A speech recognition method and system for engineering project management based on artificial intelligence

By dismantling the project management process into a minimum process and combining artificial intelligence recognition, the problems of low efficiency and high cost of speech recognition in engineering project management are solved, and efficient and low-cost management process push and recognition effects are achieved.

CN118761620BActive Publication Date: 2024-11-29XIAMEN LIANJIAN TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411244352.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-11-29
Estimated Expiration
2044-09-06

AI Technical Summary

Technical Problem

In the management of existing engineering projects, speech recognition is low efficiency and high cost, mainly due to the increased possibility of recognition failure due to multi-level calls, accent problems and noisy environment affect information transmission efficiency, and algorithm development costs are high.

Method used

Adopt artificial intelligence-based voice recognition method for engineering project management and control, and summarize and decompose management processes into minimum processes, combine identity information, positioning and time marking to identify and sort voice information, reduce noise impact, improve recognition accuracy, and reduce Process content and improve management efficiency.

Benefits of technology

By dismantling the management process into a minimum process, we can improve screening speed and management efficiency, reduce recognition failure rate, overcome accent problems, reduce costs, improve voice recognition efficiency, and achieve low-cost engineering project management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761620B_ABST
    Figure CN118761620B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of project management, and in particular to an artificial intelligence-based engineering project management and control voice recognition method and system, the method comprising: a staff member sends voice information and positioning information to a processing end for the first time through a bound terminal; the artificial intelligence decomposes all management processes into minimum processes, marks the work location, working time period and a number of staff members allowed to access each minimum process to form a minimum access block; counts the number of times staff members allowed to access all minimum access blocks; the artificial intelligence determines the identity of the staff member through voice information; pushes in different ways according to the number of times the staff member is allowed to access; the present invention decomposes all management processes into minimum processes and forms minimum access blocks, that is, switches each minimum process into an independent calling process; directly predicts the minimum process that it can access through voice recognition, so as to greatly improve management efficiency and response speed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of project management, and in particular to an artificial intelligence-based speech recognition method and system for engineering project management and control. Background Art

[0002] Existing projects are often divided into multiple construction projects and different staff members. Each construction project has its own management process, and each staff member has different permissions to access the management process, which leads to confusion in project management. Existing technologies require step-by-step invocation of processes through voice recognition (a question-and-answer model), requiring repeated voice recognition. This increased likelihood of recognition failures leads to decreased efficiency and reduced staff trust in voice recognition, which is why voice recognition is difficult to promote in project management. Furthermore, since most staff members come from all corners of the country and have accents, their voices are unclear, and the noisy work environment results in inefficient information transmission. To overcome these issues, voice recognition training must be strengthened, which also leads to increased costs. Active noise reduction is required to reduce noise, but this requires high chip computing power and algorithms, resulting in high costs. Furthermore, the diverse workplaces and staff members make algorithm development extremely expensive. The industry consensus is that universal voice recognition is extremely expensive, so achieving project management at a lower cost is an urgent issue that needs to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an engineering project management and control speech recognition method and system based on artificial intelligence.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] An artificial intelligence-based speech recognition method for engineering project management and control, comprising:

[0006] The staff sends the voice message and location information to the processing end for the first time through the bound terminal;

[0007] Through artificial intelligence, all management processes included in the project are summarized to generate a process summary table. All management processes in the process summary table are decomposed until the minimum process is obtained. The work location, working time period and number of staff members allowed to access each minimum process are marked to form a minimum access block. The number of staff members allowed to access all minimum access blocks is counted.

[0008] Entering the staff member's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice recordings of the staff member reading words into the database;

[0009] Artificial intelligence timestamps the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, and obtains the staff identity information by comparing the second terminal information with the database. Artificial intelligence recognizes the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if so, it determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value. If so, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp and the voice in the database through artificial intelligence. The voice information is sorted after semantic recognition; if the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time stamp and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and it is determined whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, the voice information is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, the voice information sent for the second time is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to push it to the terminal.

[0010] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0011] An artificial intelligence-based engineering project management and control speech recognition system, comprising:

[0012] Terminal: The staff sends voice information and positioning information to the processing end for the first time through the bound terminal;

[0013] The processing end, including artificial intelligence and databases, uses artificial intelligence to summarize all management processes included in the project to generate a process summary table. All management processes in the process summary table are decomposed until the minimum process is obtained. The work location, working time period, and number of staff members allowed to access each minimum process are marked to form a minimum access block. The number of staff members allowed to access each minimum access block is counted.

[0014] The database contains the staff member's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice of the staff member reading words;

[0015] The artificial intelligence performs a time stamp on the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, obtains the staff identity information by comparing the second terminal information with the database, and identifies the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if so, it determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value. If so, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp and the voice in the database through artificial intelligence. The voice information is sorted after semantic recognition; if the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time stamp and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and it is determined whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, the voice information is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, the voice information sent for the second time is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to push it to the terminal.

[0016] The beneficial effects of the present invention are: by summarizing all management processes to generate a process summary table and breaking it down into minimum processes and marking each minimum process with the work location, work time period and a number of staff members allowed to access, that is, switching each minimum process into an independent calling process; in order to cooperate with the increasing number of calling processes, i.e., minimum processes, a marking method is used to improve the screening speed. Since the existing engineering project management is managed in a pyramid manner, the higher the level, the greater the calling authority and the more minimum processes that can be called, and the lower the level, the lower the calling authority and the fewer minimum processes that can be called, the staff members are graded according to the number of minimum processes that can be accessed. For example, the minimum process that an excavator supervisor can call may be only one. For such staff members with a small number of minimum processes, i.e., a number less than or equal to the first preset value, who are also the largest number of staff members, their identity information can be directly recognized by voice. Since the number of minimum processes required is small, it can be directly predicted that the minimum processes they can access are given. The process can greatly improve management efficiency and response speed, and since it only performs identity recognition actions, it can greatly reduce the processing content of the processing end, leaving sufficient computing power for other voice recognition; combined with the bound first terminal information, positioning information, time stamp, and the voice information sent by the terminal for the first time, since most middle-level staff will perform engineering project management operations at the work location and working hours in actual work, the work location and working time period of the minimum process block are combined as screening conditions to filter and push the minimum process called by the personnel to improve efficiency; the greater the authority, the more difficult it is to predict the number of minimum accessible processes, and the required minimum process is determined through semantic recognition; this application can greatly improve recognition efficiency by predicting the need for calling after identifying the source of the staff who sent the voice information for the first time; by using the pre-recorded voice of the staff reading words as the basis for artificial intelligence recognition, it greatly overcomes various accent problems and improves voice recognition efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a flowchart of a speech recognition method for engineering project management based on artificial intelligence according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0018] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0019] Please refer to Figure 1 , an artificial intelligence-based speech recognition method for engineering project management and control, comprising:

[0020] The staff sends the voice message and location information to the processing end for the first time through the bound terminal;

[0021] Through artificial intelligence, all management processes included in the project are summarized to generate a process summary table. All management processes in the process summary table are decomposed until the minimum process is obtained. The work location, working time period and number of staff members allowed to access each minimum process are marked to form a minimum access block. The number of staff members allowed to access all minimum access blocks is counted.

[0022] Entering the staff member's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice recordings of the staff member reading words into the database;

[0023] Artificial intelligence timestamps the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, and obtains the staff identity information by comparing the second terminal information with the database. Artificial intelligence recognizes the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if so, it determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value. If so, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp and the voice in the database through artificial intelligence. The voice information is sorted after semantic recognition; if the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time stamp and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and it is determined whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, the voice information is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, the voice information sent for the second time is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to push it to the terminal.

[0024] From the above description, it can be seen that by summarizing all management processes to generate a process summary table and breaking it down into minimum processes and marking each minimum process with the work location, work time period and a number of staff members allowed to access, each minimum process is switched into an independent calling process; in order to cope with the increasing number of calling processes, i.e., minimum processes, the speed of screening is improved by marking. Since the existing engineering project management is managed in a pyramid manner, the higher the level, the greater the calling authority and the more minimum processes that can be called; the lower the level, the lower the calling authority and the fewer minimum processes that can be called. Staff members are graded according to the number of processes they can access. For example, there may be only one minimum process that an excavator supervisor can call. For such staff members with a small number of minimum processes, i.e., a number less than or equal to the first preset value, who are also the largest number of staff members, their identity information can be directly recognized by voice. Since the number of minimum processes required is small, it is possible to directly predict the minimum process they can access. It can greatly improve management efficiency and response speed, and because it only performs identity recognition actions, it can greatly reduce the processing content of the processing end, leaving enough computing power for other voice recognition; combined with the bound first terminal information, positioning information, time stamp, and the voice information sent by the terminal for the first time, since most middle-level staff will perform engineering project management operations at the work location and working hours in actual work, the work location and working time period of the minimum process block are combined as screening conditions to filter and push the minimum process called by the personnel to improve efficiency; the greater the authority, the more difficult it is to predict the number of minimum accessible processes, and the required minimum process is determined through semantic recognition; this application can greatly improve the recognition efficiency by predicting the need for calling after identifying the source of the staff who sent the voice information for the first time; by using the pre-recorded voice of the staff reading words as the basis for artificial intelligence recognition, it greatly overcomes the problems of various accents and improves the efficiency of voice recognition.

[0025] Furthermore, when the number of times the staff allows access is greater than the first preset value and less than the second preset value, the staff sends voice information and positioning information to the processing end for the second time through the bound terminal. The artificial intelligence timestamps the voice information sent for the second time, and performs semantic recognition on the voice information sent for the second time based on the voice in the database, and then calls the minimum process with the strongest semantic relevance to push it to the terminal.

[0026] From the above description, it can be seen that the voice information and positioning information sent for the second time indicate that the first push is inaccurate, so the push is performed based on semantics to ensure the accuracy of the push.

[0027] Furthermore, if no third voice message is sent within the preset time interval after the second voice message is sent, the push is deemed successful. The artificial intelligence records the minimum process of the second push and records the voice message, positioning information, and time stamp sent for the second time, and marks the minimum process as the priority push process. If the staff member accesses the processing end through the terminal again within the preset range of the same time stamp and positioning information, the artificial intelligence directly pushes the priority push process.

[0028] From the above description, it can be seen that by recording the visiting habits of the staff, the location range and time period are locked for push and feedback correction.

[0029] Furthermore, the time interval between the first sending of the voice message and the second sending of the voice message is within a third preset value range. If it exceeds the third preset value, the time interval is reset to the first sending of the voice message.

[0030] From the above description, it can be seen that due to normal usage habits of people, if the pushed information is incorrect, the voice message will be sent again immediately. If the voice message is not sent again within the third preset value, it is considered that the push is successful.

[0031] Furthermore, when the artificial intelligence recognizes the voice information, it selects the clearest voice to be recognized and compares the recognition result with the first voiceprint information to obtain the second voiceprint information.

[0032] An artificial intelligence-based engineering project management and control speech recognition system, comprising:

[0033] Terminal: The staff sends voice information and positioning information to the processing end for the first time through the bound terminal;

[0034] The processing end, including artificial intelligence and databases, uses artificial intelligence to summarize all management processes included in the project to generate a process summary table. All management processes in the process summary table are decomposed until the minimum process is obtained. The work location, working time period, and number of staff members allowed to access each minimum process are marked to form a minimum access block. The number of staff members allowed to access each minimum access block is counted.

[0035] The database contains the staff member's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice of the staff member reading words;

[0036] The artificial intelligence performs a time stamp on the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, obtains the staff identity information by comparing the second terminal information with the database, and identifies the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if so, it determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value. If so, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp and the voice in the database through artificial intelligence. The voice information is sorted after semantic recognition; if the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time stamp and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and it is determined whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, the voice information is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, the voice information sent for the second time is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to push it to the terminal.

[0037] From the above description, it can be seen that by summarizing all management processes to generate a process summary table and breaking it down into minimum processes and marking each minimum process with the work location, work time period and a number of staff members allowed to access, each minimum process is switched into an independent calling process; in order to cope with the increasing number of calling processes, i.e., minimum processes, the speed of screening is improved by marking. Since the existing engineering project management is managed in a pyramid manner, the higher the level, the greater the calling authority and the more minimum processes that can be called; the lower the level, the lower the calling authority and the fewer minimum processes that can be called. Staff members are graded according to the number of processes they can access. For example, there may be only one minimum process that an excavator supervisor can call. For such staff members with a small number of minimum processes, i.e., a number less than or equal to the first preset value, who are also the largest number of staff members, their identity information can be directly recognized by voice. Since the number of minimum processes required is small, it is possible to directly predict the minimum process they can access. It can greatly improve management efficiency and response speed, and because it only performs identity recognition actions, it can greatly reduce the processing content of the processing end, leaving enough computing power for other voice recognition; combined with the bound first terminal information, positioning information, time stamp, and the voice information sent by the terminal for the first time, since most middle-level staff will perform engineering project management operations at the work location and working hours in actual work, the work location and working time period of the minimum process block are combined as screening conditions to filter and push the minimum process called by the personnel to improve efficiency; the greater the authority, the more difficult it is to predict the number of minimum accessible processes, and the required minimum process is determined through semantic recognition; this application can greatly improve the recognition efficiency by predicting the need for calling after identifying the source of the staff who sent the voice information for the first time; by using the pre-recorded voice of the staff reading words as the basis for artificial intelligence recognition, it greatly overcomes the problems of various accents and improves the efficiency of voice recognition.

[0038] Furthermore, when the number of times the staff allows access is greater than the first preset value and less than the second preset value, the staff sends voice information and positioning information to the processing end for the second time through the bound terminal. The artificial intelligence timestamps the voice information sent for the second time, and performs semantic recognition on the voice information sent for the second time based on the voice in the database, and then calls the minimum process with the strongest semantic relevance to push it to the terminal.

[0039] From the above description, it can be seen that the voice information and positioning information sent for the second time indicate that the first push is inaccurate, so the push is performed based on semantics to ensure the accuracy of the push.

[0040] Furthermore, if no third voice message is sent within the preset time interval after the second voice message is sent, the push is deemed successful. The artificial intelligence records the minimum process of the second push and records the voice message, positioning information, and time stamp sent for the second time, and marks the minimum process as the priority push process. If the staff member accesses the processing end through the terminal again within the preset range of the same time stamp and positioning information, the artificial intelligence directly pushes the priority push process.

[0041] From the above description, it can be seen that by recording the visiting habits of the staff, the location range and time period are locked for push and feedback correction.

[0042] Furthermore, the time interval between the first sending of the voice message and the second sending of the voice message is within a third preset value range. If it exceeds the third preset value, the time interval is reset to the first sending of the voice message.

[0043] From the above description, it can be seen that due to normal usage habits of people, if the pushed information is incorrect, the voice message will be sent again immediately. If the voice message is not sent again within the third preset value, it is considered that the push is successful.

[0044] Furthermore, when the artificial intelligence recognizes the voice information, it selects the clearest voice to be recognized and compares the recognition result with the first voiceprint information to obtain the second voiceprint information. Example 1

[0045] An artificial intelligence-based speech recognition method for engineering project management and control, comprising:

[0046] The staff sends the voice message and location information to the processing end for the first time through the bound terminal;

[0047] Through artificial intelligence, all management processes included in the project are summarized to generate a process summary table. All management processes in the process summary table are decomposed until the minimum process is obtained. The work location, working time period and number of staff members allowed to access each minimum process are marked to form a minimum access block. The number of staff members allowed to access all minimum access blocks is counted.

[0048] Entering the staff member's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice recordings of the staff member reading words into the database;

[0049] Artificial intelligence timestamps the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, and obtains the staff identity information by comparing the second terminal information with the database. Artificial intelligence recognizes the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if so, it determines whether the number of times the staff member is allowed to access is less than or equal to the first preset value (3 (or times)). If so, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp and semantic recognition of the voice information based on the voice in the database by artificial intelligence. If the number of staff members is greater than the first preset value (3 (numbers or times)) and less than the second preset value (10 (numbers or times)), the minimum process that meets the requirements is screened by comparing the time stamp and location information with the working hours and working locations, and judging whether the number of minimum processes that meet the requirements is less than or equal to the first preset value (3 (numbers or times)). If so, the minimum process that meets the requirements is called and sent to the terminal. If not, the voice information is semantically recognized based on the voice in the database by artificial intelligence, and the minimum process with the strongest semantic relevance is called to sort and push to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value (10 (numbers or times)), the voice information sent for the second time is semantically recognized based on the voice in the database by artificial intelligence, and the minimum process with the strongest semantic relevance is called to push to the terminal.

[0050] When the number of times the staff member allows access is greater than the first preset value (3 (times)) and less than the second preset value (10 (times)), the staff member sends the voice information and positioning information to the processing end for the second time through the bound terminal. The artificial intelligence timestamps the second sent voice information, and performs semantic recognition on the second sent voice information based on the voice in the database, and then calls the minimum process with the strongest semantic relevance to push it to the terminal.

[0051] If no third voice message is sent within the preset time after the second voice message is sent, the push is deemed successful. The artificial intelligence records the minimum process of the second push and records the voice message, positioning information, and time stamp sent for the second time, and marks the minimum process as the priority push process. If the staff member accesses the processing end through the terminal again within the preset range of the same time stamp and positioning information, the artificial intelligence directly pushes the priority push process.

[0052] The time interval between the first voice message sending and the second voice message sending is within a third preset value range. If it exceeds the third preset value, the time interval is reset to the first voice message sending.

[0053] When artificial intelligence recognizes voice information, it selects the clearest voice to be recognized and compares the recognition result with the first voiceprint information to obtain the second voiceprint information. Example 2

[0054] An artificial intelligence-based engineering project management and control speech recognition system, comprising:

[0055] Terminal: The staff sends voice information and positioning information to the processing end for the first time through the bound terminal;

[0056] The processing end, including artificial intelligence and databases, uses artificial intelligence to summarize all management processes included in the project to generate a process summary table. All management processes in the process summary table are decomposed until the minimum process is obtained. The work location, working time period, and number of staff members allowed to access each minimum process are marked to form a minimum access block. The number of staff members allowed to access each minimum access block is counted.

[0057] The database contains the staff member's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice of the staff member reading words;

[0058] The artificial intelligence performs a time stamp on the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, obtains the staff identity information by comparing the second terminal information with the database, and identifies the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if so, it determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value. If so, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp and the voice in the database through artificial intelligence. The voice information is sorted after semantic recognition; if the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time stamp and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and it is determined whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, the voice information is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, the voice information sent for the second time is semantically recognized based on the voice in the database through artificial intelligence, and the minimum process with the strongest semantic relevance is called to push it to the terminal.

[0059] in:

[0060] When the number of times the staff allows access is greater than the first preset value and less than the second preset value, the staff sends voice information and positioning information to the processing end for the second time through the bound terminal. The artificial intelligence timestamps the second sent voice information, and performs semantic recognition on the second sent voice information based on the voice in the database, and then calls the minimum process with the strongest semantic relevance to push it to the terminal.

[0061] If no third voice message is sent within the preset time after the second voice message is sent, the push is deemed successful. The artificial intelligence records the minimum process of the second push and records the voice message, positioning information, and time stamp sent for the second time, and marks the minimum process as the priority push process. If the staff member accesses the processing end through the terminal again within the preset range of the same time stamp and positioning information, the artificial intelligence directly pushes the priority push process.

[0062] The time interval between the first voice message sending and the second voice message sending is within a third preset value range. If it exceeds the third preset value, the time interval is reset to the first voice message sending.

[0063] When artificial intelligence recognizes voice information, it selects the clearest voice to be recognized and compares the recognition result with the first voiceprint information to obtain the second voiceprint information.

[0064] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A speech recognition method for engineering project management based on artificial intelligence, characterized in that: include: The staff sends voice information and location information to the processing end for the first time through the bound terminal; Through artificial intelligence, all management processes included in the project are summarized to generate a process summary table, and all management processes in the process summary table are decomposed until the minimum process cannot be decomposed. The work location, working time period and several staff members allowed to access each minimum process are marked to form a minimum access block; the number of staff members allowed to access all minimum access blocks is counted; Enter the staff's identity information, first voiceprint information, bound first terminal information, and pre-recorded voice of the staff reading words into the database; The artificial intelligence performs a time stamp on the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, obtains the staff identity information by comparing the second terminal information with the database, and identifies the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information. If not, a rejection instruction is sent to the terminal; if yes, it determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value. If yes, all the minimum processes allowed to be accessed by the staff member are called and sent to the terminal. If the minimum process is 1, no sorting is required. If it is greater than 1, sorting is performed. The sorting order is based on the positioning information, time stamp, and the artificial intelligence sorting the voice information after semantic recognition based on the voice in the database; If the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time mark and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and determine whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, artificial intelligence is used to perform semantic recognition on the voice information based on the voice in the database, and then the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, artificial intelligence is used to perform semantic recognition on the voice information sent for the second time based on the voice in the database, and then the minimum process with the strongest semantic relevance is called to push it to the terminal.

2. The method for speech recognition of engineering project management based on artificial intelligence according to claim 1 is characterized in that: When the number of times the staff allows access is greater than the first preset value and less than the second preset value, the staff sends voice information and positioning information to the processing end for the second time through the bound terminal. The artificial intelligence timestamps the voice information sent for the second time, and performs semantic recognition on the voice information sent for the second time based on the voice in the database, and then calls the minimum process with the strongest semantic relevance to push it to the terminal.

3. The method for speech recognition of engineering project management and control based on artificial intelligence according to claim 2 is characterized in that: If no third voice message is sent within the preset time after the second voice message is sent, the push is deemed successful. The artificial intelligence records the minimum process of the second push and records the voice message, positioning information, and time stamp sent for the second time, and marks the minimum process as the priority push process. If the staff member accesses the processing end through the terminal again within the preset range of the same time stamp and positioning information, the artificial intelligence directly pushes the priority push process.

4. The method for speech recognition of engineering project management and control based on artificial intelligence according to claim 3 is characterized in that: The time interval between sending the voice message for the first time and sending the voice message for the second time is within a third preset value range. If it exceeds the third preset value, the time interval is reset to the first time sending the voice message.

5. The method for speech recognition of engineering project management based on artificial intelligence according to claim 4 is characterized in that: When artificial intelligence recognizes voice information, it selects the clearest voice for recognition and compares the recognition result with the first voiceprint information to obtain the second voiceprint information.

6. An engineering project management and control speech recognition system based on artificial intelligence, characterized in that: include: Terminal: The staff sends voice information and positioning information to the processing end for the first time through the bound terminal; The processing end includes artificial intelligence and databases. Through artificial intelligence, all management processes included in the project are summarized to generate a process summary table, and all management processes in the process summary table are decomposed until the minimum process cannot be decomposed. The work location, working time period and several staff members allowed to access each minimum process are marked to form a minimum access block; the number of staff members allowed to access all minimum access blocks is counted; The database contains the staff's identity information, first voiceprint information, bound first terminal information, and pre-recorded speech of the staff reading words; The artificial intelligence performs a time stamp on the voice information sent by the terminal, identifies the source of the terminal to obtain the second terminal information, obtains the staff identity information by comparing the second terminal information with the database, and the artificial intelligence identifies the voice information and compares it with the first voiceprint information to obtain the second voiceprint information, and determines whether the second voiceprint information is consistent with the first terminal information contained in the staff information, and if not, sends a rejection instruction to the terminal; if so, determines whether the number of times the staff member is allowed to access is less than or equal to a first preset value, and if so, calls all the minimum processes allowed to be accessed by the staff member and sends them to the terminal, if the minimum process is 1, no sorting is required, and if it is greater than 1, sorting is performed, and the sorting order is based on the positioning information, the time stamp, and the artificial intelligence sorting the voice information after semantic recognition based on the voice in the database; If the number of times the staff member appears is greater than the first preset value and less than the second preset value, the time mark and location information are compared with the working hours and work locations to screen the minimum process that meets the requirements, and determine whether the number of minimum processes that meet the requirements is less than or equal to the first preset value. If so, the minimum process that meets the requirements is called and sent to the terminal. Otherwise, artificial intelligence is used to perform semantic recognition on the voice information based on the voice in the database, and then the minimum process with the strongest semantic relevance is called to sort and push it to the terminal; if the number of times the staff member appears is greater than or equal to the second preset value, artificial intelligence is used to perform semantic recognition on the voice information sent for the second time based on the voice in the database, and then the minimum process with the strongest semantic relevance is called to push it to the terminal.

7. The artificial intelligence-based engineering project management and control speech recognition system according to claim 6 is characterized in that: When the number of times the staff allows access is greater than the first preset value and less than the second preset value, the staff sends voice information and positioning information to the processing end for the second time through the bound terminal. The artificial intelligence timestamps the voice information sent for the second time, and performs semantic recognition on the voice information sent for the second time based on the voice in the database, and then calls the minimum process with the strongest semantic relevance to push it to the terminal.

8. The artificial intelligence-based engineering project management and control speech recognition system according to claim 7 is characterized in that: If no third voice message is sent within the preset time after the second voice message is sent, the push is deemed successful. The artificial intelligence records the minimum process of the second push and records the voice message, positioning information, and time stamp sent for the second time, and marks the minimum process as the priority push process. If the staff member accesses the processing end through the terminal again within the preset range of the same time stamp and positioning information, the artificial intelligence directly pushes the priority push process.

9. The artificial intelligence-based engineering project management and control speech recognition system according to claim 7 is characterized in that: The time interval between sending the voice message for the first time and sending the voice message for the second time is within a third preset value range. If it exceeds the third preset value, the time interval is reset to the first time sending the voice message.

10. The artificial intelligence-based engineering project management and control speech recognition system according to claim 9, characterized in that: When artificial intelligence recognizes voice information, it selects the clearest voice for recognition and compares the recognition result with the first voiceprint information to obtain the second voiceprint information.

Citation Information

Patent Citations

  • Big data based construction management method of engineering project with standard technology and storage medium

    CN108764610A

  • Directional personnel communication method and system, storage medium and intelligent voice equipment

    CN111161742A