes natural language processing technology to perform appropriate translations according to the context.

[0402] The converted text data is reconstructed and returned to its original document format. For example, if the original format was PDF, the converted text data is converted back into a PDF file. The server saves the reconstructed document file and prepares it for later download by the user.

[0403] Finally, a download link appears on the device, and the user downloads the converted document file. In this way, the user can avoid manual errors and perform document conversion efficiently and accurately.

[0404] Specific example

[0405] For example, suppose the manufacturer provides the following document:

[0406] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0407] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." It also changes the phrase "inquiry" to "contact," resulting in the following document:

[0408] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0409] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a distributor.

[0410] The following describes the processing flow.

[0411] Step 1:

[0412] The user selects a document file provided by the manufacturer using a terminal and uploads it to the system interface.

[0413] Step 2:

[0414] The device sends the uploaded document file to the server.

[0415] Step 3:

[0416] The server reads the received document file as text data. If the document is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0417] Step 4:

[0418] The server tokenizes the extracted text data. Tokenization is the process of breaking down text data into words and phrases and identifying each one.

[0419] Step 5:

[0420] The server detects specific keywords or phrases from tokenized text data. This may involve using natural language processing techniques.

[0421] Step 6:

[0422] The server applies predefined conversion rules to the keywords and phrases it detects. Specifically, it performs processes such as replacing "our company" with "manufacturer."

[0423] Step 7:

[0424] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact".

[0425] Step 8:

[0426] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated as a PDF file again.

[0427] Step 9:

[0428] The server saves the reconstructed document files and prepares them for users to download.

[0429] Step 10:

[0430] A download link will appear on the device, and the user will download the converted document file. This file can then be used by the distributor.

[0431] (Example 1)

[0432] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0433] Traditional manual document conversion processes are time-consuming, labor-intensive, and prone to errors. Furthermore, consistent and improper replacement of specific keywords and phrases is often lacking, making them inefficient, especially when processing large volumes of documents. To address these challenges, a system is needed that automatically and accurately converts specific keywords and phrases within document files.

[0434] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0435] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data into word or phrase units; means for detecting specific keywords or phrases from the tokenized text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to automatically detect specific keywords or phrases and replace them accurately and efficiently.

[0436] A "document file" is an electronic file that contains text data, and includes formats such as PDF and Word files.

[0437] "Text data" refers to string data extracted or parsed from document files, and includes words and phrases.

[0438] "Tokenization" is the process of dividing text data into words or phrases and identifying each one individually.

[0439] A "keyword or phrase" is a word or phrase that has a specific meaning and is an important element within a document.

[0440] "Conversion rules" refer to predefined rules for replacing specific keywords or phrases with new expressions.

[0441] "Reconstruction" is the process of restoring the replaced text data to its original document format.

[0442] "Output" refers to providing the user with the completed document file.

[0443] This invention relates to a system for automatically converting specific keywords or phrases within document files. This significantly reduces manual data entry and allows users to efficiently and accurately convert documents. The specific procedures and processes for implementing this system are detailed below.

[0444] This system includes the user's terminal, a server, appropriate analysis modules, and natural language processing technology. The user selects a document file provided by the manufacturer from their terminal and uploads it to the system. The terminal then sends the uploaded document file to the server.

[0445] The server receives document files in formats such as PDF and Word, selects the appropriate analysis module, and extracts the text data. For example, it uses the PDFLib or pdfminer.six library for PDF files and the python-docx library for Word files.

[0446] The extracted text data is tokenized by the server. Natural language processing libraries (such as NLTK or spaCy) are used for tokenization, dividing the text data into words and phrases. The server then detects specific keywords and phrases from the tokenized text. A predefined keyword list is used for this detection.

[0447] The server applies pre-configured translation rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." Furthermore, it uses custom translation functions as needed to perform contextually appropriate translations using natural language processing techniques (such as BERT or GPT).

[0448] The converted text data is then reconstructed by the server into its original document format. For PDFs, the PDFLib library is used to generate a new PDF file, and for Word files, the python-docx library is used to generate a new Word file.

[0449] Finally, the server saves the reconstructed document file and provides a download link to the user's terminal. The user can then use their terminal to download the converted document file via this link.

[0450] As a concrete example, consider a case where the manufacturer provides the following document:

[0451] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0452] When a user uploads this document to the system, the server extracts the text data and converts the keyword "our company" to "manufacturer" and the phrase "inquiry" to "contact". As a result, a document like the following is generated:

[0453] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0454] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can download the converted document using this link.

[0455] Examples of prompts to input into a generative AI model are as follows:

[0456] Please translate the phrases "our company," "standard warranty period," and "inquiries" into language suitable for sales companies.

[0457] In this way, the system enables efficient and accurate conversion of specific keywords and phrases within a document.

[0458] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0459] Step 1:

[0460] The user selects a document file provided by the manufacturer and uploads it from their device using the system interface. The device processes this upload request and sends the document file to the server.

[0461] Input: Document file selected from the user's terminal (PDF or Word format)

[0462] Output: Document file sent to the server

[0463] Step 2:

[0464] The server receives document files sent from the terminal. It recognizes the format of the document file and selects the appropriate parsing module (PDF parsing module or Word parsing module). For PDF files, it uses the PDFLib or pdfminer.six library to extract text data; for Word files, it uses the python-docx library.

[0465] Input: Document file sent to the server

[0466] Output: Extracted text data

[0467] Specific operation: The PDF parsing module uses PDFLib, and the Word parsing module uses python-docx to extract text data from documents.

[0468] Step 3:

[0469] The server tokenizes the extracted text data. This tokenization uses natural language processing libraries (such as NLTK or spaCy) to divide the text into words and phrases.

[0470] Input: Extracted text data

[0471] Output: Tokenized text data

[0472] Specific operation: The text is split into words and phrases using NLTK's word_tokenize function or spaCy's nlp function.

[0473] Step 4:

[0474] The server detects specific keywords or phrases from tokenized text data. It uses a predefined keyword list to check whether each token is included in the list.

[0475] Input: Tokenized text data

[0476] Output: Detected keywords and phrases

[0477] Specific operation: Loop through the tokenized text data and verify whether each token is present in the keyword list.

[0478] Step 5:

[0479] The server applies predefined conversion rules to detected keywords and phrases. For example, it converts "our company" to "manufacturer." For more complex conversions, it utilizes natural language processing techniques (such as BERT and GPT).

[0480] Input: Detected keywords or phrases

[0481] Output: Converted text data

[0482] Specific operation: Basic substitutions are performed using Python's str.replace function, and context-aware transformations are performed using a transformer model.

[0483] Step 6:

[0484] The server reconstructs the converted text data into the original document format. For PDFs, it uses the PDFLib library to generate a new PDF file, and for Word files, it uses the python-docx library to generate a new Word file.

[0485] Input: Converted text data

[0486] Output: Reconstructed document file (PDF or Word format)

[0487] Specific operation: Insert text data into a new document object and format the entire document.

[0488] Step 7:

[0489] The server saves the reconstructed document file and displays a download link on the user's terminal. The user can then download the converted document file from their terminal via this link.

[0490] Input: Reconstructed document file

[0491] Output: Download link displayed on the user's terminal

[0492] Specific operation: The server saves the document file to the storage directory, generates a download link containing the path, and notifies the user.

[0493] (Application Example 1)

[0494] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0495] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions suitable for sales companies, and further outputs the converted content as audio. Conventional systems require manual modification of the wording in document files, which is inefficient and prone to errors. Furthermore, because they are limited to visual information, it is difficult for visually impaired individuals to obtain the information.

[0496] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0497] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined conversion rules; means for reconstructing the replaced text data as a document file; means for outputting the reconstructed document file; means for converting the converted text data into audio data; and means for outputting the converted audio data. This makes it possible to automatically and efficiently convert the expression of a document file for sales companies and further output its contents as audio.

[0498] A "document file" is a file containing text or image data stored in an electronic format.

[0499] "Text data" refers to information expressed as a string of characters, specifically the content extracted from a document file.

[0500] "Analyzing" means breaking down given data and processing it in order to understand its meaning and structure.

[0501] A "keyword or phrase" refers to a word or short sentence with a specific meaning, representing an important part of a document.

[0502] "Detecting" means finding data that has specific conditions or characteristics.

[0503] "Conversion rules" are pre-defined guidelines or rules for replacing specific keywords or phrases with other expressions.

[0504] "Substitution" means replacing one element with another.

[0505] "Reconstructing" means putting data back together to return it to its original state or format.

[0506] "Outputting" means providing processed data in a format that can be used by the user.

[0507] "Audio data" refers to digital data containing audio information, in a format that can be played back by an audio playback device.

[0508] This invention is a system for efficiently converting the expression of business documents for specific purposes, particularly for generating appropriate expressions for customers and outputting them as audio. The following describes specific embodiments for carrying out the invention.

[0509] Program Overview

[0510] The server system and terminals operate in coordination. The system according to the present invention includes the following main means:

[0511] A method for receiving document files and analyzing them as text data.

[0512] Means for detecting specific keywords or phrases from analyzed text data

[0513] A means of replacing detected keywords or phrases with new expressions according to predefined transformation rules.

[0514] A method for reconstructing the replaced text data as a document file.

[0515] Means for outputting a reconstructed document file

[0516] A means of converting converted text data into audio data.

[0517] Means for outputting converted audio data

[0518] Hardware and software to be used

[0519] The server is hosted using Google Cloud Platform, and utilizes SpaCy for natural language processing and the Google Text-to-Speech API for speech synthesis. The device is a smartphone or tablet with an application for uploading document files installed.

[0520] Data processing and data calculation

[0521] 1. Upload document files:

[0522] The user uploads document files (such as PDFs or Word documents) provided by the manufacturer to the system interface using their device. The device then sends the uploaded document files to the server.

[0523] 2. Extraction and analysis of text data:

[0524] The server receives document files and extracts text data using pdfminer (for PDF files) or python-docx (for Word files). The extracted text data is tokenized using the SpaCy library, and specific keywords or phrases are detected and analyzed.

[0525] 3. Text data conversion:

[0526] The server uses natural language processing technology to replace detected keywords and phrases with new expressions according to predefined conversion rules. For example, it might replace "our company" with "manufacturer" and "inquiry" with "contact".

[0527] 4. Reconstruction and output of document files:

[0528] The replaced text data is then reconstructed back into its original format and saved as a PDF or Word file. The user downloads the reconstructed document file from the server via their device.

[0529] 5. Generation and output of audio data:

[0530] The converted text data is then converted into audio data using the Google Text-to-Speech API. This audio data can then be played back by the user on their device.

[0531] Specific example

[0532] For example, if a staff member uploads a document like this:

[0533] We offer a standard warranty period. If you encounter any problems during this warranty period, please contact us.

[0534] This document will be transformed as follows:

[0535] The manufacturer provides a standard warranty period. If any problems occur during this warranty period, please contact us.

[0536] Furthermore, this converted text is also output as audio data.

[0537] Example of a prompt

[0538] The user is a store employee. They uploaded a document file received from the manufacturer. Please translate the content of this document into language that is easy for customers to understand, and also output the content as audio.

[0539] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0540] Step 1:

[0541] The user uploads a document file provided by the manufacturer using their device. The device sends the file to the server regardless of its format (PDF, Word, etc.). The input is the document file, and the output is the file uploaded to the server.

[0542] Step 2:

[0543] The server parses the received document files. Specifically, it uses pdfminer to extract text data from PDF files and python-docx to extract text data from Word files. The input is document files, and the output is text data.

[0544] Step 3:

[0545] The server analyzes the extracted text data and performs tokenization. The SpaCy library is used for this process. Tokenization divides the text data into words and phrases. The input is the text data, and the output is the tokenized data.

[0546] Step 4:

[0547] The server detects specific keywords and phrases from tokenized data. This is done using predefined lists and natural language processing techniques. The input is tokenized data, and the output is the detected keywords and phrases.

[0548] Step 5:

[0549] The server replaces detected keywords and phrases with new expressions according to predefined conversion rules. For example, it replaces "our company" with "manufacturer." The input is the detected keywords and phrases, and the output is the text data after the replacement.

[0550] Step 6:

[0551] The server reconstructs the replaced text data and returns it to the original document format (PDF or Word file). This reconstruction uses the corresponding formatting library. The input is the replaced text data, and the output is the reconstructed document file.

[0552] Step 7:

[0553] The server provides a download link to the terminal for the reconstructed document file, allowing the user to retrieve the file. The input is the reconstructed document file, and the output is the download link.

[0554] Step 8:

[0555] The server converts the converted text data into audio data. It uses the Google Text-to-Speech API to convert the text data into audio data. The input is the replaced text data, and the output is the audio data.

[0556] Step 9:

[0557] The device provides the user with the ability to play audio data. The audio data is streamed and can be listened to by the user. The input is audio data, and the output is audio playback.

[0558] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0559] This invention relates to a system for automatically replacing specific keywords or phrases within a document file, and further incorporates an emotion engine to recognize user emotions and optimize the conversion process. The following describes specific embodiments of the system of this invention.

[0560] The process begins with the user using a terminal to select a document file provided by the manufacturer and uploading it to the system interface. The terminal then sends the uploaded document file to the server.

[0561] The server receives document files and extracts text data using the appropriate method according to the file format. For example, if the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0562] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. Natural language processing techniques may be used for this detection.

[0563] Next, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's emotional state from documents uploaded by the user and their activity history. For example, if the user is in a hurry, the emotion engine will detect this and adjust the settings to expedite the conversion process.

[0564] The server applies predefined conversion rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." It can also dynamically adjust conversion rules based on the user's emotions, as analyzed by the emotion engine. For instance, if the user is calm, it will select more polite language; if they are in a hurry, it will convert to more concise language.

[0565] Next, the server uses a custom translation function to make any other necessary changes to the wording. Here too, the results of the sentiment engine's analysis are reflected. The converted text data is then reconstructed and returned to its original document format. If it was in PDF format, it will be generated again as a PDF file.

[0566] The server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, and the user downloads the converted document file. This file can be used by the reseller.

[0567] As a concrete example, consider a case where the manufacturer provides the following document.

[0568] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0569] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." Additionally, by changing "inquiry" to "contact," the following document is generated:

[0570] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0571] Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is streamlined and supported to ensure a smooth user experience. The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a reseller.

[0572] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can also respond flexibly to the user's emotional state.

[0573] The following describes the processing flow.

[0574] Step 1:

[0575] The user selects a document file provided by the manufacturer using their device and uploads it to the system interface. The uploaded document files can be in formats such as PDF or Word.

[0576] Step 2:

[0577] The terminal sends the uploaded document file to the server. The server receives this file and begins processing it.

[0578] Step 3:

[0579] The server determines the format of the received document file and performs text extraction processing according to that format. For example, if the file is in PDF format, the PDF parsing module is used to extract the text data; if it is in Word format, the Word parsing module is used.

[0580] Step 4:

[0581] The server analyzes the extracted text data and performs tokenization. Tokenization is the process of breaking down text data into words and phrases and identifying each token.

[0582] Step 5:

[0583] The server detects specific keywords and phrases from the tokenized text data. Examples of keywords and phrases detected include "our company" and "contact us."

[0584] Step 6:

[0585] The server uses an emotion engine to recognize the user's emotional state. The emotion engine analyzes the user's facial expressions, voice, and operation history as they interact with the system to estimate their emotions.

[0586] Step 7:

[0587] The server applies predefined conversion rules to the keywords and phrases it detects. For example, it might replace "our company" with "manufacturer." In this process, the sentiment engine may dynamically adjust the conversion rules based on the user's emotions as analyzed. For instance, if the user is in a hurry, the conversion will prioritize speed.

[0588] Step 8:

[0589] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact." The sentiment engine's analysis results are reflected here as well.

[0590] Step 9:

[0591] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated again as a PDF file.

[0592] Step 10:

[0593] The server saves the reconstructed document files and prepares them for users to download. The generated files are provided as download links for easy access by users.

[0594] Step 11:

[0595] A download link appears on the terminal, and the user downloads the converted document file. This file can then be used by the reseller. Providing documents quickly converted to the format the user needs significantly improves operational efficiency.

[0596] As a concrete example, consider the following document provided by the manufacturer:

[0597] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0598] When a user uploads this document to the system, the server extracts the text data and converts "our company" to "manufacturer" and "inquiry" to "contact". Furthermore, if the emotion engine detects that the user is in a hurry, it speeds up the conversion process and generates a document like this:

[0599] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0600] This document file will be saved again in PDF format, and a download link will be provided to the user's device. The user can click the link to download the converted document and use it as a distributor.

[0601] (Example 2)

[0602] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0603] Conventional document conversion systems can replace specific keywords and phrases, but they have a weakness in that they cannot perform dynamic conversions that take into account the user's emotional state. In particular, appropriate text conversion is needed depending on whether the user is in a hurry or calm, and there is a need for a way to do this automatically.

[0604] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0605] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data; means for detecting specific keywords or phrases from the tokenized text data; means for analyzing the user's emotional state; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules, and further means for dynamically adjusting the transformation rules based on the user's emotional state; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This enables flexible and efficient document transformation in response to the user's emotional state.

[0606] A "document file" is a general term for electronic file formats that contain text data. Examples include PDF and Word files.

[0607] "Text data" refers to data that represents textual information as a string of characters.

[0608] "Analysis" is the process of extracting information from document files and identifying and understanding their contents.

[0609] "Tokenization" is the process of breaking down text data into units of words or phrases.

[0610] "Keywords" are words or phrases that are considered particularly important within a document.

[0611] A "phrase" is a unit of language in which two or more words combine to form a single meaning.

[0612] "Emotional state" refers to the user's current emotional and psychological state.

[0613] A "conversion rule" is a predefined set of rules for substituting specific keywords or phrases with other expressions.

[0614] "Dynamic adjustment" means changing settings and behavior in real time according to the situation and conditions.

[0615] "Reconstruction" is the process of returning converted text data to its original document format.

[0616] "Output" refers to the act of providing the user with the processing results.

[0617] This invention is a system that automatically performs the process from uploading a document file to analyzing, converting, and downloading its contents. The system consists of a terminal and a server; the terminal provides an interface for receiving user input, and the server performs the analysis, conversion, and output of the document file.

[0618] First, the user accesses the system interface using a terminal and uploads a document file. Document file formats include PDF and Word. The uploaded file is sent to the server, which temporarily stores it.

[0619] Next, the server analyzes the received document file. The appropriate analysis module is selected depending on the format of the document file. Specifically, PyPDF2 is used for PDF files, and python-docx is used for Word files to extract text data. The extracted text data is tokenized and broken down into words and phrases. Natural language processing libraries such as NLTK and SpaCy are used for this tokenization process.

[0620] The server detects specific keywords and phrases from tokenized text data. Regular expressions and predefined keyword lists are used for detection. At this stage, an emotion engine is utilized to analyze the user's emotional state. The emotion engine uses Hugging Face's Transformers model to identify the emotional state from the user's interaction history and uploaded documents.

[0621] Detected keywords and phrases are replaced with new expressions according to predefined conversion rules. These rules are dynamically adjusted based on the user's emotional state. For example, if the user is in a hurry, the emotion engine detects this and adjusts the settings to expedite the conversion process.

[0622] The replaced text data is then subjected to a custom translation function to modify any other necessary phrasing. This converted text data is then reconstructed and returned to its original document format (such as PDF or Word). Libraries such as ReportLab are used for this reconstruction.

[0623] The reconstructed document file is saved on the server and prepared for user download. A download link is displayed on the terminal, and the user downloads the converted document file. The downloaded file can then be used by the user as a reseller.

[0624] As a concrete example, consider the case where the following document is uploaded by a user.

[0625] "We offer a standard warranty period. To ensure your peace of mind, please direct any inquiries to our support team."

[0626] The server analyzes this document and replaces "our company" with "manufacturer" and "inquiry" with "contact". Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is simplified, and a document like the following is generated.

[0627] "The manufacturer provides a standard warranty period. To ensure your peace of mind, please contact the manufacturer's support team."

[0628] Here are some examples of input prompts for a generative AI model.

[0629] Prompt example:

[0630] "Read the following document and replace specific keywords and phrases (for example, replace "our company" with "manufacturer"). Also, change the wording to be concise if the user is in a hurry, and more polite if they are calm. Document: We offer a standard warranty period. For your peace of mind, please contact our support team. User's emotional state: Urgent Result Document: The manufacturer offers a standard warranty period. For your peace of mind, please contact the manufacturer's support team."

[0631] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can respond flexibly according to the user's emotional state.

[0632] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0633] Step 1:

[0634] The user uploads a document file.

[0635] Input: A document file (such as a PDF or Word document) selected by the user using the terminal interface.

[0636] Specific operation: The user clicks the upload button and uploads the selected document file to the system interface. This operation sends the file data to the server as an HTTP request.

[0637] Output: The document file received by the server.

[0638] Step 2:

[0639] The server receives the document file.

[0640] Input: Uploaded document file.

[0641] Specific operation: The server temporarily stores the received document file in storage. The system automatically identifies the file format (PDF or Word).

[0642] Output: Document files saved on the server.

[0643] Step 3:

[0644] The server extracts the text data from the document file.

[0645] Input: Saved document file.

[0646] Specific operation: The server extracts text data using an analysis module (such as PyPDF2 or python-docx) appropriate for the document file format. For PDFs, PyPDF2 is used to extract text page by page, and for Word files, python-docx is used to extract text.

[0647] Output: Extracted text data.

[0648] Step 4:

[0649] The server tokenizes the text data.

[0650] Input: Extracted text data.

[0651] Specific operation: The server uses natural language processing libraries such as NLTK and SpaCy to tokenize text data. Tokenization breaks down sentences into units of words and phrases. Specifically, the SpaCy nlp function is used to tokenize text data.

[0652] Output: Tokenized text data.

[0653] Step 5:

[0654] The server detects specific keywords or phrases.

[0655] Input: Tokenized text data.

[0656] Specific operation: The server uses regular expressions or predefined keyword lists to detect specific keywords or phrases from text data. For example, to detect the keyword "our company," it uses the regular expression findall function.

[0657] Output: Detected keywords and phrases.

[0658] Step 6:

[0659] The server analyzes the user's emotional state.

[0660] Input: User activity history and text data of uploaded documents.

[0661] Specific operation: The server uses the Hugging Face Transformers model to analyze emotional states. Specifically, it uses the Transformers pipeline function to perform sentiment analysis.

[0662] Output: Analyzed user emotional state.

[0663] Step 7:

[0664] The server replaces keywords and phrases.

[0665] Input: Detected keywords and phrases, and the analyzed user's emotional state.

[0666] Specific operation: The server replaces keywords and phrases with new expressions according to predefined conversion rules. It also dynamically adjusts the replacement rules based on the user's emotional state. For example, it may replace "our company" with "manufacturer" or "inquiry" with "contact".

[0667] Output: Text data after replacement.

[0668] Step 8:

[0669] The server applies the custom translation.

[0670] Input: The text data to be replaced.

[0671] Specific operation: The server uses a custom translation function to further adjust the text data as needed. For example, if the user is in a hurry, it might change "Contact Us" to "Contact Us" to make the text more concise.

[0672] Output: Text data after custom translation has been applied.

[0673] Step 9:

[0674] The server reconstructs the converted document.

[0675] Input: Text data after custom translation has been applied.

[0676] Specific operation: The server reconstructs the text data into its original document format (PDF or Word). In the case of PDF, it uses the ReportLab library for reconstruction.

[0677] Output: Reconstructed document file.

[0678] Step 10:

[0679] The server generates a download link.

[0680] Input: Reconstructed document file.

[0681] Specific operation: The server saves the reconstructed document file and generates a URL so that the user can download the file. This URL is sent to the user's device.

[0682] Output: The generated download link.

[0683] Step 11:

[0684] The user downloads the converted document.

[0685] Input: Download link provided by the server.

[0686] Specific action: The user clicks the download link displayed on the terminal and downloads the converted document file.

[0687] Output: Document files downloaded by the user.

[0688] (Application Example 2)

[0689] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0690] Conventional document conversion systems lacked flexibility and usability because they replaced keywords and phrases based on static rules without considering user sentiment. Furthermore, especially in virtual stores, the accurate and effective conversion of product descriptions is crucial, but current systems struggled to meet this requirement.

[0691] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0692] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for analyzing the user's emotions and dynamically adjusting the document transformation rules based on those emotions; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to optimize the document transformation process according to the user's emotional state, enabling it to be performed more flexibly and efficiently.

[0693] A "document file" is an electronically stored document containing text data.

[0694] "Text data" refers to data that is stored as character information.

[0695] "Analysis" is the process of breaking down data and analyzing its individual elements.

[0696] "Keywords" are particularly important words or phrases within a document.

[0697] A "phrase" is a combination of multiple words that are grammatically or semantically coherent.

[0698] "Conversion rules" are a set of predefined rules for substituting specific keywords or phrases with other expressions.

[0699] "User emotions" refers to the emotional state a user exhibits while using the system, including states such as being in a hurry or being calm.

[0700] "Emotional analysis" refers to the process of determining a user's emotional state based on their actions and input data.

[0701] "Substitution" is the process of replacing one word or phrase with another.

[0702] "Dynamic adjustment" refers to the process of changing conversion rules in real time according to conditions such as the user's emotional state.

[0703] "Reconstruction" refers to the process of converting replaced text data back into a new document file format.

[0704] A "generative AI model" is an artificial intelligence built using deep learning that performs text generation and analysis according to various tasks.

[0705] A "prompt statement" is a sentence or phrase used as input in natural language processing.

[0706] To realize this invention, the following system configuration and processing are necessary. The system is designed so that users can operate it using a terminal such as a smartphone or PC.

[0707] A user accesses the system interface and uploads a document file. This document file is sent to the server. The server extracts the text data using the appropriate module depending on the format of the uploaded document file. For example, a PDF file uses the PDF parsing module, and a Word file uses the Word parsing module.

[0708] The analyzed text data is tokenized using natural language processing techniques. Specifically, it is broken down into words and phrases, and each is processed to identify it. Next, the server detects specific keywords and phrases from the extracted tokenized data. Transformation rules are predefined, and the server uses these rules to detect keywords and phrases.

[0709] Next, the server uses an emotion engine to analyze the user's emotions. The emotion engine uses a generative AI model to analyze the user's emotional state. Based on this emotion analysis, the conversion rules are dynamically adjusted. For example, if the user is in a hurry, it generates concise sentences; if they are calm, it generates polite sentences.

[0710] The server replaces detected keywords and phrases with new expressions according to pre-defined conversion rules. The sentiment engine's analysis results are also reflected in this process. The converted text data is then reconstructed and returned to its original format. For example, if it was in PDF format, it will be generated again as a PDF file.

[0711] Finally, the server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, allowing the user to download the converted document file.

[0712] The hardware and software used include the following:

[0713] Hardware: Smartphones (iOS, Android), PC

[0714] Software: Python, Hugging Face transformers library, TextBlob library

[0715] As a concrete example, consider the case where a user uploads the following document:

[0716] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0717] The server analyzes this document and transforms specific keywords and phrases as follows:

[0718] The virtual store offers a standard warranty period. For your peace of mind, please contact the virtual store's support team.

[0719] Furthermore, prompts are used to analyze user sentiment. Examples of specific prompts are shown below:

[0720] Text for when the user is in a hurry

[0721] By inputting this prompt into the AI ​​model, the user's emotional state of being in a hurry is analyzed. Based on this information, the server performs appropriate text conversion.

[0722] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0723] Step 1:

[0724] The user uploads a document file using their terminal. The input is the document file itself, which is sent to the server. The output is the document file stored on the server.

[0725] Step 2:

[0726] The server extracts text data in an appropriate manner depending on the format of the uploaded document file. The input is the document file format (PDF, Word, etc.), and the data processing involves using PDF parsing modules or Word parsing modules to obtain the text data. The output is the extracted text data.

[0727] Step 3:

[0728] The server analyzes and tokenizes the extracted text data. The input is the extracted text data, and the data processing utilizes natural language processing techniques to break down the data into words and phrases. The output is the tokenized text data.

[0729] Step 4:

[0730] The server detects specific keywords and phrases from tokenized text data. The input is tokenized text data, and the data operation involves detecting keywords and phrases based on predefined transformation rules. The output is the detected keywords and phrases.

[0731] Step 5:

[0732] The server uses an emotion engine (generative AI model) to analyze the user's emotions. The input consists of the user's operation history and uploaded documents. As part of the data processing, prompt sentences are generated and input into the generative AI model to analyze the user's emotional state. The output is the result of the user's emotion analysis.

[0733] Step 6:

[0734] The server replaces detected keywords and phrases with new expressions based on predefined transformation rules. The input consists of detected keywords and phrases, as well as the user sentiment analysis results. The server performs the replacement process based on predefined transformation rules as a data calculation. The output is the replaced text data.

[0735] Step 7:

[0736] The server reconstructs the replaced text data and converts it back to the original document format. The input is the replaced text data, and the data processing involves converting it back to the original document format (PDF, Word, etc.). The output is the reconstructed document file.

[0737] Step 8:

[0738] The server saves the reconstructed document file and prepares it for the user to download. The input is the reconstructed document file, and a download link to the user's terminal is generated. The output is the download link provided to the user.

[0739] Specific operations include: users dragging and dropping files to upload them; extracting and analyzing text data on the server; analyzing responses to prompts using AI models generated by the sentiment engine; dynamic text replacement processing according to transformation rules; and providing reconstructed documents via generated download links.

[0740] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0741] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0742] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0743] [Third Embodiment]

[0744] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0745] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0746] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0747] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0748] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0749] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0750] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0751] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0752] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0753] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0754] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0755] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0756] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions for use by distributors. The specific procedures and processes for implementing this system are detailed below.

[0757] This system begins with the user uploading document files provided by the manufacturer. The user selects the document files using a terminal and uploads them to the system interface. The terminal then sends the uploaded document files to the server.

[0758] The server receives document files and extracts text data using the appropriate method depending on the file format. If the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0759] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. These keywords and phrases are generally expressions such as "our company" and "standard warranty period."

[0760] The server applies predefined translation rules to detected keywords and phrases. Specifically, it performs processes such as replacing "our company" with "manufacturer." Furthermore, the server uses a custom translation function to make other necessary changes to the wording. This custom translation function utilizes natural language processing technology to perform appropriate translations according to the context.

[0761] The converted text data is reconstructed and returned to its original document format. For example, if the original format was PDF, the converted text data is converted back into a PDF file. The server saves the reconstructed document file and prepares it for later download by the user.

[0762] Finally, a download link appears on the device, and the user downloads the converted document file. In this way, the user can avoid manual errors and perform document conversion efficiently and accurately.

[0763] Specific example

[0764] For example, suppose the manufacturer provides the following document:

[0765] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0766] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." It also changes the phrase "inquiry" to "contact," resulting in the following document:

[0767] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0768] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a distributor.

[0769] The following describes the processing flow.

[0770] Step 1:

[0771] The user selects a document file provided by the manufacturer using a terminal and uploads it to the system interface.

[0772] Step 2:

[0773] The device sends the uploaded document file to the server.

[0774] Step 3:

[0775] The server reads the received document file as text data. If the document is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0776] Step 4:

[0777] The server tokenizes the extracted text data. Tokenization is the process of breaking down text data into words and phrases and identifying each one.

[0778] Step 5:

[0779] The server detects specific keywords or phrases from tokenized text data. This may involve using natural language processing techniques.

[0780] Step 6:

[0781] The server applies predefined conversion rules to the keywords and phrases it detects. Specifically, it performs processes such as replacing "our company" with "manufacturer."

[0782] Step 7:

[0783] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact".

[0784] Step 8:

[0785] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated as a PDF file again.

[0786] Step 9:

[0787] The server saves the reconstructed document files and prepares them for users to download.

[0788] Step 10:

[0789] A download link will appear on the device, and the user will download the converted document file. This file can then be used by the distributor.

[0790] (Example 1)

[0791] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0792] Traditional manual document conversion processes are time-consuming, labor-intensive, and prone to errors. Furthermore, consistent and improper replacement of specific keywords and phrases is often lacking, making them inefficient, especially when processing large volumes of documents. To address these challenges, a system is needed that automatically and accurately converts specific keywords and phrases within document files.

[0793] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0794] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data into word or phrase units; means for detecting specific keywords or phrases from the tokenized text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to automatically detect specific keywords or phrases and replace them accurately and efficiently.

[0795] A "document file" is an electronic file that contains text data, and includes formats such as PDF and Word files.

[0796] "Text data" refers to string data extracted or parsed from document files, and includes words and phrases.

[0797] "Tokenization" is the process of dividing text data into words or phrases and identifying each one individually.

[0798] A "keyword or phrase" is a word or phrase that has a specific meaning and is an important element within a document.

[0799] "Conversion rules" refer to predefined rules for replacing specific keywords or phrases with new expressions.

[0800] "Reconstruction" is the process of restoring the replaced text data to its original document format.

[0801] "Output" refers to providing the user with the completed document file.

[0802] This invention relates to a system for automatically converting specific keywords or phrases within document files. This significantly reduces manual data entry and allows users to efficiently and accurately convert documents. The specific procedures and processes for implementing this system are detailed below.

[0803] This system includes the user's terminal, a server, appropriate analysis modules, and natural language processing technology. The user selects a document file provided by the manufacturer from their terminal and uploads it to the system. The terminal then sends the uploaded document file to the server.

[0804] The server receives document files in formats such as PDF and Word, selects the appropriate analysis module, and extracts the text data. For example, it uses the PDFLib or pdfminer.six library for PDF files and the python-docx library for Word files.

[0805] The extracted text data is tokenized by the server. Natural language processing libraries (such as NLTK or spaCy) are used for tokenization, dividing the text data into words and phrases. The server then detects specific keywords and phrases from the tokenized text. A predefined keyword list is used for this detection.

[0806] The server applies pre-configured translation rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." Furthermore, it uses custom translation functions as needed to perform contextually appropriate translations using natural language processing techniques (such as BERT or GPT).

[0807] The converted text data is then reconstructed by the server into its original document format. For PDFs, the PDFLib library is used to generate a new PDF file, and for Word files, the python-docx library is used to generate a new Word file.

[0808] Finally, the server saves the reconstructed document file and provides a download link to the user's terminal. The user can then use their terminal to download the converted document file via this link.

[0809] As a concrete example, consider a case where the manufacturer provides the following document:

[0810] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0811] When a user uploads this document to the system, the server extracts the text data and converts the keyword "our company" to "manufacturer" and the phrase "inquiry" to "contact". As a result, a document like the following is generated:

[0812] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0813] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can download the converted document using this link.

[0814] Examples of prompts to input into a generative AI model are as follows:

[0815] Please translate the phrases "our company," "standard warranty period," and "inquiries" into language suitable for sales companies.

[0816] In this way, the system enables efficient and accurate conversion of specific keywords and phrases within a document.

[0817] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0818] Step 1:

[0819] The user selects a document file provided by the manufacturer and uploads it from their device using the system interface. The device processes this upload request and sends the document file to the server.

[0820] Input: Document file selected from the user's terminal (PDF or Word format)

[0821] Output: Document file sent to the server

[0822] Step 2:

[0823] The server receives document files sent from the terminal. It recognizes the format of the document file and selects the appropriate parsing module (PDF parsing module or Word parsing module). For PDF files, it uses the PDFLib or pdfminer.six library to extract text data; for Word files, it uses the python-docx library.

[0824] Input: Document file sent to the server

[0825] Output: Extracted text data

[0826] Specific operation: The PDF parsing module uses PDFLib, and the Word parsing module uses python-docx to extract text data from documents.

[0827] Step 3:

[0828] The server tokenizes the extracted text data. This tokenization uses natural language processing libraries (such as NLTK or spaCy) to divide the text into words and phrases.

[0829] Input: Extracted text data

[0830] Output: Tokenized text data

[0831] Specific operation: The text is split into words and phrases using NLTK's word_tokenize function or spaCy's nlp function.

[0832] Step 4:

[0833] The server detects specific keywords or phrases from tokenized text data. It uses a predefined keyword list to check whether each token is included in the list.

[0834] Input: Tokenized text data

[0835] Output: Detected keywords and phrases

[0836] Specific operation: Loop through the tokenized text data and verify whether each token is present in the keyword list.

[0837] Step 5:

[0838] The server applies predefined conversion rules to detected keywords and phrases. For example, it converts "our company" to "manufacturer." For more complex conversions, it utilizes natural language processing techniques (such as BERT and GPT).

[0839] Input: Detected keywords or phrases

[0840] Output: Converted text data

[0841] Specific operation: Basic substitutions are performed using Python's str.replace function, and context-aware transformations are performed using a transformer model.

[0842] Step 6:

[0843] The server reconstructs the converted text data into the original document format. For PDFs, it uses the PDFLib library to generate a new PDF file, and for Word files, it uses the python-docx library to generate a new Word file.

[0844] Input: Converted text data

[0845] Output: Reconstructed document file (PDF or Word format)

[0846] Specific operation: Insert text data into a new document object and format the entire document.

[0847] Step 7:

[0848] The server saves the reconstructed document file and displays a download link on the user's terminal. The user can then download the converted document file from their terminal via this link.

[0849] Input: Reconstructed document file

[0850] Output: Download link displayed on the user's terminal

[0851] Specific operation: The server saves the document file to the storage directory, generates a download link containing the path, and notifies the user.

[0852] (Application Example 1)

[0853] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0854] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions suitable for sales companies, and further outputs the converted content as audio. Conventional systems require manual modification of the wording in document files, which is inefficient and prone to errors. Furthermore, because they are limited to visual information, it is difficult for visually impaired individuals to obtain the information.

[0855] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0856] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined conversion rules; means for reconstructing the replaced text data as a document file; means for outputting the reconstructed document file; means for converting the converted text data into audio data; and means for outputting the converted audio data. This makes it possible to automatically and efficiently convert the expression of a document file for sales companies and further output its contents as audio.

[0857] A "document file" is a file containing text or image data stored in an electronic format.

[0858] "Text data" refers to information expressed as a string of characters, specifically the content extracted from a document file.

[0859] "Analyzing" means breaking down given data and processing it in order to understand its meaning and structure.

[0860] A "keyword or phrase" refers to a word or short sentence with a specific meaning, representing an important part of a document.

[0861] "Detecting" means finding data that has specific conditions or characteristics.

[0862] "Conversion rules" are pre-defined guidelines or rules for replacing specific keywords or phrases with other expressions.

[0863] "Substitution" means replacing one element with another.

[0864] "Reconstructing" means putting data back together to return it to its original state or format.

[0865] "Outputting" means providing processed data in a format that can be used by the user.

[0866] "Audio data" refers to digital data containing audio information, in a format that can be played back by an audio playback device.

[0867] This invention is a system for efficiently converting the expression of business documents for specific purposes, particularly for generating appropriate expressions for customers and outputting them as audio. The following describes specific embodiments for carrying out the invention.

[0868] Program Overview

[0869] The server system and terminals operate in coordination. The system according to the present invention includes the following main means:

[0870] A method for receiving document files and analyzing them as text data.

[0871] Means for detecting specific keywords or phrases from analyzed text data

[0872] A means of replacing detected keywords or phrases with new expressions according to predefined transformation rules.

[0873] A method for reconstructing the replaced text data as a document file.

[0874] Means for outputting a reconstructed document file

[0875] A means of converting converted text data into audio data.

[0876] Means for outputting converted audio data

[0877] Hardware and software to be used

[0878] The server is hosted using Google Cloud Platform, and utilizes SpaCy for natural language processing and the Google Text-to-Speech API for speech synthesis. The device is a smartphone or tablet with an application for uploading document files installed.

[0879] Data processing and data calculation

[0880] 1. Upload document files:

[0881] The user uploads document files (such as PDFs or Word documents) provided by the manufacturer to the system interface using their device. The device then sends the uploaded document files to the server.

[0882] 2. Extraction and analysis of text data:

[0883] The server receives document files and extracts text data using pdfminer (for PDF files) or python-docx (for Word files). The extracted text data is tokenized using the SpaCy library, and specific keywords or phrases are detected and analyzed.

[0884] 3. Text data conversion:

[0885] The server uses natural language processing technology to replace detected keywords and phrases with new expressions according to predefined conversion rules. For example, it might replace "our company" with "manufacturer" and "inquiry" with "contact".

[0886] 4. Reconstruction and output of document files:

[0887] The replaced text data is then reconstructed back into its original format and saved as a PDF or Word file. The user downloads the reconstructed document file from the server via their device.

[0888] 5. Generation and output of audio data:

[0889] The converted text data is then converted into audio data using the Google Text-to-Speech API. This audio data can then be played back by the user on their device.

[0890] Specific example

[0891] For example, if a staff member uploads a document like this:

[0892] We offer a standard warranty period. If you encounter any problems during this warranty period, please contact us.

[0893] This document will be transformed as follows:

[0894] The manufacturer provides a standard warranty period. If any problems occur during this warranty period, please contact us.

[0895] Furthermore, this converted text is also output as audio data.

[0896] Example of a prompt

[0897] The user is a store employee. They uploaded a document file received from the manufacturer. Please translate the content of this document into language that is easy for customers to understand, and also output the content as audio.

[0898] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0899] Step 1:

[0900] The user uploads a document file provided by the manufacturer using their device. The device sends the file to the server regardless of its format (PDF, Word, etc.). The input is the document file, and the output is the file uploaded to the server.

[0901] Step 2:

[0902] The server parses the received document files. Specifically, it uses pdfminer to extract text data from PDF files and python-docx to extract text data from Word files. The input is document files, and the output is text data.

[0903] Step 3:

[0904] The server analyzes the extracted text data and performs tokenization. The SpaCy library is used for this process. Tokenization divides the text data into words and phrases. The input is the text data, and the output is the tokenized data.

[0905] Step 4:

[0906] The server detects specific keywords and phrases from tokenized data. This is done using predefined lists and natural language processing techniques. The input is tokenized data, and the output is the detected keywords and phrases.

[0907] Step 5:

[0908] The server replaces detected keywords and phrases with new expressions according to predefined conversion rules. For example, it replaces "our company" with "manufacturer." The input is the detected keywords and phrases, and the output is the text data after the replacement.

[0909] Step 6:

[0910] The server reconstructs the replaced text data and returns it to the original document format (PDF or Word file). This reconstruction uses the corresponding formatting library. The input is the replaced text data, and the output is the reconstructed document file.

[0911] Step 7:

[0912] The server provides a download link to the terminal for the reconstructed document file, allowing the user to retrieve the file. The input is the reconstructed document file, and the output is the download link.

[0913] Step 8:

[0914] The server converts the converted text data into audio data. It uses the Google Text-to-Speech API to convert the text data into audio data. The input is the replaced text data, and the output is the audio data.

[0915] Step 9:

[0916] The device provides the user with the ability to play audio data. The audio data is streamed and can be listened to by the user. The input is audio data, and the output is audio playback.

[0917] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0918] This invention relates to a system for automatically replacing specific keywords or phrases within a document file, and further incorporates an emotion engine to recognize user emotions and optimize the conversion process. The following describes specific embodiments of the system of this invention.

[0919] The process begins with the user using a terminal to select a document file provided by the manufacturer and uploading it to the system interface. The terminal then sends the uploaded document file to the server.

[0920] The server receives document files and extracts text data using the appropriate method according to the file format. For example, if the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[0921] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. Natural language processing techniques may be used for this detection.

[0922] Next, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's emotional state from documents uploaded by the user and their activity history. For example, if the user is in a hurry, the emotion engine will detect this and adjust the settings to expedite the conversion process.

[0923] The server applies predefined conversion rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." It can also dynamically adjust conversion rules based on the user's emotions, as analyzed by the emotion engine. For instance, if the user is calm, it will select more polite language; if they are in a hurry, it will convert to more concise language.

[0924] Next, the server uses a custom translation function to make any other necessary changes to the wording. Here too, the results of the sentiment engine's analysis are reflected. The converted text data is then reconstructed and returned to its original document format. If it was in PDF format, it will be generated again as a PDF file.

[0925] The server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, and the user downloads the converted document file. This file can be used by the reseller.

[0926] As a concrete example, consider a case where the manufacturer provides the following document.

[0927] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0928] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." Additionally, by changing "inquiry" to "contact," the following document is generated:

[0929] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0930] Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is streamlined and supported to ensure a smooth user experience. The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a reseller.

[0931] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can also respond flexibly to the user's emotional state.

[0932] The following describes the processing flow.

[0933] Step 1:

[0934] The user selects a document file provided by the manufacturer using their device and uploads it to the system interface. The uploaded document files can be in formats such as PDF or Word.

[0935] Step 2:

[0936] The terminal sends the uploaded document file to the server. The server receives this file and begins processing it.

[0937] Step 3:

[0938] The server determines the format of the received document file and performs text extraction processing according to that format. For example, if the file is in PDF format, the PDF parsing module is used to extract the text data; if it is in Word format, the Word parsing module is used.

[0939] Step 4:

[0940] The server analyzes the extracted text data and performs tokenization. Tokenization is the process of breaking down text data into words and phrases and identifying each token.

[0941] Step 5:

[0942] The server detects specific keywords and phrases from the tokenized text data. Examples of keywords and phrases detected include "our company" and "contact us."

[0943] Step 6:

[0944] The server uses an emotion engine to recognize the user's emotional state. The emotion engine analyzes the user's facial expressions, voice, and operation history as they interact with the system to estimate their emotions.

[0945] Step 7:

[0946] The server applies predefined conversion rules to the keywords and phrases it detects. For example, it might replace "our company" with "manufacturer." In this process, the sentiment engine may dynamically adjust the conversion rules based on the user's emotions as analyzed. For instance, if the user is in a hurry, the conversion will prioritize speed.

[0947] Step 8:

[0948] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact." The sentiment engine's analysis results are reflected here as well.

[0949] Step 9:

[0950] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated again as a PDF file.

[0951] Step 10:

[0952] The server saves the reconstructed document files and prepares them for users to download. The generated files are provided as download links for easy access by users.

[0953] Step 11:

[0954] A download link appears on the terminal, and the user downloads the converted document file. This file can then be used by the reseller. Providing documents quickly converted to the format the user needs significantly improves operational efficiency.

[0955] As a concrete example, consider the following document provided by the manufacturer:

[0956] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[0957] When a user uploads this document to the system, the server extracts the text data and converts "our company" to "manufacturer" and "inquiry" to "contact". Furthermore, if the emotion engine detects that the user is in a hurry, it speeds up the conversion process and generates a document like this:

[0958] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[0959] This document file will be saved again in PDF format, and a download link will be provided to the user's device. The user can click the link to download the converted document and use it as a distributor.

[0960] (Example 2)

[0961] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0962] Conventional document conversion systems can replace specific keywords and phrases, but they have a weakness in that they cannot perform dynamic conversions that take into account the user's emotional state. In particular, appropriate text conversion is needed depending on whether the user is in a hurry or calm, and there is a need for a way to do this automatically.

[0963] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0964] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data; means for detecting specific keywords or phrases from the tokenized text data; means for analyzing the user's emotional state; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules, and further means for dynamically adjusting the transformation rules based on the user's emotional state; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This enables flexible and efficient document transformation in response to the user's emotional state.

[0965] A "document file" is a general term for electronic file formats that contain text data. Examples include PDF and Word files.

[0966] "Text data" refers to data that represents textual information as a string of characters.

[0967] "Analysis" is the process of extracting information from document files and identifying and understanding their contents.

[0968] "Tokenization" is the process of breaking down text data into units of words or phrases.

[0969] "Keywords" are words or phrases that are considered particularly important within a document.

[0970] A "phrase" is a unit of language in which two or more words combine to form a single meaning.

[0971] "Emotional state" refers to the user's current emotional and psychological state.

[0972] A "conversion rule" is a predefined set of rules for substituting specific keywords or phrases with other expressions.

[0973] "Dynamic adjustment" means changing settings and behavior in real time according to the situation and conditions.

[0974] "Reconstruction" is the process of returning converted text data to its original document format.

[0975] "Output" refers to the act of providing the user with the processing results.

[0976] This invention is a system that automatically performs the process from uploading a document file to analyzing, converting, and downloading its contents. The system consists of a terminal and a server; the terminal provides an interface for receiving user input, and the server performs the analysis, conversion, and output of the document file.

[0977] First, the user accesses the system interface using a terminal and uploads a document file. Document file formats include PDF and Word. The uploaded file is sent to the server, which temporarily stores it.

[0978] Next, the server analyzes the received document file. The appropriate analysis module is selected depending on the format of the document file. Specifically, PyPDF2 is used for PDF files, and python-docx is used for Word files to extract text data. The extracted text data is tokenized and broken down into words and phrases. Natural language processing libraries such as NLTK and SpaCy are used for this tokenization process.

[0979] The server detects specific keywords and phrases from tokenized text data. Regular expressions and predefined keyword lists are used for detection. At this stage, an emotion engine is utilized to analyze the user's emotional state. The emotion engine uses Hugging Face's Transformers model to identify the emotional state from the user's interaction history and uploaded documents.

[0980] Detected keywords and phrases are replaced with new expressions according to predefined conversion rules. These rules are dynamically adjusted based on the user's emotional state. For example, if the user is in a hurry, the emotion engine detects this and adjusts the settings to expedite the conversion process.

[0981] The replaced text data is then subjected to a custom translation function to modify any other necessary phrasing. This converted text data is then reconstructed and returned to its original document format (such as PDF or Word). Libraries such as ReportLab are used for this reconstruction.

[0982] The reconstructed document file is saved on the server and prepared for user download. A download link is displayed on the terminal, and the user downloads the converted document file. The downloaded file can then be used by the user as a reseller.

[0983] As a concrete example, consider the case where the following document is uploaded by a user.

[0984] "We offer a standard warranty period. To ensure your peace of mind, please direct any inquiries to our support team."

[0985] The server analyzes this document and replaces "our company" with "manufacturer" and "inquiry" with "contact". Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is simplified, and a document like the following is generated.

[0986] "The manufacturer provides a standard warranty period. To ensure your peace of mind, please contact the manufacturer's support team."

[0987] Here are some examples of input prompts for a generative AI model.

[0988] Prompt example:

[0989] "Read the following document and replace specific keywords and phrases (for example, replace "our company" with "manufacturer"). Also, change the wording to be concise if the user is in a hurry, and more polite if they are calm. Document: We offer a standard warranty period. For your peace of mind, please contact our support team. User's emotional state: Urgent Result Document: The manufacturer offers a standard warranty period. For your peace of mind, please contact the manufacturer's support team."

[0990] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can respond flexibly according to the user's emotional state.

[0991] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0992] Step 1:

[0993] The user uploads a document file.

[0994] Input: A document file (such as a PDF or Word document) selected by the user using the terminal interface.

[0995] Specific operation: The user clicks the upload button and uploads the selected document file to the system interface. This operation sends the file data to the server as an HTTP request.

[0996] Output: The document file received by the server.

[0997] Step 2:

[0998] The server receives the document file.

[0999] Input: Uploaded document file.

[1000] Specific operation: The server temporarily stores the received document file in storage. The system automatically identifies the file format (PDF or Word).

[1001] Output: Document files saved on the server.

[1002] Step 3:

[1003] The server extracts the text data from the document file.

[1004] Input: Saved document file.

[1005] Specific operation: The server extracts text data using an analysis module (such as PyPDF2 or python-docx) appropriate for the document file format. For PDFs, PyPDF2 is used to extract text page by page, and for Word files, python-docx is used to extract text.

[1006] Output: Extracted text data.

[1007] Step 4:

[1008] The server tokenizes the text data.

[1009] Input: Extracted text data.

[1010] Specific operation: The server uses natural language processing libraries such as NLTK and SpaCy to tokenize text data. Tokenization breaks down sentences into units of words and phrases. Specifically, the SpaCy nlp function is used to tokenize text data.

[1011] Output: Tokenized text data.

[1012] Step 5:

[1013] The server detects specific keywords or phrases.

[1014] Input: Tokenized text data.

[1015] Specific operation: The server uses regular expressions or predefined keyword lists to detect specific keywords or phrases from text data. For example, to detect the keyword "our company," it uses the regular expression findall function.

[1016] Output: Detected keywords and phrases.

[1017] Step 6:

[1018] The server analyzes the user's emotional state.

[1019] Input: User activity history and text data of uploaded documents.

[1020] Specific operation: The server uses the Hugging Face Transformers model to analyze emotional states. Specifically, it uses the Transformers pipeline function to perform sentiment analysis.

[1021] Output: Analyzed user emotional state.

[1022] Step 7:

[1023] The server replaces keywords and phrases.

[1024] Input: Detected keywords and phrases, and the analyzed user's emotional state.

[1025] Specific operation: The server replaces keywords and phrases with new expressions according to predefined conversion rules. It also dynamically adjusts the replacement rules based on the user's emotional state. For example, it may replace "our company" with "manufacturer" or "inquiry" with "contact".

[1026] Output: Text data after replacement.

[1027] Step 8:

[1028] The server applies the custom translation.

[1029] Input: The text data to be replaced.

[1030] Specific operation: The server uses a custom translation function to further adjust the text data as needed. For example, if the user is in a hurry, it might change "Contact Us" to "Contact Us" to make the text more concise.

[1031] Output: Text data after custom translation has been applied.

[1032] Step 9:

[1033] The server reconstructs the converted document.

[1034] Input: Text data after custom translation has been applied.

[1035] Specific operation: The server reconstructs the text data into its original document format (PDF or Word). In the case of PDF, it uses the ReportLab library for reconstruction.

[1036] Output: Reconstructed document file.

[1037] Step 10:

[1038] The server generates a download link.

[1039] Input: Reconstructed document file.

[1040] Specific operation: The server saves the reconstructed document file and generates a URL so that the user can download the file. This URL is sent to the user's device.

[1041] Output: The generated download link.

[1042] Step 11:

[1043] The user downloads the converted document.

[1044] Input: Download link provided by the server.

[1045] Specific action: The user clicks the download link displayed on the terminal and downloads the converted document file.

[1046] Output: Document files downloaded by the user.

[1047] (Application Example 2)

[1048] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1049] Conventional document conversion systems lacked flexibility and usability because they replaced keywords and phrases based on static rules without considering user sentiment. Furthermore, especially in virtual stores, the accurate and effective conversion of product descriptions is crucial, but current systems struggled to meet this requirement.

[1050] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1051] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for analyzing the user's emotions and dynamically adjusting the document transformation rules based on those emotions; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to optimize the document transformation process according to the user's emotional state, enabling it to be performed more flexibly and efficiently.

[1052] A "document file" is an electronically stored document containing text data.

[1053] "Text data" refers to data that is stored as character information.

[1054] "Analysis" is the process of breaking down data and analyzing its individual elements.

[1055] "Keywords" are particularly important words or phrases within a document.

[1056] A "phrase" is a combination of multiple words that are grammatically or semantically coherent.

[1057] "Conversion rules" are a set of predefined rules for substituting specific keywords or phrases with other expressions.

[1058] "User emotions" refers to the emotional state a user exhibits while using the system, including states such as being in a hurry or being calm.

[1059] "Emotional analysis" refers to the process of determining a user's emotional state based on their actions and input data.

[1060] "Substitution" is the process of replacing one word or phrase with another.

[1061] "Dynamic adjustment" refers to the process of changing conversion rules in real time according to conditions such as the user's emotional state.

[1062] "Reconstruction" refers to the process of converting replaced text data back into a new document file format.

[1063] A "generative AI model" is an artificial intelligence built using deep learning that performs text generation and analysis according to various tasks.

[1064] A "prompt statement" is a sentence or phrase used as input in natural language processing.

[1065] To realize this invention, the following system configuration and processing are necessary. The system is designed so that users can operate it using a terminal such as a smartphone or PC.

[1066] A user accesses the system interface and uploads a document file. This document file is sent to the server. The server extracts the text data using the appropriate module depending on the format of the uploaded document file. For example, a PDF file uses the PDF parsing module, and a Word file uses the Word parsing module.

[1067] The analyzed text data is tokenized using natural language processing techniques. Specifically, it is broken down into words and phrases, and each is processed to identify it. Next, the server detects specific keywords and phrases from the extracted tokenized data. Transformation rules are predefined, and the server uses these rules to detect keywords and phrases.

[1068] Next, the server uses an emotion engine to analyze the user's emotions. The emotion engine uses a generative AI model to analyze the user's emotional state. Based on this emotion analysis, the conversion rules are dynamically adjusted. For example, if the user is in a hurry, it generates concise sentences; if they are calm, it generates polite sentences.

[1069] The server replaces detected keywords and phrases with new expressions according to pre-defined conversion rules. The sentiment engine's analysis results are also reflected in this process. The converted text data is then reconstructed and returned to its original format. For example, if it was in PDF format, it will be generated again as a PDF file.

[1070] Finally, the server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, allowing the user to download the converted document file.

[1071] The hardware and software used include the following:

[1072] Hardware: Smartphones (iOS, Android), PC

[1073] Software: Python, Hugging Face transformers library, TextBlob library

[1074] As a concrete example, consider the case where a user uploads the following document:

[1075] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[1076] The server analyzes this document and transforms specific keywords and phrases as follows:

[1077] The virtual store offers a standard warranty period. For your peace of mind, please contact the virtual store's support team.

[1078] Furthermore, prompts are used to analyze user sentiment. Examples of specific prompts are shown below:

[1079] Text for when the user is in a hurry

[1080] By inputting this prompt into the AI ​​model, the user's emotional state of being in a hurry is analyzed. Based on this information, the server performs appropriate text conversion.

[1081] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1082] Step 1:

[1083] The user uploads a document file using their terminal. The input is the document file itself, which is sent to the server. The output is the document file stored on the server.

[1084] Step 2:

[1085] The server extracts text data in an appropriate manner depending on the format of the uploaded document file. The input is the document file format (PDF, Word, etc.), and the data processing involves using PDF parsing modules or Word parsing modules to obtain the text data. The output is the extracted text data.

[1086] Step 3:

[1087] The server analyzes and tokenizes the extracted text data. The input is the extracted text data, and the data processing utilizes natural language processing techniques to break down the data into words and phrases. The output is the tokenized text data.

[1088] Step 4:

[1089] The server detects specific keywords and phrases from tokenized text data. The input is tokenized text data, and the data operation involves detecting keywords and phrases based on predefined transformation rules. The output is the detected keywords and phrases.

[1090] Step 5:

[1091] The server uses an emotion engine (generative AI model) to analyze the user's emotions. The input consists of the user's operation history and uploaded documents. As part of the data processing, prompt sentences are generated and input into the generative AI model to analyze the user's emotional state. The output is the result of the user's emotion analysis.

[1092] Step 6:

[1093] The server replaces detected keywords and phrases with new expressions based on predefined transformation rules. The input consists of detected keywords and phrases, as well as the user sentiment analysis results. The server performs the replacement process based on predefined transformation rules as a data calculation. The output is the replaced text data.

[1094] Step 7:

[1095] The server reconstructs the replaced text data and converts it back to the original document format. The input is the replaced text data, and the data processing involves converting it back to the original document format (PDF, Word, etc.). The output is the reconstructed document file.

[1096] Step 8:

[1097] The server saves the reconstructed document file and prepares it for the user to download. The input is the reconstructed document file, and a download link to the user's terminal is generated. The output is the download link provided to the user.

[1098] Specific operations include: users dragging and dropping files to upload them; extracting and analyzing text data on the server; analyzing responses to prompts using AI models generated by the sentiment engine; dynamic text replacement processing according to transformation rules; and providing reconstructed documents via generated download links.

[1099] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1100] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1101] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1102] [Fourth Embodiment]

[1103] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1104] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1105] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1106] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1107] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1108] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1109] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1110] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1111] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1112] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1113] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1114] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1115] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1116] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions for use by distributors. The specific procedures and processes for implementing this system are detailed below.

[1117] This system begins with the user uploading document files provided by the manufacturer. The user selects the document files using a terminal and uploads them to the system interface. The terminal then sends the uploaded document files to the server.

[1118] The server receives document files and extracts text data using the appropriate method depending on the file format. If the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[1119] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. These keywords and phrases are generally expressions such as "our company" and "standard warranty period."

[1120] The server applies predefined translation rules to detected keywords and phrases. Specifically, it performs processes such as replacing "our company" with "manufacturer." Furthermore, the server uses a custom translation function to make other necessary changes to the wording. This custom translation function utilizes natural language processing technology to perform appropriate translations according to the context.

[1121] The converted text data is reconstructed and returned to its original document format. For example, if the original format was PDF, the converted text data is converted back into a PDF file. The server saves the reconstructed document file and prepares it for later download by the user.

[1122] Finally, a download link appears on the device, and the user downloads the converted document file. In this way, the user can avoid manual errors and perform document conversion efficiently and accurately.

[1123] Specific example

[1124] For example, suppose the manufacturer provides the following document:

[1125] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[1126] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." It also changes the phrase "inquiry" to "contact," resulting in the following document:

[1127] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[1128] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a distributor.

[1129] The following describes the processing flow.

[1130] Step 1:

[1131] The user selects a document file provided by the manufacturer using a terminal and uploads it to the system interface.

[1132] Step 2:

[1133] The device sends the uploaded document file to the server.

[1134] Step 3:

[1135] The server reads the received document file as text data. If the document is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[1136] Step 4:

[1137] The server tokenizes the extracted text data. Tokenization is the process of breaking down text data into words and phrases and identifying each one.

[1138] Step 5:

[1139] The server detects specific keywords or phrases from tokenized text data. This may involve using natural language processing techniques.

[1140] Step 6:

[1141] The server applies predefined conversion rules to the keywords and phrases it detects. Specifically, it performs processes such as replacing "our company" with "manufacturer."

[1142] Step 7:

[1143] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact".

[1144] Step 8:

[1145] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated as a PDF file again.

[1146] Step 9:

[1147] The server saves the reconstructed document files and prepares them for users to download.

[1148] Step 10:

[1149] A download link will appear on the device, and the user will download the converted document file. This file can then be used by the distributor.

[1150] (Example 1)

[1151] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1152] Traditional manual document conversion processes are time-consuming, labor-intensive, and prone to errors. Furthermore, consistent and improper replacement of specific keywords and phrases is often lacking, making them inefficient, especially when processing large volumes of documents. To address these challenges, a system is needed that automatically and accurately converts specific keywords and phrases within document files.

[1153] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1154] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data into word or phrase units; means for detecting specific keywords or phrases from the tokenized text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to automatically detect specific keywords or phrases and replace them accurately and efficiently.

[1155] A "document file" is an electronic file that contains text data, and includes formats such as PDF and Word files.

[1156] "Text data" refers to string data extracted or parsed from document files, and includes words and phrases.

[1157] "Tokenization" is the process of dividing text data into words or phrases and identifying each one individually.

[1158] A "keyword or phrase" is a word or phrase that has a specific meaning and is an important element within a document.

[1159] "Conversion rules" refer to predefined rules for replacing specific keywords or phrases with new expressions.

[1160] "Reconstruction" is the process of restoring the replaced text data to its original document format.

[1161] "Output" refers to providing the user with the completed document file.

[1162] This invention relates to a system for automatically converting specific keywords or phrases within document files. This significantly reduces manual data entry and allows users to efficiently and accurately convert documents. The specific procedures and processes for implementing this system are detailed below.

[1163] This system includes the user's terminal, a server, appropriate analysis modules, and natural language processing technology. The user selects a document file provided by the manufacturer from their terminal and uploads it to the system. The terminal then sends the uploaded document file to the server.

[1164] The server receives document files in formats such as PDF and Word, selects the appropriate analysis module, and extracts the text data. For example, it uses the PDFLib or pdfminer.six library for PDF files and the python-docx library for Word files.

[1165] The extracted text data is tokenized by the server. Natural language processing libraries (such as NLTK or spaCy) are used for tokenization, dividing the text data into words and phrases. The server then detects specific keywords and phrases from the tokenized text. A predefined keyword list is used for this detection.

[1166] The server applies pre-configured translation rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." Furthermore, it uses custom translation functions as needed to perform contextually appropriate translations using natural language processing techniques (such as BERT or GPT).

[1167] The converted text data is then reconstructed by the server into its original document format. For PDFs, the PDFLib library is used to generate a new PDF file, and for Word files, the python-docx library is used to generate a new Word file.

[1168] Finally, the server saves the reconstructed document file and provides a download link to the user's terminal. The user can then use their terminal to download the converted document file via this link.

[1169] As a concrete example, consider a case where the manufacturer provides the following document:

[1170] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[1171] When a user uploads this document to the system, the server extracts the text data and converts the keyword "our company" to "manufacturer" and the phrase "inquiry" to "contact". As a result, a document like the following is generated:

[1172] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[1173] The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can download the converted document using this link.

[1174] Examples of prompts to input into a generative AI model are as follows:

[1175] Please translate the phrases "our company," "standard warranty period," and "inquiries" into language suitable for sales companies.

[1176] In this way, the system enables efficient and accurate conversion of specific keywords and phrases within a document.

[1177] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1178] Step 1:

[1179] The user selects a document file provided by the manufacturer and uploads it from their device using the system interface. The device processes this upload request and sends the document file to the server.

[1180] Input: Document file selected from the user's terminal (PDF or Word format)

[1181] Output: Document file sent to the server

[1182] Step 2:

[1183] The server receives document files sent from the terminal. It recognizes the format of the document file and selects the appropriate parsing module (PDF parsing module or Word parsing module). For PDF files, it uses the PDFLib or pdfminer.six library to extract text data; for Word files, it uses the python-docx library.

[1184] Input: Document file sent to the server

[1185] Output: Extracted text data

[1186] Specific operation: The PDF parsing module uses PDFLib, and the Word parsing module uses python-docx to extract text data from documents.

[1187] Step 3:

[1188] The server tokenizes the extracted text data. This tokenization uses natural language processing libraries (such as NLTK or spaCy) to divide the text into words and phrases.

[1189] Input: Extracted text data

[1190] Output: Tokenized text data

[1191] Specific operation: The text is split into words and phrases using NLTK's word_tokenize function or spaCy's nlp function.

[1192] Step 4:

[1193] The server detects specific keywords or phrases from tokenized text data. It uses a predefined keyword list to check whether each token is included in the list.

[1194] Input: Tokenized text data

[1195] Output: Detected keywords and phrases

[1196] Specific operation: Loop through the tokenized text data and verify whether each token is present in the keyword list.

[1197] Step 5:

[1198] The server applies predefined conversion rules to detected keywords and phrases. For example, it converts "our company" to "manufacturer." For more complex conversions, it utilizes natural language processing techniques (such as BERT and GPT).

[1199] Input: Detected keywords or phrases

[1200] Output: Converted text data

[1201] Specific operation: Basic substitutions are performed using Python's str.replace function, and context-aware transformations are performed using a transformer model.

[1202] Step 6:

[1203] The server reconstructs the converted text data into the original document format. For PDFs, it uses the PDFLib library to generate a new PDF file, and for Word files, it uses the python-docx library to generate a new Word file.

[1204] Input: Converted text data

[1205] Output: Reconstructed document file (PDF or Word format)

[1206] Specific operation: Insert text data into a new document object and format the entire document.

[1207] Step 7:

[1208] The server saves the reconstructed document file and displays a download link on the user's terminal. The user can then download the converted document file from their terminal via this link.

[1209] Input: Reconstructed document file

[1210] Output: Download link displayed on the user's terminal

[1211] Specific operation: The server saves the document file to the storage directory, generates a download link containing the path, and notifies the user.

[1212] (Application Example 1)

[1213] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1214] This invention relates to a system that automatically converts specific keywords and phrases within document files provided by manufacturers into expressions suitable for sales companies, and further outputs the converted content as audio. Conventional systems require manual modification of the wording in document files, which is inefficient and prone to errors. Furthermore, because they are limited to visual information, it is difficult for visually impaired individuals to obtain the information.

[1215] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1216] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined conversion rules; means for reconstructing the replaced text data as a document file; means for outputting the reconstructed document file; means for converting the converted text data into audio data; and means for outputting the converted audio data. This makes it possible to automatically and efficiently convert the expression of a document file for sales companies and further output its contents as audio.

[1217] A "document file" is a file containing text or image data stored in an electronic format.

[1218] "Text data" refers to information expressed as a string of characters, specifically the content extracted from a document file.

[1219] "Analyzing" means breaking down given data and processing it in order to understand its meaning and structure.

[1220] A "keyword or phrase" refers to a word or short sentence with a specific meaning, representing an important part of a document.

[1221] "Detecting" means finding data that has specific conditions or characteristics.

[1222] "Conversion rules" are pre-defined guidelines or rules for replacing specific keywords or phrases with other expressions.

[1223] "Substitution" means replacing one element with another.

[1224] "Reconstructing" means putting data back together to return it to its original state or format.

[1225] "Outputting" means providing processed data in a format that can be used by the user.

[1226] "Audio data" refers to digital data containing audio information, in a format that can be played back by an audio playback device.

[1227] This invention is a system for efficiently converting the expression of business documents for specific purposes, particularly for generating appropriate expressions for customers and outputting them as audio. The following describes specific embodiments for carrying out the invention.

[1228] Program Overview

[1229] The server system and terminals operate in coordination. The system according to the present invention includes the following main means:

[1230] A method for receiving document files and analyzing them as text data.

[1231] Means for detecting specific keywords or phrases from analyzed text data

[1232] A means of replacing detected keywords or phrases with new expressions according to predefined transformation rules.

[1233] A method for reconstructing the replaced text data as a document file.

[1234] Means for outputting a reconstructed document file

[1235] A means of converting converted text data into audio data.

[1236] Means for outputting converted audio data

[1237] Hardware and software to be used

[1238] The server is hosted using Google Cloud Platform, and utilizes SpaCy for natural language processing and the Google Text-to-Speech API for speech synthesis. The device is a smartphone or tablet with an application for uploading document files installed.

[1239] Data processing and data calculation

[1240] 1. Upload document files:

[1241] The user uploads document files (such as PDFs or Word documents) provided by the manufacturer to the system interface using their device. The device then sends the uploaded document files to the server.

[1242] 2. Extraction and analysis of text data:

[1243] The server receives document files and extracts text data using pdfminer (for PDF files) or python-docx (for Word files). The extracted text data is tokenized using the SpaCy library, and specific keywords or phrases are detected and analyzed.

[1244] 3. Text data conversion:

[1245] The server uses natural language processing technology to replace detected keywords and phrases with new expressions according to predefined conversion rules. For example, it might replace "our company" with "manufacturer" and "inquiry" with "contact".

[1246] 4. Reconstruction and output of document files:

[1247] The replaced text data is then reconstructed back into its original format and saved as a PDF or Word file. The user downloads the reconstructed document file from the server via their device.

[1248] 5. Generation and output of audio data:

[1249] The converted text data is then converted into audio data using the Google Text-to-Speech API. This audio data can then be played back by the user on their device.

[1250] Specific example

[1251] For example, if a staff member uploads a document like this:

[1252] We offer a standard warranty period. If you encounter any problems during this warranty period, please contact us.

[1253] This document will be transformed as follows:

[1254] The manufacturer provides a standard warranty period. If any problems occur during this warranty period, please contact us.

[1255] Furthermore, this converted text is also output as audio data.

[1256] Example of a prompt

[1257] The user is a store employee. They uploaded a document file received from the manufacturer. Please translate the content of this document into language that is easy for customers to understand, and also output the content as audio.

[1258] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1259] Step 1:

[1260] The user uploads a document file provided by the manufacturer using their device. The device sends the file to the server regardless of its format (PDF, Word, etc.). The input is the document file, and the output is the file uploaded to the server.

[1261] Step 2:

[1262] The server parses the received document files. Specifically, it uses pdfminer to extract text data from PDF files and python-docx to extract text data from Word files. The input is document files, and the output is text data.

[1263] Step 3:

[1264] The server analyzes the extracted text data and performs tokenization. The SpaCy library is used for this process. Tokenization divides the text data into words and phrases. The input is the text data, and the output is the tokenized data.

[1265] Step 4:

[1266] The server detects specific keywords and phrases from tokenized data. This is done using predefined lists and natural language processing techniques. The input is tokenized data, and the output is the detected keywords and phrases.

[1267] Step 5:

[1268] The server replaces detected keywords and phrases with new expressions according to predefined conversion rules. For example, it replaces "our company" with "manufacturer." The input is the detected keywords and phrases, and the output is the text data after the replacement.

[1269] Step 6:

[1270] The server reconstructs the replaced text data and returns it to the original document format (PDF or Word file). This reconstruction uses the corresponding formatting library. The input is the replaced text data, and the output is the reconstructed document file.

[1271] Step 7:

[1272] The server provides a download link to the terminal for the reconstructed document file, allowing the user to retrieve the file. The input is the reconstructed document file, and the output is the download link.

[1273] Step 8:

[1274] The server converts the converted text data into audio data. It uses the Google Text-to-Speech API to convert the text data into audio data. The input is the replaced text data, and the output is the audio data.

[1275] Step 9:

[1276] The device provides the user with the ability to play audio data. The audio data is streamed and can be listened to by the user. The input is audio data, and the output is audio playback.

[1277] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1278] This invention relates to a system for automatically replacing specific keywords or phrases within a document file, and further incorporates an emotion engine to recognize user emotions and optimize the conversion process. The following describes specific embodiments of the system of this invention.

[1279] The process begins with the user using a terminal to select a document file provided by the manufacturer and uploading it to the system interface. The terminal then sends the uploaded document file to the server.

[1280] The server receives document files and extracts text data using the appropriate method according to the file format. For example, if the document file is in PDF format, the PDF parsing module is used to extract the text data; if it is a Word file, the Word parsing module is used.

[1281] The extracted text data is parsed and tokenized on the server. Tokenization is the process of breaking down text data into words and phrases and identifying each one. The server then detects specific keywords and phrases from the tokenized text data. Natural language processing techniques may be used for this detection.

[1282] Next, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's emotional state from documents uploaded by the user and their activity history. For example, if the user is in a hurry, the emotion engine will detect this and adjust the settings to expedite the conversion process.

[1283] The server applies predefined conversion rules to detected keywords and phrases. For example, it might replace "our company" with "manufacturer." It can also dynamically adjust conversion rules based on the user's emotions, as analyzed by the emotion engine. For instance, if the user is calm, it will select more polite language; if they are in a hurry, it will convert to more concise language.

[1284] Next, the server uses a custom translation function to make any other necessary changes to the wording. Here too, the results of the sentiment engine's analysis are reflected. The converted text data is then reconstructed and returned to its original document format. If it was in PDF format, it will be generated again as a PDF file.

[1285] The server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, and the user downloads the converted document file. This file can be used by the reseller.

[1286] As a concrete example, consider a case where the manufacturer provides the following document.

[1287] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[1288] When a user uploads this document to the system, the server extracts the text data and changes the keyword "our company" to "manufacturer." Additionally, by changing "inquiry" to "contact," the following document is generated:

[1289] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[1290] Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is streamlined and supported to ensure a smooth user experience. The generated document file is saved again as a PDF or Word document, and a download link is provided to the user's device. The user can use this link to download the converted document and use it as a reseller.

[1291] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can also respond flexibly to the user's emotional state.

[1292] The following describes the processing flow.

[1293] Step 1:

[1294] The user selects a document file provided by the manufacturer using their device and uploads it to the system interface. The uploaded document files can be in formats such as PDF or Word.

[1295] Step 2:

[1296] The terminal sends the uploaded document file to the server. The server receives this file and begins processing it.

[1297] Step 3:

[1298] The server determines the format of the received document file and performs text extraction processing according to that format. For example, if the file is in PDF format, the PDF parsing module is used to extract the text data; if it is in Word format, the Word parsing module is used.

[1299] Step 4:

[1300] The server analyzes the extracted text data and performs tokenization. Tokenization is the process of breaking down text data into words and phrases and identifying each token.

[1301] Step 5:

[1302] The server detects specific keywords and phrases from the tokenized text data. Examples of keywords and phrases detected include "our company" and "contact us."

[1303] Step 6:

[1304] The server uses an emotion engine to recognize the user's emotional state. The emotion engine analyzes the user's facial expressions, voice, and operation history as they interact with the system to estimate their emotions.

[1305] Step 7:

[1306] The server applies predefined conversion rules to the keywords and phrases it detects. For example, it might replace "our company" with "manufacturer." In this process, the sentiment engine may dynamically adjust the conversion rules based on the user's emotions as analyzed. For instance, if the user is in a hurry, the conversion will prioritize speed.

[1307] Step 8:

[1308] The server uses a custom translation function to make other necessary changes to the wording. For example, it might change "inquiry" to "contact." The sentiment engine's analysis results are reflected here as well.

[1309] Step 9:

[1310] The server reconstructs the text data after the conversion process is complete. This process restores it to its original document format; if it was in PDF format, it will be generated again as a PDF file.

[1311] Step 10:

[1312] The server saves the reconstructed document files and prepares them for users to download. The generated files are provided as download links for easy access by users.

[1313] Step 11:

[1314] A download link appears on the terminal, and the user downloads the converted document file. This file can then be used by the reseller. Providing documents quickly converted to the format the user needs significantly improves operational efficiency.

[1315] As a concrete example, consider the following document provided by the manufacturer:

[1316] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[1317] When a user uploads this document to the system, the server extracts the text data and converts "our company" to "manufacturer" and "inquiry" to "contact". Furthermore, if the emotion engine detects that the user is in a hurry, it speeds up the conversion process and generates a document like this:

[1318] The manufacturer provides a standard warranty period. For your peace of mind, please contact the manufacturer's support team.

[1319] This document file will be saved again in PDF format, and a download link will be provided to the user's device. The user can click the link to download the converted document and use it as a distributor.

[1320] (Example 2)

[1321] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1322] Conventional document conversion systems can replace specific keywords and phrases, but they have a weakness in that they cannot perform dynamic conversions that take into account the user's emotional state. In particular, appropriate text conversion is needed depending on whether the user is in a hurry or calm, and there is a need for a way to do this automatically.

[1323] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1324] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for tokenizing the analyzed text data; means for detecting specific keywords or phrases from the tokenized text data; means for analyzing the user's emotional state; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules, and further means for dynamically adjusting the transformation rules based on the user's emotional state; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This enables flexible and efficient document transformation in response to the user's emotional state.

[1325] A "document file" is a general term for electronic file formats that contain text data. Examples include PDF and Word files.

[1326] "Text data" refers to data that represents textual information as a string of characters.

[1327] "Analysis" is the process of extracting information from document files and identifying and understanding their contents.

[1328] "Tokenization" is the process of breaking down text data into units of words or phrases.

[1329] "Keywords" are words or phrases that are considered particularly important within a document.

[1330] A "phrase" is a unit of language in which two or more words combine to form a single meaning.

[1331] "Emotional state" refers to the user's current emotional and psychological state.

[1332] A "conversion rule" is a predefined set of rules for substituting specific keywords or phrases with other expressions.

[1333] "Dynamic adjustment" means changing settings and behavior in real time according to the situation and conditions.

[1334] "Reconstruction" is the process of returning converted text data to its original document format.

[1335] "Output" refers to the act of providing the user with the processing results.

[1336] This invention is a system that automatically performs the process from uploading a document file to analyzing, converting, and downloading its contents. The system consists of a terminal and a server; the terminal provides an interface for receiving user input, and the server performs the analysis, conversion, and output of the document file.

[1337] First, the user accesses the system interface using a terminal and uploads a document file. Document file formats include PDF and Word. The uploaded file is sent to the server, which temporarily stores it.

[1338] Next, the server analyzes the received document file. The appropriate analysis module is selected depending on the format of the document file. Specifically, PyPDF2 is used for PDF files, and python-docx is used for Word files to extract text data. The extracted text data is tokenized and broken down into words and phrases. Natural language processing libraries such as NLTK and SpaCy are used for this tokenization process.

[1339] The server detects specific keywords and phrases from tokenized text data. Regular expressions and predefined keyword lists are used for detection. At this stage, an emotion engine is utilized to analyze the user's emotional state. The emotion engine uses Hugging Face's Transformers model to identify the emotional state from the user's interaction history and uploaded documents.

[1340] Detected keywords and phrases are replaced with new expressions according to predefined conversion rules. These rules are dynamically adjusted based on the user's emotional state. For example, if the user is in a hurry, the emotion engine detects this and adjusts the settings to expedite the conversion process.

[1341] The replaced text data is then subjected to a custom translation function to modify any other necessary phrasing. This converted text data is then reconstructed and returned to its original document format (such as PDF or Word). Libraries such as ReportLab are used for this reconstruction.

[1342] The reconstructed document file is saved on the server and prepared for user download. A download link is displayed on the terminal, and the user downloads the converted document file. The downloaded file can then be used by the user as a reseller.

[1343] As a concrete example, consider the case where the following document is uploaded by a user.

[1344] "We offer a standard warranty period. To ensure your peace of mind, please direct any inquiries to our support team."

[1345] The server analyzes this document and replaces "our company" with "manufacturer" and "inquiry" with "contact". Furthermore, if the emotion engine detects that the user is in a hurry, the conversion process is simplified, and a document like the following is generated.

[1346] "The manufacturer provides a standard warranty period. To ensure your peace of mind, please contact the manufacturer's support team."

[1347] Here are some examples of input prompts for a generative AI model.

[1348] Prompt example:

[1349] "Read the following document and replace specific keywords and phrases (for example, replace "our company" with "manufacturer"). Also, change the wording to be concise if the user is in a hurry, and more polite if they are calm. Document: We offer a standard warranty period. For your peace of mind, please contact our support team. User's emotional state: Urgent Result Document: The manufacturer offers a standard warranty period. For your peace of mind, please contact the manufacturer's support team."

[1350] In this way, the system of the present invention can perform document conversion work efficiently and accurately, and can respond flexibly according to the user's emotional state.

[1351] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1352] Step 1:

[1353] The user uploads a document file.

[1354] Input: A document file (such as a PDF or Word document) selected by the user using the terminal interface.

[1355] Specific operation: The user clicks the upload button and uploads the selected document file to the system interface. This operation sends the file data to the server as an HTTP request.

[1356] Output: The document file received by the server.

[1357] Step 2:

[1358] The server receives the document file.

[1359] Input: Uploaded document file.

[1360] Specific operation: The server temporarily stores the received document file in storage. The system automatically identifies the file format (PDF or Word).

[1361] Output: Document files saved on the server.

[1362] Step 3:

[1363] The server extracts the text data from the document file.

[1364] Input: Saved document file.

[1365] Specific operation: The server extracts text data using an analysis module (such as PyPDF2 or python-docx) appropriate for the document file format. For PDFs, PyPDF2 is used to extract text page by page, and for Word files, python-docx is used to extract text.

[1366] Output: Extracted text data.

[1367] Step 4:

[1368] The server tokenizes the text data.

[1369] Input: Extracted text data.

[1370] Specific operation: The server uses natural language processing libraries such as NLTK and SpaCy to tokenize text data. Tokenization breaks down sentences into units of words and phrases. Specifically, the SpaCy nlp function is used to tokenize text data.

[1371] Output: Tokenized text data.

[1372] Step 5:

[1373] The server detects specific keywords or phrases.

[1374] Input: Tokenized text data.

[1375] Specific operation: The server uses regular expressions or predefined keyword lists to detect specific keywords or phrases from text data. For example, to detect the keyword "our company," it uses the regular expression findall function.

[1376] Output: Detected keywords and phrases.

[1377] Step 6:

[1378] The server analyzes the user's emotional state.

[1379] Input: User activity history and text data of uploaded documents.

[1380] Specific operation: The server uses the Hugging Face Transformers model to analyze emotional states. Specifically, it uses the Transformers pipeline function to perform sentiment analysis.

[1381] Output: Analyzed user emotional state.

[1382] Step 7:

[1383] The server replaces keywords and phrases.

[1384] Input: Detected keywords and phrases, and the analyzed user's emotional state.

[1385] Specific operation: The server replaces keywords and phrases with new expressions according to predefined conversion rules. It also dynamically adjusts the replacement rules based on the user's emotional state. For example, it may replace "our company" with "manufacturer" or "inquiry" with "contact".

[1386] Output: Text data after replacement.

[1387] Step 8:

[1388] The server applies the custom translation.

[1389] Input: The text data to be replaced.

[1390] Specific operation: The server uses a custom translation function to further adjust the text data as needed. For example, if the user is in a hurry, it might change "Contact Us" to "Contact Us" to make the text more concise.

[1391] Output: Text data after custom translation has been applied.

[1392] Step 9:

[1393] The server reconstructs the converted document.

[1394] Input: Text data after custom translation has been applied.

[1395] Specific operation: The server reconstructs the text data into its original document format (PDF or Word). In the case of PDF, it uses the ReportLab library for reconstruction.

[1396] Output: Reconstructed document file.

[1397] Step 10:

[1398] The server generates a download link.

[1399] Input: Reconstructed document file.

[1400] Specific operation: The server saves the reconstructed document file and generates a URL so that the user can download the file. This URL is sent to the user's device.

[1401] Output: The generated download link.

[1402] Step 11:

[1403] The user downloads the converted document.

[1404] Input: Download link provided by the server.

[1405] Specific action: The user clicks the download link displayed on the terminal and downloads the converted document file.

[1406] Output: Document files downloaded by the user.

[1407] (Application Example 2)

[1408] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1409] Conventional document conversion systems lacked flexibility and usability because they replaced keywords and phrases based on static rules without considering user sentiment. Furthermore, especially in virtual stores, the accurate and effective conversion of product descriptions is crucial, but current systems struggled to meet this requirement.

[1410] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1411] In this invention, the server includes means for receiving a document file and analyzing it as text data; means for detecting specific keywords or phrases from the analyzed text data; means for replacing the detected keywords or phrases with new expressions according to predefined transformation rules; means for analyzing the user's emotions and dynamically adjusting the document transformation rules based on those emotions; means for reconstructing the replaced text data as a document file; and means for outputting the reconstructed document file. This makes it possible to optimize the document transformation process according to the user's emotional state, enabling it to be performed more flexibly and efficiently.

[1412] A "document file" is an electronically stored document containing text data.

[1413] "Text data" refers to data that is stored as character information.

[1414] "Analysis" is the process of breaking down data and analyzing its individual elements.

[1415] "Keywords" are particularly important words or phrases within a document.

[1416] A "phrase" is a combination of multiple words that are grammatically or semantically coherent.

[1417] "Conversion rules" are a set of predefined rules for substituting specific keywords or phrases with other expressions.

[1418] "User emotions" refers to the emotional state a user exhibits while using the system, including states such as being in a hurry or being calm.

[1419] "Emotional analysis" refers to the process of determining a user's emotional state based on their actions and input data.

[1420] "Substitution" is the process of replacing one word or phrase with another.

[1421] "Dynamic adjustment" refers to the process of changing conversion rules in real time according to conditions such as the user's emotional state.

[1422] "Reconstruction" refers to the process of converting replaced text data back into a new document file format.

[1423] A "generative AI model" is an artificial intelligence built using deep learning that performs text generation and analysis according to various tasks.

[1424] A "prompt statement" is a sentence or phrase used as input in natural language processing.

[1425] To realize this invention, the following system configuration and processing are necessary. The system is designed so that users can operate it using a terminal such as a smartphone or PC.

[1426] A user accesses the system interface and uploads a document file. This document file is sent to the server. The server extracts the text data using the appropriate module depending on the format of the uploaded document file. For example, a PDF file uses the PDF parsing module, and a Word file uses the Word parsing module.

[1427] The analyzed text data is tokenized using natural language processing techniques. Specifically, it is broken down into words and phrases, and each is processed to identify it. Next, the server detects specific keywords and phrases from the extracted tokenized data. Transformation rules are predefined, and the server uses these rules to detect keywords and phrases.

[1428] Next, the server uses an emotion engine to analyze the user's emotions. The emotion engine uses a generative AI model to analyze the user's emotional state. Based on this emotion analysis, the conversion rules are dynamically adjusted. For example, if the user is in a hurry, it generates concise sentences; if they are calm, it generates polite sentences.

[1429] The server replaces detected keywords and phrases with new expressions according to pre-defined conversion rules. The sentiment engine's analysis results are also reflected in this process. The converted text data is then reconstructed and returned to its original format. For example, if it was in PDF format, it will be generated again as a PDF file.

[1430] Finally, the server saves the reconstructed document file and prepares it for the user to download. A download link appears on the terminal, allowing the user to download the converted document file.

[1431] The hardware and software used include the following:

[1432] Hardware: Smartphones (iOS, Android), PC

[1433] Software: Python, Hugging Face transformers library, TextBlob library

[1434] As a concrete example, consider the case where a user uploads the following document:

[1435] We offer a standard warranty period. For your peace of mind, please direct any inquiries to our support team.

[1436] The server analyzes this document and transforms specific keywords and phrases as follows:

[1437] The virtual store offers a standard warranty period. For your peace of mind, please contact the virtual store's support team.

[1438] Furthermore, prompts are used to analyze user sentiment. Examples of specific prompts are shown below:

[1439] Text for when the user is in a hurry

[1440] By inputting this prompt into the AI ​​model, the user's emotional state of being in a hurry is analyzed. Based on this information, the server performs appropriate text conversion.

[1441] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1442] Step 1:

[1443] The user uploads a document file using their terminal. The input is the document file itself, which is sent to the server. The output is the document file stored on the server.

[1444] Step 2:

[1445] The server extracts text data in an appropriate manner depending on the format of the uploaded document file. The input is the document file format (PDF, Word, etc.), and the data processing involves using PDF parsing modules or Word parsing modules to obtain the text data. The output is the extracted text data.

[1446] Step 3:

[1447] The server analyzes and tokenizes the extracted text data. The input is the extracted text data, and the data processing utilizes natural language processing techniques to break down the data into words and phrases. The output is the tokenized text data.

[1448] Step 4:

[1449] The server detects specific keywords and phrases from tokenized text data. The input is tokenized text data, and the data operation involves detecting keywords and phrases based on predefined transformation rules. The output is the detected keywords and phrases.

[1450] Step 5:

[1451] The server uses an emotion engine (generative AI model) to analyze the user's emotions. The input consists of the user's operation history and uploaded documents. As part of the data processing, prompt sentences are generated and input into the generative AI model to analyze the user's emotional state. The output is the result of the user's emotion analysis.

[1452] Step 6:

[1453] The server replaces detected keywords and phrases with new expressions based on predefined transformation rules. The input consists of detected keywords and phrases, as well as the user sentiment analysis results. The server performs the replacement process based on predefined transformation rules as a data calculation. The output is the replaced text data.

[1454] Step 7:

[1455] The server reconstructs the replaced text data and converts it back to the original document format. The input is the replaced text data, and the data processing involves converting it back to the original document format (PDF, Word, etc.). The output is the reconstructed document file.

[1456] Step 8:

[1457] The server saves the reconstructed document file and prepares it for the user to download. The input is the reconstructed document file, and a download link to the user's terminal is generated. The output is the download link provided to the user.

[1458] Specific operations include: users dragging and dropping files to upload them; extracting and analyzing text data on the server; analyzing responses to prompts using AI models generated by the sentiment engine; dynamic text replacement processing according to transformation rules; and providing reconstructed documents via generated download links.

[1459] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1460] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1461] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1462] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1463] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1464] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1465] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1466] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1467] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1468] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1469] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1470] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1471] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1472] 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.

[1473] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1474] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1475] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1476] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1477] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1478] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1479] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1480] The following is further disclosed regarding the embodiments described above.

[1481] (Claim 1)

[1482] A means of receiving a document file and analyzing it as text data,

[1483] A means for detecting specific keywords or phrases from analyzed text data,

[1484] A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules,

[1485] A means of reconstructing the replaced text data as a document file,

[1486] A means of outputting a reconstructed document file,

[1487] A system that includes this.

[1488] (Claim 2)

[1489] The system according to claim 1, further comprising means for extracting an uploaded document file as text data in an appropriate manner, depending on the format of the document file.

[1490] (Claim 3)

[1491] The system according to claim 1, further comprising means for utilizing natural language processing techniques to replace detected keywords or phrases.

[1492] "Example 1"

[1493] (Claim 1)

[1494] A means of receiving a document file and analyzing it as text data,

[1495] A tokenization method that divides the analyzed text data into word or phrase units,

[1496] A means for detecting specific keywords or phrases from tokenized text data,

[1497] A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules,

[1498] A means of reconstructing the replaced text data as a document file,

[1499] A means of outputting a reconstructed document file,

[1500] A system that includes this.

[1501] (Claim 2)

[1502] The system according to claim 1, further comprising means for extracting an uploaded document file as text data in an appropriate manner, depending on the format of the document file.

[1503] (Claim 3)

[1504] The system according to claim 1, further comprising means for utilizing natural language processing techniques to replace detected keywords or phrases.

[1505] "Application Example 1"

[1506] (Claim 1)

[1507] A means of receiving a document file and analyzing it as text data,

[1508] A means for detecting specific keywords or phrases from analyzed text data,

[1509] A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules,

[1510] A means of reconstructing the replaced text data as a document file,

[1511] A means of outputting a reconstructed document file,

[1512] A means of converting the converted text data into audio data,

[1513] A means for outputting the converted audio data,

[1514] A system that includes this.

[1515] (Claim 2)

[1516] The system according to claim 1, further comprising means for extracting an uploaded document file as text data in an appropriate manner, depending on the format of the document file.

[1517] (Claim 3)

[1518] The system according to claim 1, further comprising means for utilizing natural language processing techniques to replace detected keywords or phrases.

[1519] "Example 2 of combining an emotion engine"

[1520] (Claim 1)

[1521] A means of receiving a document file and analyzing it as text data,

[1522] A means of tokenizing the analyzed text data,

[1523] A means for detecting specific keywords or phrases from tokenized text data,

[1524] A means of analyzing the emotional state of users,

[1525] A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules, and further dynamically adjusting the transformation rules based on the user's emotional state,

[1526] A means of reconstructing the replaced text data as a document file,

[1527] A means of outputting a reconstructed document file,

[1528] A system that includes this.

[1529] (Claim 2)

[1530] The system according to claim 1, further comprising means for extracting an uploaded document file as text data in an appropriate manner, depending on the format of the document file.

[1531] (Claim 3)

[1532] The system according to claim 1, further comprising means for utilizing natural language processing techniques to replace detected keywords or phrases.

[1533] "Application example 2 when combining with an emotional engine"

[1534] (Claim 1)

[1535] A means of receiving a document file and analyzing it as text data,

[1536] A means for detecting specific keywords or phrases from analyzed text data,

[1537] A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules,

[1538] A means of analyzing user emotions and dynamically adjusting document transformation rules based on those emotions,

[1539] A means of reconstructing the replaced text data as a document file,

[1540] A means of outputting a reconstructed document file,

[1541] A system that includes this.

[1542] (Claim 2)

[1543] The system according to claim 1, further comprising means for extracting an uploaded document file as text data in an appropriate manner, depending on the format of the document file.

[1544] (Claim 3)

[1545] The system according to claim 1, further comprising means for utilizing natural language processing techniques to replace detected keywords or phrases.

[1546] (Claim 4)

[1547] The system according to claim 1, further comprising means for utilizing a generative AI model to analyze user emotions.

[1548] (Claim 5)

[1549] The system according to claim 1, further comprising means for generating prompt statements to support the user's smooth operation of the converted document. [Explanation of symbols]

[1550] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving a document file and analyzing it as text data, A means for detecting specific keywords or phrases from analyzed text data, A means for replacing detected keywords or phrases with new expressions according to predefined transformation rules, A means of reconstructing the replaced text data as a document file, A means of outputting a reconstructed document file, A system that includes this.

2. The system according to claim 1, further comprising means for extracting an uploaded document file as text data in an appropriate manner, depending on the format of the document file.

3. The system according to claim 1, further comprising means for using natural language processing techniques to replace detected keywords or phrases.

Images