Intelligent network invoice issuing method, device and system based on AI large model

Abstract

The invention discloses an intelligent network invoice issuing method, device and system based on an AI large model. The method comprises the following steps: acquiring an invoice issuing related file input by a user from an AI invoice issuing page; generating a recognition request according to the invoice issuing related file, sending the recognition request to an OCR character recognition system, and receiving analyzed character information returned by the OCR character recognition system; calling a GPT finance and taxation industry large language model service according to the analyzed character information, and obtaining invoice data generated by the GPT finance and taxation industry large language model service for the analyzed character information, the invoice data including at least one of the following items: purchaser information, seller information, invoice details and amount; and filling the invoice data into an AI invoice issuing page, and after obtaining a manual confirmation message, generating a formatted invoice aiming at the invoice issuing related file input by the user. Therefore, the working efficiency of the billing service can be improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and invoice issuance technology, and in particular to an intelligent network invoice issuance method, device and system based on an AI big model. Background Art

[0002] The existing online invoice system provides online invoice issuance services to millions of users, and the daily business volume processed is roughly several million invoices.

[0003] When issuing an online invoice, the user is required to manually enter the invoice information details on the human-computer interaction page of the online invoice system. The user can click the issue button only after checking that the invoice information is correct. The online invoice system will then issue a formatted invoice based on the invoice information details.

[0004] Therefore, when a large number of invoices need to be issued, users need to invest a lot of time and effort to complete it, and there is a possibility of errors. Summary of the invention

[0005] In view of this, the present invention proposes an intelligent network invoice issuance method, device and system based on an AI big model, aiming to solve the problem of time-consuming and labor-intensive issuance of a large number of invoices in the prior art.

[0006] In a first aspect, the present invention provides an intelligent network invoice issuance method based on an AI big model, comprising:

[0007] Obtain invoice issuance related documents input by the user from the AI ​​invoice issuance page, wherein the invoice issuance related documents include at least one of the following: bill picture, mobile phone screenshot, handwritten bill picture, contract picture, word document, excel document;

[0008] Generate a recognition request based on the invoice issuance related documents and send it to the OCR text recognition system, and receive the parsed text information returned by the OCR text recognition system;

[0009] Calling the GPT finance and tax industry language model service according to the parsed text information to obtain the invoice data generated by the GPT finance and tax industry language model service for the parsed text information, wherein the invoice data includes at least one of the following: buyer information, seller information, invoice details, and amount;

[0010] The invoice data is filled into the AI ​​invoice issuance page, and after obtaining the manual confirmation message, a format invoice is generated for the invoice issuance related documents input by the user.

[0011] Furthermore, the OCR text recognition system executes an image preprocessing algorithm, a text detection algorithm and a character segmentation algorithm, performs intelligent denoising, tilt correction and / or dictionary matching on the invoice issuance related documents, and generates parsed text information corresponding to the invoice issuance related documents.

[0012] Furthermore, the GPT finance and tax industry language model is an AI model.

[0013] The GPT finance and tax industry language model analyzes and identifies the parsed text information, extracts invoice data related to invoice issuance, and assembles the invoice data into json format data and returns it.

[0014] Furthermore, the AI ​​invoice issuance page adopts a distributed microservice architecture to call the GPT finance and taxation industry large language model service or the OCR text recognition system.

[0015] In a second aspect, the present invention provides an intelligent network invoice issuing device based on an AI big model, comprising:

[0016] An acquisition unit, used to acquire invoice issuance related documents input by the user from the AI ​​invoice issuance page, wherein the invoice issuance related documents include at least one of the following: a bill picture, a mobile phone screenshot, a handwritten bill picture, a contract picture, a word document, and an excel document;

[0017] A parsing unit, used to generate a recognition request according to the invoice issuance related documents and send it to the OCR text recognition system, and receive the parsed text information returned by the OCR text recognition system;

[0018] An invoice data generating unit is used to call the GPT finance and tax industry large language model service according to the parsed text information, and obtain the invoice data generated by the GPT finance and tax industry large language model service for the parsed text information, wherein the invoice data includes at least one of the following: purchaser information, seller information, invoice details, and amount;

[0019] The invoice issuing unit is used to fill the invoice data into the AI ​​invoice issuing page, and after obtaining the manual confirmation message, generate a format invoice for the invoice issuance related documents input by the user.

[0020] Furthermore, the OCR text recognition system executes an image preprocessing algorithm, a text detection algorithm and a character segmentation algorithm, performs intelligent denoising, tilt correction and / or dictionary matching on the invoice issuance related documents, and generates parsed text information corresponding to the invoice issuance related documents.

[0021] Furthermore, the GPT finance and tax industry language model is an AI model.

[0022] The GPT finance and tax industry language model analyzes and identifies the parsed text information, extracts invoice data related to invoice issuance, and assembles the invoice data into json format data and returns it.

[0023] Furthermore, the AI ​​invoice issuance page adopts a distributed microservice architecture to call the GPT finance and taxation industry large language model service or the OCR text recognition system.

[0024] In a third aspect, the present invention provides a terminal, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0025] In a fourth aspect, the present invention provides a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the method described in the first aspect.

[0026] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0028] Figure 1 A flowchart of an intelligent network invoice issuance method based on an AI big model according to an embodiment of the present invention;

[0029] Figure 2 This is a schematic diagram of the use of the intelligent network invoice issuance system based on the AI ​​big model according to an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram of the composition of an intelligent network invoice issuing device based on an AI big model according to an embodiment of the present invention;

[0031] Figure 4 The present invention is a schematic diagram of the components of a terminal to which the method of the present invention is applied. DETAILED DESCRIPTION

[0032] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0033] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0034] Generative Pre-trained Transformer model, Generative Pre-trained Transformer, referred to as GPT. Optical Character Recognition, Optical Character Recognition, referred to as OCR.

[0035] Currently, when issuing invoices, users need to manually fill in a large amount of invoice information such as buyer information, seller information, invoice details, amount, etc. When the invoice details issued by the user are very complicated and the number of invoices to be issued is very large, it takes a lot of manpower, time, effort and responsibility to complete the work.

[0036] To this end, the present invention proposes a technical solution for intelligent network invoice issuance based on AI big model, including an intelligent network invoice issuance method, device and system based on AI big model. The intelligent network invoice issuance method based on AI big model uses AI big model technology and algorithm design to intelligently and automatically identify and parse invoice data, and intelligently and automatically issue invoices. In this way, by using the developed intelligent network invoice issuance device and system based on AI big model, it is possible to intelligently and automatically issue invoices on the network, improve the work efficiency of the invoice issuance business, save labor costs, and realize the intelligence of network invoice issuance business products.

[0037] like Figure 1 As shown, the intelligent network invoice issuance method based on the AI ​​big model of the embodiment of the present invention includes the following steps:

[0038] S100: Acquire invoice issuance related documents input by the user from the AI ​​invoice issuance page, wherein the invoice issuance related documents include at least one of the following: a bill image, a mobile phone screenshot, a handwritten bill image, a contract image, a word document, and an excel document;

[0039] S200: generating a recognition request according to the invoice issuance related documents and sending it to the OCR text recognition system, and receiving the parsed text information returned by the OCR text recognition system;

[0040] S300: Calling the GPT finance and tax industry language model service according to the parsed text information to obtain the invoice data generated by the GPT finance and tax industry language model service for the parsed text information, wherein the invoice data includes at least one of the following: buyer information, seller information, invoice details, and amount;

[0041] S400: Fill the invoice data into the AI ​​invoice issuance page, and after obtaining the manual confirmation message, generate a format invoice for the invoice issuance related documents input by the user.

[0042] Furthermore, the OCR text recognition system executes an image preprocessing algorithm, a text detection algorithm and a character segmentation algorithm, performs intelligent denoising, tilt correction and / or dictionary matching on the invoice issuance related documents, and generates parsed text information corresponding to the invoice issuance related documents.

[0043] Furthermore, the GPT finance and tax industry language model is an AI model.

[0044] The GPT finance and tax industry language model analyzes and identifies the parsed text information, extracts invoice data related to invoice issuance, and assembles the invoice data into json format data and returns it.

[0045] Furthermore, the AI ​​invoice issuance page adopts a distributed microservice architecture to call the GPT finance and taxation industry large language model service or the OCR text recognition system.

[0046] Compared with the existing online invoice system, the intelligent online invoice issuance method based on the AI ​​big model in the embodiment of the present invention realizes the automation and intelligence of online invoice issuance, which can greatly improve work efficiency and save labor costs.

[0047] like Figure 2 As shown in the figure, the design and development of an intelligent network invoice issuance system based on an AI big model can include the following steps:

[0048] S1. Build a big AI model. Build the GPT finance and taxation industry language model (referred to as finance and taxation GPT). Based on years of accumulated business experience in the finance and taxation industry and massive finance and taxation data, generate and continuously update the finance and taxation GPT through reinforcement learning technologies such as supervised fine-tuning, reward learning, and human alignment.

[0049] In this way, massive amounts of financial and tax data are used to train a large AI model, namely the financial and tax GPT. The AI ​​large model, the financial and tax GPT, is used to identify and extract financial and tax or invoice data.

[0050] The finance and taxation GPT can also provide intelligent finance and taxation consulting, automated finance and taxation processing, personalized finance and taxation solutions and other services to corporate and institutional users in the finance and taxation field.

[0051] Furthermore, if Figure 2 As shown in the figure, based on the GPT finance and tax industry language model, an AI big model service is developed. The AI ​​big model service will train a large amount of finance and taxation-related data, and will also do a lot of data testing. When it is found that the invoice information extracted by the AI ​​big model for certain text content is inaccurate, the model recognition and extraction algorithm is adjusted to finally reach the requirement of accurately extracting the data content related to the invoice. In this way, intelligent recognition of invoice data and automatic filling of invoice information can be achieved.

[0052] S2. Develop an OCR text recognition system. Develop a high-performance OCR text recognition system to recognize and analyze the text information recorded in various invoice-related documents, and achieve accurate text recognition and analysis. For example, it can recognize and analyze various types of image information, such as bill images, mobile phone screenshots, handwritten bill images (such as scans or photos), contract images (such as scans or photos); it can recognize various types of files, such as word documents, excel documents, etc.

[0053] S3. Build an invoice issuance system with a distributed microservice architecture, such as building a distributed microservice architecture. The high-performance and highly available distributed microservice architecture can support billions of visits and millions of invoices per day, and provide high-performance and stable invoice system services.

[0054] In this way, invoices can be automatically issued, and massive invoice issuance business requests can be accepted, ensuring efficient and stable operation of the system.

[0055] In this way, the intelligent network invoice issuance method based on the AI ​​big model of the embodiment of the present invention can use the AI ​​big model, OCR text recognition system, and distributed microservice architecture to realize the issuance of invoices for massive invoice issuance services. Compared with the existing network invoice system, the work efficiency of the invoice issuance business is greatly improved, and the AI ​​intelligentization and automation of the invoice issuance business are realized at the same time.

[0056] like Figure 2 As shown, when a user uses the intelligent network invoice issuance system based on the AI ​​big model to execute the intelligent network invoice issuance method based on the AI ​​big model, the following steps may be included:

[0057] Step 1: Extract the text information recorded in the invoice issuance documents

[0058] The user opens the AI ​​invoice issuance page in the invoice issuance system and uploads invoice issuance related files such as word documents, excel documents, pictures, etc. The invoice issuance related files can be table data, receipt information, contract information, screenshot information, etc.

[0059] After the user uploads these contents, the invoice issuance system calls the OCR text recognition system to recognize the text content in word documents, excel documents, pictures, etc. Specifically, the pictures can be irregular pictures, blurred pictures, pictures with handwritten content, etc.

[0060] The OCR text recognition system executes image preprocessing algorithms, text detection algorithms and character segmentation algorithms to perform intelligent denoising, tilt correction, and dictionary matching on invoice-related documents uploaded by users, such as various irregular images, blurred images, and images with handwritten content, and ultimately accurately identifies and extracts the text information recorded in the documents.

[0061] Step 2: Generate invoice data using text information

[0062] After accurately parsing the text content (such as text information) in various files uploaded by users through the OCR text recognition system, the AI ​​big model service is called using the parsed text content.

[0063] The AI ​​big model service performs intelligent analysis and recognition on the text content parsed from the files uploaded by users, accurately extracts invoice-related data content, such as buyer information, seller information, invoice details, amount, etc., and then intelligently assembles it into JSON format data as invoice issuance information (such as invoice data) and returns it to the invoice issuance system.

[0064] Step 3: Generate a formatted invoice and store, display and send it to the designated recipient

[0065] When the invoice issuance system receives the assembled invoice issuance information returned by the AI ​​big model service, it automatically fills it into the corresponding position or control of the invoice issuance page and displays it as the corresponding invoice information. In this way, when issuing an invoice, the user no longer needs to manually enter the invoice issuance information. After the user checks and verifies that the automatically filled invoice information is correct, he can directly click the invoice issuance button to complete the invoice issuance operation.

[0066] After using the intelligent invoice issuance system based on AI big model, users only need to upload some form data, receipt information, contract information, screenshot information and other files. The online invoice issuance system will automatically and intelligently identify the information that needs to be invoiced in the uploaded file, and then intelligently parse and extract the invoice data, automatically fill it into the invoice issuance page information, and finally complete the invoice issuance. This process does not require users to manually enter invoice information, which greatly improves work efficiency and reduces labor costs.

[0067] The present invention uses the GPT finance and taxation industry large language model to realize the true AI and intelligence of invoice issuance products, leading the industry trend and indicating the future development direction and potential of the finance and taxation industry.

[0068] like Figure 3 As shown, the intelligent network invoice issuing device 1000 based on the AI ​​big model according to the embodiment of the present invention includes:

[0069] The acquisition unit 10 is used to acquire the invoice issuance related documents input by the user from the AI ​​invoice issuance page, wherein the invoice issuance related documents include at least one of the following: a bill picture, a mobile phone screenshot, a handwritten bill picture, a contract picture, a word document, and an excel document;

[0070] The parsing unit 20 is used to generate a recognition request according to the invoice issuance related documents and send it to the OCR text recognition system, and receive the parsed text information returned by the OCR text recognition system;

[0071] The invoice data generating unit 30 is used to call the GPT finance and tax industry language model service according to the parsed text information, and obtain the invoice data generated by the GPT finance and tax industry language model service for the parsed text information, wherein the invoice data includes at least one of the following: buyer information, seller information, invoice details, and amount;

[0072] The invoice issuing unit 40 is used to fill the invoice data into the AI ​​invoice issuing page, and after obtaining the manual confirmation message, generate a format invoice for the invoice issuance related documents input by the user.

[0073] Furthermore, the OCR text recognition system executes an image preprocessing algorithm, a text detection algorithm and a character segmentation algorithm, performs intelligent denoising, tilt correction and / or dictionary matching on the invoice issuance related documents, and generates parsed text information corresponding to the invoice issuance related documents.

[0074] Furthermore, the GPT finance and tax industry language model is an AI model.

[0075] The GPT finance and tax industry language model analyzes and identifies the parsed text information, extracts invoice data related to invoice issuance, and assembles the invoice data into json format data and returns it.

[0076] Furthermore, the AI ​​invoice issuance page adopts a distributed microservice architecture to call the GPT finance and taxation industry large language model service or the OCR text recognition system.

[0077] The embodiment of the present invention also provides a terminal to execute the method. Figure 4 It shows a schematic diagram of a terminal provided by some embodiments of the present invention. Figure 4 As shown, the terminal 8 includes: a processor 800, a memory 801, a bus 802 and a communication interface 803, wherein the processor 800, the communication interface 803 and the memory 801 are connected via the bus 802; the memory 801 stores a computer program that can be run on the processor 800, and the processor 800 executes the method provided by any embodiment of the present invention when running the computer program.

[0078] The memory 801 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the device network element and at least one other network element is realized through at least one communication interface 803 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0079] The bus 802 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 801 is used to store programs, and the processor 800 executes the programs after receiving execution instructions. The method disclosed in any implementation of the embodiment of the present invention may be applied to the processor 800, or implemented by the processor 800.

[0080] The processor 800 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the method can be completed by the hardware integrated logic circuit in the processor 800 or the instruction in the form of software. The processor 800 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention may be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in a decoding processor. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 801, and the processor 800 reads the information in the memory 801 and completes the steps of the method in combination with its hardware.

[0081] The terminal provided by the embodiment of the present invention and the method provided by the embodiment of the present invention are based on the same inventive concept and have the same beneficial effects as the method adopted, run or implemented by them.

[0082] An embodiment of the present invention also provides a computer-readable storage medium corresponding to the method provided by the embodiment, and the computer-readable storage medium is a CD, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method provided by any of the embodiments.

[0083] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0084] The computer-readable storage medium provided by the embodiment of the present invention is based on the same inventive concept as the method of the embodiment of the present invention, and has the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0085] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these adjustments and modifications of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these changes and modifications.

Claims

1. An intelligent network invoice issuance method based on AI big model, characterized in that: include: Obtain invoice issuance related documents input by the user from the AI ​​invoice issuance page, wherein the invoice issuance related documents include at least one of the following: bill picture, mobile phone screenshot, handwritten bill picture, contract picture, word document, excel document; Generate a recognition request based on the invoice issuance related documents and send it to the OCR text recognition system, and receive the parsed text information returned by the OCR text recognition system; Calling the GPT finance and tax industry language model service according to the parsed text information to obtain the invoice data generated by the GPT finance and tax industry language model service for the parsed text information, wherein the invoice data includes at least one of the following: buyer information, seller information, invoice details, and amount; The invoice data is filled into the AI ​​invoice issuance page, and after obtaining the manual confirmation message, a format invoice is generated for the invoice issuance related documents input by the user.

2. The intelligent network invoice issuance method based on AI big model as claimed in claim 1 is characterized in that: The OCR text recognition system executes an image preprocessing algorithm, a text detection algorithm and a character segmentation algorithm, performs intelligent denoising, tilt correction and / or dictionary matching on the invoice issuance related documents, and generates parsed text information corresponding to the invoice issuance related documents.

3. The intelligent network invoice issuance method based on AI big model as claimed in claim 2 is characterized in that: The GPT finance and tax industry language model is an AI model. The GPT finance and tax industry language model analyzes and identifies the parsed text information, extracts invoice data related to invoice issuance, and assembles the invoice data into json format data and returns it.

4. The intelligent network invoice issuance method based on AI big model as claimed in claim 3 is characterized in that: The AI ​​invoice issuance page uses a distributed microservice architecture to call the GPT finance and taxation industry large language model service or the OCR text recognition system.

5. An intelligent network invoice issuing device based on AI big model, characterized in that: include: An acquisition unit, used to acquire invoice issuance related documents input by the user from the AI ​​invoice issuance page, wherein the invoice issuance related documents include at least one of the following: a bill picture, a mobile phone screenshot, a handwritten bill picture, a contract picture, a word document, and an excel document; A parsing unit, used to generate a recognition request according to the invoice issuance related documents and send it to the OCR text recognition system, and receive the parsed text information returned by the OCR text recognition system; An invoice data generating unit is used to call the GPT finance and tax industry large language model service according to the parsed text information, and obtain the invoice data generated by the GPT finance and tax industry large language model service for the parsed text information, wherein the invoice data includes at least one of the following: purchaser information, seller information, invoice details, and amount; The invoice issuing unit is used to fill the invoice data into the AI ​​invoice issuing page, and after obtaining the manual confirmation message, generate a format invoice for the invoice issuance related documents input by the user.

6. The intelligent network invoice issuing device based on the AI ​​big model as claimed in claim 5 is characterized in that: The OCR text recognition system executes an image preprocessing algorithm, a text detection algorithm and a character segmentation algorithm, performs intelligent denoising, tilt correction and / or dictionary matching on the invoice issuance related documents, and generates parsed text information corresponding to the invoice issuance related documents.

7. The intelligent network invoice issuing device based on AI big model as claimed in claim 5, characterized in that: The GPT finance and tax industry language model is an AI model. The GPT finance and tax industry language model analyzes and identifies the parsed text information, extracts invoice data related to invoice issuance, and assembles the invoice data into json format data and returns it.

8. The intelligent network invoice issuing device based on AI big model as claimed in claim 5, characterized in that: The AI ​​invoice issuance page uses a distributed microservice architecture to call the GPT finance and taxation industry large language model service or the OCR text recognition system.

9. A terminal, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any one of claims 1 to 4.

10. A computer storage medium, characterized in that: Computer executable instructions are stored, and the computer executable instructions are used to execute the method according to any one of claims 1 to 4.

Images