CN117764542A - Language model driven enterprise virtual office assistant system and implementation method - Google Patents

Abstract

The invention discloses a language model driven enterprise virtual office assistant system and an implementation method thereof in the technical field of virtual office assistants, wherein the system comprises a dialogue acquisition module, a dialogue recording module, a dialogue identification module, a backlog module, a query system module, a printing module and an office terminal module; the implementation method comprises the following steps: firstly, collecting dialogue content and chat records of staff by using a dialogue collection module; recording dialogue contents of staff by using a dialogue recording module, and intelligently identifying by using a dialogue identification module; and thirdly, identifying the dialogue content according to the dialogue identification module, and respectively realizing the task of creating the backlog, inquiring the company system and printing the file by using the backlog module, the inquiring system module and the printing module. The invention is simple and easy to operate, and can solve a series of working problems of enterprise staff in an office scene through cooperation among the modules, thereby improving the working efficiency.

Description

Language model driven enterprise virtual office assistant system and implementation method

Technical Field

The invention belongs to the technical field of virtual office assistants, and particularly relates to a language model driven enterprise virtual office assistant system and an implementation method.

Background

The office system is a comprehensive management method for providing data information for enterprises by using enterprise management software system engineering, enterprise management requirements as a basis and IT technology as a support. Most of the traditional office systems are based on web, and are relatively cool pages and lack of humanized communication, so that the thought of a team is broken in design, and the most original and effective communication between people is destroyed by the traditional web office systems.

In the related technology, various screen-carrying sound boxes with intelligent dialogue interaction systems exist in the market at present, the screen-carrying sound boxes are used for effectively displaying the existing intelligent dialogue technology through a screen, and the screen-carrying sound boxes are used for providing better information interaction experience, are used for entertainment and education scenes, and can be used for directly listening to songs, inquiring weather, express information, daily boring, encyclopedic knowledge answering and the like.

The scheme in the prior art cannot further expand the use scene, such as intelligent office and OA system information inquiry.

Disclosure of Invention

In order to solve the problem of further expanding the use scene in the prior art, the invention aims to provide a language model driven enterprise virtual office assistant system and an implementation method, and a series of working problems of enterprise staff in an office scene are solved through cooperation among the modules, so that the working efficiency is improved.

In order to achieve the above object, the technical scheme of the present invention is as follows: a language model driven enterprise virtual office assistant system comprises a dialogue acquisition module, a dialogue recording module, a dialogue identification module, a backlog module, a query system module, a printing module and an office terminal module;

the dialogue collection module is used for collecting dialogue and chat content initiated by staff;

the dialogue recording module is used for storing and recording the message records of the staff and is used for the system to understand the intention of the staff with the memory capacity;

the dialogue identification module is used for analyzing dialogue characters and languages initiated by the staff by utilizing the large model and judging things the staff wants to do;

the system comprises a backlog module, a backlog module and a backlog terminal module, wherein the backlog module is used for receiving identification information at the backlog identification module, when the identification information is identified as creating a backlog, the backlog acquisition module continues to talk with staff and acquire dialogue content until the staff speaks backlogs and specific reminding time, and after the backlog time and the backlog are collected, the information is transmitted to the office terminal module;

the inquiry system module is used for receiving the identification information at the dialogue identification module, when the identification information is identified as the inquiry company system, the inquiry system module is used for calling the corresponding system content according to the staff dialogue content, and when the corresponding system content is not available, a friendly prompt is given;

the printing module is used for receiving the identification information at the dialogue identification module, prompting staff to send the file to be printed and receiving the file content when the information is identified as the printed file, calling a corresponding printer driver to send the file and print, simultaneously reading the working state of the printer, and notifying the staff of the printing result after the printing is finished;

and the office terminal module is used for receiving the information at each module and working together with the modules to complete the corresponding dialogue instruction.

The basic scheme is as follows: firstly, staff sends out a conversation according to own needs, a conversation acquisition module in the system acquires conversation content of the staff and transmits the content to a conversation recording module, the conversation recording module records and stores the conversation content and transmits conversation information to a conversation identification module, and the conversation identification module identifies the conversation of the staff and identifies corresponding staff needs at the position;

when the requirement of the staff is identified to be creating the to-do reminding, the backlog module receives the backlog information at the moment, identifies and collects the project content and the project time required to be sponsored by the staff, and transmits the backlog to the office terminal module for backlog sponsoring;

when the requirement of the staff is identified as a system for inquiring the company, the inquiry system module receives the inquiry requirement of the staff at the moment and correspondingly invokes the system for inquiring the company required by the staff;

when the requirement of the staff is identified as the printing file, the printing module receives the printing information of the staff at the moment, calls the corresponding printing file, transmits the corresponding file to the printer, sends the file to the corresponding printer driver for printing, reads the working state of the printer, and informs the staff of the printing result after the printing is finished.

The basic scheme has the beneficial effects that: 1. the office assistant can normally exchange dialogue with staff through cooperation among the modules, recognize staff intention in the dialogue process, provide functional services such as relevant system inquiry, file printing, to-do task creation, knowledge question and answer, flow submission and approval for staff, and help staff administrative staff write documents, send notification, and staff birthday blessing and other matters.

2. Through the function, the office assistant can solve a series of working problems of enterprise staff in an office scene, so that working efficiency is improved, and the other party can help the enterprise to analyze staff working states and timely give help and correct through data collection.

And the system further comprises a proxy auditing module, wherein the proxy auditing module is used for auditing whether the identified proxy transaction is accurate or not, auditing the project content and time of the proxy transaction, outputting the auditing content to the office terminal module after the auditing content is accurate, and transmitting the information back to the proxy transaction module for re-auditing when the auditing is failed.

The basic scheme has the beneficial effects that: the generation checking module can check the created generation contents sent by the staff, so that the error judgment of the office assistant on the generation contents sent by the staff is reduced, and the accuracy of the office assistant in processing is improved.

And the system comprises a system classification module, wherein the system classification module is used for classifying the system of the company, correspondingly receiving the calling information at the query system module, and recording the queried frequencies of the system of the company with different classifications.

The basic scheme has the beneficial effects that: the system classification module can classify the system of the company, so that the system classification module can conveniently, efficiently and quickly retrieve corresponding system information on one hand, and can record and analyze the retrieval fear frequency of different classification system information on the other hand, and further can analyze management requirements of staff.

And when the secondary frequency reaches a threshold value, the system supplementing module transmits the system information to the office terminal module so as to remind a manager of supplementing and updating the system.

The basic scheme has the beneficial effects that: the system supplementing module can collect the system according to the company system which is not queried by staff, and after the query frequency of the system reaches a threshold value, the system supplementing module transmits corresponding information to the terminal so as to promote updating and supplementing of the system of the company.

Further, the printing module comprises a printer sequence unit and a printing sequence unit;

the printer sequence unit is used for arranging the printers in real time according to the working states of a plurality of printers at the moment and queuing the idle printers to the front row;

and the printing sequence unit is used for queuing the printing items on the printer to be utilized at the moment and identifying the waiting time among the items.

The basic scheme has the beneficial effects that: the printer sequence unit can queue different printers, so that the most idle printers are arranged under the upcoming print task, and further, the faster and more efficient printing is realized; meanwhile, the printing sequence unit can correspondingly arrange printing items at the corresponding printer, and feed back waiting time of different items, so that staff can know time required by different printing tasks at the moment more quickly, and the staff can arrange work time of the staff reasonably.

And the printing module further comprises a key judging module, wherein the key judging module is used for identifying whether the file to be printed is an encrypted file or not, if the encrypted file is needed to be provided with a password with staff, the printing is directly carried out, and if the encrypted file is not the encrypted file.

The basic scheme has the beneficial effects that: the key judgment module is designed to judge the state of the printed file, so that the important file of the company can be correspondingly kept secret, and the possibility of leakage of the important file of the company is reduced.

Further, the printer further comprises a Bluetooth connection module, wherein the Bluetooth connection module is used for wirelessly connecting the printer with the staff end so as to realize that the printer successfully receives and prints the file.

The basic scheme has the beneficial effects that: the Bluetooth connection module can enable the printer to be connected with the terminal, and further facilitates smooth printing of a subsequent printer.

The printer and the staff end are connected successfully, if the printer and the staff end are connected successfully, the printer and the staff end directly speak the file to print, and if the printer and the staff end are not connected smoothly, the printer and the staff end are connected through Bluetooth again.

The basic scheme has the beneficial effects that: the connection judging module can judge the connection condition of Bluetooth, so that whether the printer prints or not is judged, and smooth printing of the printer is realized.

Further, the method for realizing the language model driven enterprise virtual office assistant system comprises the following steps:

firstly, collecting dialogue content and chat records of staff by using a dialogue collection module;

recording dialogue contents of staff by using a dialogue recording module, intelligently identifying the dialogue contents by using a dialogue identification module, and transmitting the corresponding identified contents to the next stage;

and thirdly, identifying the dialogue content according to the dialogue identification module, and respectively realizing the task of creating the backlog, inquiring the company system and printing the file by using the backlog module, the inquiring system module and the printing module.

The basic scheme has the beneficial effects that: the system recognizes the intention of the staff through a natural language dialogue mode with the staff, integrates software and hardware such as a nailing system, a document management system and a printer, and provides corresponding functional services for different working demands of the staff, so that the system helps the enterprise to analyze the working state of the staff and timely gives help and correction.

Drawings

FIG. 1 is a system diagram of a language model driven enterprise virtual office assistant system in accordance with an embodiment of the present invention.

Fig. 2 is a schematic diagram of a method for implementing a language model driven enterprise virtual office assistant system according to an embodiment of the present invention.

Detailed Description

The following is a further detailed description of the embodiments:

example 1

Substantially as shown in figure 1: a language model driven enterprise virtual office assistant system comprises a dialogue acquisition module, a dialogue recording module, a dialogue identification module, a proxy auditing module, a proxy item module, a query system module, a system classification module, a system supplementing module, a printing module and an office terminal module;

the dialogue collection module is used for collecting dialogue and chat content initiated by staff;

the dialogue recording module is used for storing and recording the message records of the staff and is used for the system to understand the intention of the staff with the memory capacity;

the dialogue identification module is used for analyzing dialogue characters and languages initiated by the staff by utilizing the large model and judging things the staff wants to do;

the agent auditing module is used for auditing whether the identified agent is accurate or not, auditing the item content and time of the agent, outputting the auditing content to the office terminal module after the auditing content is accurate, and transmitting the information back to the agent module for re-auditing when the auditing is failed;

the system comprises a backlog module, a backlog module and a backlog terminal module, wherein the backlog module is used for receiving identification information at the backlog identification module, when the identification information is identified as creating a backlog, the backlog acquisition module continues to talk with staff and acquire dialogue content until the staff speaks backlogs and specific reminding time, and after the backlog time and the backlog are collected, the information is transmitted to the office terminal module;

the inquiry system module is used for receiving the identification information at the dialogue identification module, when the identification information is identified as the inquiry company system, the inquiry system module is used for calling the corresponding system content according to the staff dialogue content, and when the corresponding system content is not available, a friendly prompt is given;

the system classification module is used for classifying the company systems, correspondingly receiving the calling information at the inquiry system module and recording the inquired frequencies of the company systems with different classifications;

the system supplementing module is used for receiving the system information which is not queried when the staff queries the system of the company, recording the query frequency of the system according to the staff, and transmitting the system information to the office terminal module when the secondary frequency reaches a threshold value so as to remind the manager of supplementing and updating the system;

the printing module is used for receiving the identification information at the dialogue identification module, prompting staff to send the file to be printed and receiving the file content when the information is identified as the printed file, calling a corresponding printer driver to send the file and print, simultaneously reading the working state of the printer, and notifying the staff of the printing result after the printing is finished;

and the office terminal module is used for receiving the information at each module and working together with the modules to complete the corresponding dialogue instruction.

The specific implementation process is as follows: the staff sends out dialogue according to own demand, dialogue collection module gathers staff's dialogue content, then dialogue record module keeps with staff's message record and is used for the system to take memory ability to understand staff's intention. For example, an employee firstly speaks to the system that a to-be-done reminder of 'project start meeting' is created, the reminding time is 3 pm on 6 months and 9 days, and the system searches for 'event' and 'time' information in the context by two dialogue inputs, so that the dialogue identification module carries out AI identification on the intention and judges things the employee wants to do;

the method comprises the steps that a backlog module, a query system module and a printing module are respectively utilized for realizing the things which the personnel want to do, wherein when the personnel want to create the backlog, the backlog module receives backlog information at the moment, identifies and collects project contents and project time which the personnel need to do, and meanwhile, the backlog auditing module audits the identified information so that the transmitted backlog information is accurate and real, and then the office terminal module performs backlog sponsor;

when an employee wants to inquire the company system, the inquiry system module extracts different types of company system information required by the employee through the system classification module, so that the efficient extraction of the company system is achieved; meanwhile, when the staff wants to print the file, the printing module receives the staff printing information at the moment, calls the corresponding file to be printed, transmits the corresponding file to the printer, sends the file to the corresponding printer driver for printing, reads the working state of the printer, and informs the staff of the printing result after the printing is finished.

Example 2

The difference from the above-described embodiment is that the printing module includes a printer serial unit and a printing serial unit;

the printer sequence unit is used for arranging the printers in real time according to the working states of a plurality of printers at the moment and queuing the idle printers to the front row;

and the printing sequence unit is used for queuing the printing items on the printer to be utilized at the moment and identifying the waiting time among the items.

The specific implementation process is as follows: the printer sequence unit will arrange the printers, for example, five printers are arranged at this time, wherein the printers are correspondingly arranged according to the number of the printing items of the printers, and the fewer the printing items of the printers are arranged in the front row; meanwhile, the printing sequence unit sorts the items on the corresponding printer, for example, when three printing items are printed on the printer A, the printing sequence unit analyzes and calculates the time of the three printing items respectively and the corresponding waiting time is convenient for staff to communicate.

Example 3

The printing module further comprises a key judging module, wherein the key judging module is used for identifying whether the file to be printed is an encrypted file or not, if so, the encrypted file needs to be provided with a password with staff, and if not, the printing is directly carried out.

The specific implementation process is as follows: the key judgment module is designed to judge the state of the printed file, so that the important file of the company can be correspondingly kept secret, and the possibility of leakage of the important file of the company is reduced.

Example 4

The difference with the embodiment is that the device also comprises a Bluetooth connection module and a connection judgment module;

and the Bluetooth connection module is used for wirelessly connecting the printer with the staff end so as to realize that the printer successfully receives and prints the file.

And the connection judging module is used for judging whether the printer and the staff end are successfully connected at the moment, if so, directly speaking the file for printing, and if not, transmitting the information to the staff end for Bluetooth connection again.

The specific implementation process is as follows: the printer is connected with the terminal by utilizing the Bluetooth connection module; meanwhile, the connection judging module judges the connection relation between the connection judging module and the printer in time so as to realize successful printing of the printer.

Example 5

Unlike the above embodiment, as shown in fig. 2, a method for implementing a language model driven enterprise virtual office assistant system includes the following steps:

firstly, collecting dialogue content and chat records of staff by using a dialogue collection module;

recording dialogue contents of staff by using a dialogue recording module, intelligently identifying the dialogue contents by using a dialogue identification module, and transmitting the corresponding identified contents to the next stage;

and thirdly, identifying the dialogue content according to the dialogue identification module, and respectively realizing the task of creating the backlog, inquiring the company system and printing the file by using the backlog module, the inquiring system module and the printing module.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

The foregoing is merely an embodiment of the present invention, and a specific structure and characteristics of common knowledge in the art, which are well known in the scheme, are not described herein, so that a person of ordinary skill in the art knows all the prior art in the application day or before the priority date of the present invention, and can know all the prior art in the field, and have the capability of applying the conventional experimental means before the date, so that a person of ordinary skill in the art can complete and implement the present embodiment in combination with his own capability in the light of the present application, and some typical known structures or known methods should not be an obstacle for a person of ordinary skill in the art to implement the present application. It should be noted that modifications and improvements can be made by those skilled in the art without departing from the structure of the present invention, and these should also be considered as the scope of the present invention, which does not affect the effect of the implementation of the present invention and the utility of the patent. The protection scope of the present application shall be subject to the content of the claims, and the description of the specific embodiments and the like in the specification can be used for explaining the content of the claims.

Claims (9)

1. A language model driven enterprise virtual office assistant system, characterized by: the system comprises a dialogue acquisition module, a dialogue recording module, a dialogue identification module, a backlog module, a query system module, a printing module and an office terminal module;

the dialogue collection module is used for collecting dialogue and chat content initiated by staff;

the dialogue recording module is used for storing and recording the message records of the staff and is used for the system to understand the intention of the staff with the memory capacity;

the dialogue identification module is used for analyzing dialogue characters and languages initiated by the staff by utilizing the large model and judging things the staff wants to do;

the system comprises a backlog module, a backlog module and a backlog terminal module, wherein the backlog module is used for receiving identification information at the backlog identification module, when the identification information is identified as creating a backlog, the backlog acquisition module continues to talk with staff and acquire dialogue content until the staff speaks backlogs and specific reminding time, and after the backlog time and the backlog are collected, the information is transmitted to the office terminal module;

the inquiry system module is used for receiving the identification information at the dialogue identification module, when the identification information is identified as the inquiry company system, the inquiry system module is used for calling the corresponding system content according to the staff dialogue content, and when the corresponding system content is not available, a friendly prompt is given;

the printing module is used for receiving the identification information at the dialogue identification module, prompting staff to send the file to be printed and receiving the file content when the information is identified as the printed file, calling a corresponding printer driver to send the file and print, simultaneously reading the working state of the printer, and notifying the staff of the printing result after the printing is finished;

and the office terminal module is used for receiving the information at each module and working together with the modules to complete the corresponding dialogue instruction.

2. The language model driven enterprise virtual office assistant system of claim 1, wherein: the system also comprises a proxy auditing module, wherein the proxy auditing module is used for auditing whether the identified proxy is accurate or not, auditing the project content and time of the proxy, outputting the auditing content to the office terminal module after the auditing content is accurate, and transmitting the information back to the proxy module for re-auditing when the auditing is failed.

3. The language model driven enterprise virtual office assistant system of claim 2, wherein: the system comprises a system classification module, a system query module and a system query module, wherein the system classification module is used for classifying company systems, correspondingly receiving the calling information at the query system module, and recording the queried frequencies of the company systems of different classifications.

4. The language model driven enterprise virtual office assistant system of claim 3, wherein: the system comprises a system management module, a system management module and a system management module, wherein the system management module is used for managing the system management module, and is used for managing the system management module.

5. The language model driven enterprise virtual office assistant system of claim 4, wherein: the printing module comprises a printer sequence unit and a printing sequence unit;

the printer sequence unit is used for arranging the printers in real time according to the working states of a plurality of printers at the moment and queuing the idle printers to the front row;

and the printing sequence unit is used for queuing the printing items on the printer to be utilized at the moment and identifying the waiting time among the items.

6. The language model driven enterprise virtual office assistant system of claim 5, wherein: the printing module also comprises a key judging module which is used for identifying whether the file to be printed is an encrypted file or not, if the encrypted file needs to be provided with a password with staff, the printing is directly carried out, and if the encrypted file is not the encrypted file.

7. The language model driven enterprise virtual office assistant system of claim 6, wherein: the printer is connected with the staff end in a wireless mode, and the printer is connected with the staff end in a wireless mode.

8. The language model driven enterprise virtual office assistant system of claim 7, wherein: the printer and the staff end are connected successfully, if the printer and the staff end are connected successfully, the file is directly transferred to be printed, and if the printer and the staff end are not connected successfully, the information is transferred to the staff end to be connected again through Bluetooth.

9. A method for realizing a language model driven enterprise virtual office assistant system is characterized in that: the method comprises the following steps:

firstly, collecting dialogue content and chat records of staff by using a dialogue collection module;

recording dialogue contents of staff by using a dialogue recording module, intelligently identifying the dialogue contents by using a dialogue identification module, and transmitting the corresponding identified contents to the next stage;

and thirdly, identifying the dialogue content according to the dialogue identification module, and respectively realizing the task of creating the backlog, inquiring the company system and printing the file by using the backlog module, the inquiring system module and the printing module.