Paint laboratory detection data measurement method and system based on voice recognition

Abstract

The invention relates to the technical field of data processing, in particular to a coating laboratory detection data measurement method and system based on voice recognition, and the method comprises the steps: obtaining laboratory voice data, inputting the laboratory voice data into a preset language recognition model, and obtaining a voice recognition result; obtaining text data; performing text cleaning processing on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; obtaining a corresponding specific experiment value according to the text entity feature, and obtaining a coating experiment result based on the specific experiment value; and inputting the text entity characteristics, the specific experiment numerical value and the coating experiment result into a preset experiment report template to obtain a coating experiment record report. The device has the effects of facilitating real-time recording of the paint experiment data and improving the integrity and accuracy of the paint experiment data.

Description

Technical Field

[0001] The present invention relates to the technical field of experimental data measurement and transmission, and in particular to a coating laboratory test data measurement method and system based on speech recognition. Background Art

[0002] A paint laboratory is a laboratory used for paint research and development, testing and analysis. In the laboratory, experimenters will conduct multiple tests on the paint and record various test data. Currently, most paint laboratories still use traditional paper recording methods to record data. This method is easily contaminated during operation due to the paint's easy staining, splashing, odor and other characteristics. The tester's hands are easily contaminated, and it is impossible to record the original paper records conveniently and quickly, resulting in low efficiency of recording while testing, and the inability to guarantee the timeliness of recording, and even missing key test nodes, low recording efficiency, and poor timeliness.

[0003] Existing coating experimental data measurement systems usually use manual recording methods, and researchers will manually record the data obtained during the test into paper or electronic documents. In addition, some laboratories have begun to try to use image recognition technology to automatically identify experimental equipment readings, and to assist in recording through simple automated equipment such as barcode scanners. However, these solutions mainly rely on visual input and require more manual intervention.

[0004] At present, although some coating experimental data measurement systems have emerged, they are usually complicated to operate and cannot be well integrated with laboratory workflows. Experimenters are still required to input data through a keyboard or touch screen and manually input a large amount of data, which is not convenient in a laboratory environment, especially when the hands of the inspectors may be contaminated by the coating. The operation is cumbersome and prone to errors, resulting in low detection efficiency and the inability to guarantee the timeliness of data recording. Therefore, there is room for improvement. Summary of the invention

[0005] In order to facilitate real-time recording of coating experimental data and improve the integrity and accuracy of coating experimental data, the present application provides a coating laboratory test data measurement method and system based on speech recognition.

[0006] The above-mentioned invention objective of the present application is achieved through the following technical solutions: A method for measuring coating laboratory test data based on speech recognition, the method comprising the steps of: Acquire laboratory speech data, and input the laboratory speech data into a preset language recognition model to obtain text data; Performing text cleaning on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; Acquire corresponding specific experimental values ​​according to the text entity features, and acquire coating experimental results based on the specific experimental values; The text entity features, specific experimental values ​​and coating experiment results are input into a preset experimental report template to obtain a coating experiment record report.

[0007] By adopting the above technical solution, in the coating laboratory, when the experimenter needs to record the coating experiment data, the experimenter can directly verbally explain the specific process of the coating experiment in the laboratory, and can obtain the laboratory voice data through the sound collection equipment in the laboratory and the speech recognition technology. The laboratory voice data refers to the voice data of the experimenter explaining the coating experiment process in the laboratory. The experimental voice data is input into a preset language recognition model, and the language recognition model is used to transcribe the voice into text to form text data. The obtained text data is cleaned and processed to remove irrelevant characters, punctuation marks and stop words to improve the quality of the text data, thereby improving the accuracy of the experimental data recording results. The text data is subjected to entity feature recognition to obtain text entity features. Text entity features refer to entities in the text data, such as experimental parameters. The name, unit and other information are obtained according to the text entity features, and the corresponding specific experimental values ​​are obtained. The specific experimental values ​​refer to the data corresponding to the text entity features. For example, the reaction temperature is set to 100 degrees Celsius, the reaction temperature and Celsius are the text entity features, and 100 is the corresponding specific experimental value. The specific experimental values ​​are used for analysis to obtain the coating experiment results, thereby realizing the analysis function of the coating experiment. The text entity features, specific experimental values ​​and coating experiment results are input into the preset experimental report template to form a coating experiment record report, complete the detection and recording of the coating experiment data, and form a coating experiment report to facilitate the experimenter to analyze the coating experiment process. The entire coating experiment data recording process is obtained through voice recognition, and there is no need for the experimenter to manually record the data, which is convenient for real-time recording of coating experiment data and improves the integrity and accuracy of coating experiment data.

[0008] In a preferred example, the present application may be further configured as follows: before inputting the laboratory voice data into a preset language recognition model, the coating laboratory test data measurement method based on voice recognition further includes: Performing adaptive data processing on the laboratory voice data to obtain laboratory voice data from which environmental noise outside a pickup range is filtered out; The laboratory speech data after the environmental noise is filtered out is subjected to time-frequency blind separation processing to filter out the residual noise in the sound pickup range to obtain the target laboratory speech data.

[0009] By adopting the above technical scheme, the laboratory voice data picked up by the sound collection device is adaptively processed, the environmental noise beyond the sound pickup range is preliminarily filtered out, and the laboratory voice data with the environmental noise outside the pickup range filtered out is obtained, which helps to reduce the impact of external environmental noise. Combined with the time-frequency analysis algorithm, the laboratory voice data within the sound pickup range is subjected to time-frequency blind separation processing, and the residual noise in the laboratory voice data is further filtered out again, which helps to further improve the quality of the laboratory voice data.

[0010] In a preferred example, the present application may be further configured as follows: the adaptive data processing of the laboratory voice data to obtain the laboratory voice data from which the ambient noise outside the pickup range is filtered out specifically includes: Performing fitting processing on the laboratory speech data to obtain sound source position information of the laboratory speech data; Acquire the coupling relationship between the sound source position information and the sound pickup range of the sound pickup device, and construct a noise reduction optimization criterion that conforms to the sound source position information; According to the noise reduction optimization criterion, the laboratory speech data is filtered to obtain the laboratory speech data from which the environmental noise outside the pickup range is filtered out.

[0011] By adopting the above technical solution, the laboratory voice data is fitted and the position with the highest fitting degree is taken as the sound source position, that is, the position where the experimenter speaks. The sound source position information is coupled with the sound pickup range of the sound pickup device to find the coupling relationship between the sound source position and the sound pickup range of the sound pickup device, which helps to construct a noise reduction optimization criterion that conforms to the sound source position, and helps to perform targeted filtering on the laboratory voice data picked up by the sound pickup device according to the noise reduction optimization criterion, so as to obtain the laboratory voice data with the ambient noise outside the pickup range filtered out, thereby improving the noise reduction accuracy of the laboratory voice data and thus improving the quality of experimental data recording.

[0012] In a preferred example, the present application may be further configured as follows: performing text cleaning on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data, specifically including: The text cleaning process includes an abnormality removal process and a repair process, and abnormal text features are obtained based on the text data; Acquire the text to be deleted and the text to be repaired according to the abnormal text features, perform abnormal removal processing on the text data based on the text to be deleted, and perform repair processing on the text data based on the text to be repaired, so as to obtain cleaned text data; The cleaned text data is input into a preset named entity recognition model to obtain text entity features.

[0013] By adopting the above technical scheme, in the process of extracting text data to obtain corresponding experimental data, abnormal text features are extracted according to the converted text data. Abnormal text features refer to abnormal text content and redundant content identified in the process of converting into text content. The corresponding text to be deleted and text to be repaired are determined according to the abnormal text features, and the text to be deleted is subjected to abnormal removal processing, and the text to be repaired is subjected to repair processing to form cleaned text data, thereby improving the quality of the text data. The cleaned text data is input into a preset named entity recognition model, and the NER technology in the named entity model is used to extract text entity features from the cleaned text data, so as to quickly extract experimental data from the text data.

[0014] In a preferred example, the present application may be further configured as follows: the step of obtaining corresponding specific experimental values ​​according to the text entity features and obtaining coating experimental results based on the specific experimental values ​​specifically includes: Based on a preset regular expression formula, a specific experimental value is obtained in the text data according to the text entity feature; The specific experimental values ​​are compared with the preset experimental standard values ​​to obtain the coating experimental results.

[0015] By adopting the above technical solution, in the process of acquiring experimental data, a regular expression formula is set up in the system, and the acquired text entity features are input into the regular expression formula to quickly extract specific experimental values ​​from the text data. This can help us accurately extract specific values ​​related to parameter entities from complex text data, greatly simplifying the data extraction and analysis process. The extracted specific experimental values ​​are compared with the preset experimental standard values, the coating experimental results have been analyzed, and the coating experimental data recording process is completed.

[0016] In a preferred example, the present application may be further configured as follows: before inputting the text entity features, specific experimental values ​​and coating experiment results into a preset experimental report template to obtain a coating experiment record report, the coating laboratory test data measurement method based on speech recognition also includes: Acquire coating experiment environment information, and acquire experiment environment pictures and experiment environment data based on the coating experiment environment information; The experimental environment picture and the experimental environment data are integrated to form a coating experimental environment analysis result, and input into a preset experimental report template.

[0017] By adopting the above technical solution, in the process of generating a paint experiment report, the paint experiment environment information can be obtained through the image acquisition equipment in the laboratory and the environmental monitoring equipment in the laboratory, and the experimental environment picture and experimental environment data can be extracted from the paint experiment environment information. The experimental environment picture refers to the picture information of the paint experiment process, and the experimental environment data refers to the laboratory environment conditions during the paint experiment process, such as laboratory temperature, humidity and other data. The experimental environment picture and the experimental environment data are integrated to form an experimental environment analysis result, and are input into the experimental report template, so that the formed paint experiment report can record the real-time experimental process pictures and the laboratory environment conditions, forming a more comprehensive paint experiment report.

[0018] The second object of the invention is achieved by the following technical solutions: A coating laboratory test data measurement system based on speech recognition, the coating laboratory test data measurement system based on speech recognition comprising: A text data acquisition module is used to acquire laboratory voice data, and input the laboratory voice data into a preset language recognition model to obtain text data; A text data analysis module, used for performing text cleaning processing on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; An experimental data extraction module, used to obtain corresponding specific experimental values ​​according to the text entity features, and obtain coating experimental results based on the specific experimental values; The experimental report generation module is used to input the text entity features, specific experimental values ​​and coating experimental results into a preset experimental report template to obtain a coating experiment record report.

[0019] By adopting the above technical solution, in the coating laboratory, when the experimenter needs to record the coating experiment data, the experimenter can directly verbally explain the specific process of the coating experiment in the laboratory, and can obtain the laboratory voice data through the sound collection equipment in the laboratory and the speech recognition technology. The laboratory voice data refers to the voice data of the experimenter explaining the coating experiment process in the laboratory. The experimental voice data is input into a preset language recognition model, and the language recognition model is used to transcribe the voice into text to form text data. The obtained text data is cleaned and processed to remove irrelevant characters, punctuation marks and stop words to improve the quality of the text data, thereby improving the accuracy of the experimental data recording results. The text data is subjected to entity feature recognition to obtain text entity features. Text entity features refer to entities in the text data, such as experimental parameters. The name, unit and other information are obtained according to the text entity features, and the corresponding specific experimental values ​​are obtained. The specific experimental values ​​refer to the data corresponding to the text entity features. For example, the reaction temperature is set to 100 degrees Celsius, the reaction temperature and Celsius are the text entity features, and 100 is the corresponding specific experimental value. The specific experimental values ​​are used for analysis to obtain the coating experiment results, thereby realizing the analysis function of the coating experiment. The text entity features, specific experimental values ​​and coating experiment results are input into the preset experimental report template to form a coating experiment record report, complete the detection and recording of the coating experiment data, and form a coating experiment report to facilitate the experimenter to analyze the coating experiment process. The entire coating experiment data recording process is obtained through voice recognition, and there is no need for the experimenter to manually record the data, which is convenient for real-time recording of coating experiment data and improves the integrity and accuracy of coating experiment data.

[0020] The third objective of the present application is achieved through the following technical solutions: A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for measuring coating laboratory test data based on speech recognition are implemented.

[0021] The fourth objective of the present application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for measuring coating laboratory test data based on speech recognition.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: 1. In the coating laboratory, when the experimenter needs to record the coating experiment data, the experimenter can directly verbally explain the specific process of the coating experiment in the laboratory. The laboratory voice data can be obtained through the sound collection equipment in the laboratory and the speech recognition technology. The laboratory voice data refers to the voice data of the experimenter explaining the coating experiment process in the laboratory. The experimental voice data is input into the preset language recognition model, and the language recognition model is used to transcribe the voice into text to form text data. The obtained text data is cleaned and processed to remove irrelevant characters, punctuation marks and stop words to improve the quality of the text data, thereby improving the accuracy of the experimental data recording results. The text data is subjected to entity feature recognition to obtain text entity features. Text entity features refer to entities in text data, such as experimental parameter names, units, etc. Information, according to the text entity features, the corresponding specific experimental values ​​are obtained. The specific experimental values ​​refer to the data corresponding to the text entity features. For example, if the reaction temperature is set to 100 degrees Celsius, the reaction temperature and degrees Celsius are the text entity features, and 100 is the corresponding specific experimental value. The specific experimental values ​​obtained are used for analysis to obtain the coating experiment results, and the analysis function of the coating experiment is realized. The text entity features, specific experimental values ​​and coating experiment results are input into the preset experimental report template to form a coating experiment record report, complete the detection and recording of the coating experiment data, and form a coating experiment report to facilitate the experimenter to analyze the coating experiment process. The entire coating experiment data recording process is obtained through voice recognition, and there is no need for the experimenter to manually record the data, which is convenient for real-time recording of coating experiment data and improves the integrity and accuracy of coating experiment data. 2. By adaptively processing the laboratory voice data picked up by the sound collection device, the environmental noise outside the sound pickup range is initially filtered out, and the laboratory voice data with the environmental noise outside the pickup range filtered out is obtained, which helps to reduce the marketing of external environmental noise. Combined with the time-frequency analysis algorithm, the laboratory voice data within the sound pickup range is subjected to time-frequency blind separation processing, and the residual noise in the laboratory voice data is further filtered out again, which helps to further improve the quality of the laboratory voice data; 3. In the process of extracting the text data to obtain the corresponding experimental data, the abnormal text features are extracted according to the converted text data. The abnormal text features refer to the abnormal text content and redundant content identified in the process of converting into text content. The corresponding text to be deleted and the text to be repaired are determined according to the abnormal text features. The text to be deleted is subjected to abnormal removal processing, and the text to be repaired is subjected to repair processing to form cleaned text data, thereby improving the quality of the text data. The cleaned text data is input into a preset named entity recognition model, and the NER technology in the named entity model is used to extract text entity features from the cleaned text data, so as to quickly extract experimental data from the text data; 4. In the process of acquiring experimental data, set up a regular expression formula in the system, input the acquired text entity features into the regular expression formula, and quickly extract specific experimental values ​​from the text data. This can help us accurately extract specific values ​​related to parameter entities from complex text data, greatly simplifying the data extraction and analysis process. The extracted specific experimental values ​​are compared with the preset experimental standard values, the coating experimental results have been analyzed, and the coating experimental data recording process is completed. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of a coating laboratory test data measurement method based on speech recognition in one embodiment of the present application; Figure 2 is another implementation flow chart of a coating laboratory test data measurement method based on speech recognition in an embodiment of the present application; Figure 3 It is a flowchart for implementing step S101 in the coating laboratory test data measurement method based on speech recognition in one embodiment of the present application; Figure 4 It is a flowchart for implementing step S20 in the coating laboratory test data measurement method based on speech recognition in one embodiment of the present application; Figure 5 It is a flowchart for implementing step S30 in the coating laboratory test data measurement method based on speech recognition in one embodiment of the present application; Figure 6 is another implementation flow chart of a coating laboratory test data measurement method based on speech recognition in an embodiment of the present application; Figure 7 It is a principle block diagram of a coating laboratory test data measurement system based on speech recognition in one embodiment of the present application; Figure 8 It is a schematic diagram of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0024] The present application is further described in detail below in conjunction with the accompanying drawings.

[0025] In one embodiment, if Figure 1 As shown, the present application discloses a method for measuring coating laboratory test data based on speech recognition, which specifically includes the following steps: S10: Acquire laboratory speech data, and input the laboratory speech data into a preset language recognition model to obtain text data.

[0026] In this embodiment, the laboratory voice data refers to the voice data of the experimenter explaining the coating experiment process in the laboratory.

[0027] Specifically, in a paint laboratory, when an experimenter needs to record paint experiment data, the experimenter can directly verbally explain the specific process of the paint experiment in the laboratory, and can use voice recognition technology through the sound collection equipment in the laboratory to obtain the voice data of the experimenter explaining the paint experiment process in the laboratory.

[0028] Furthermore, the experimental speech data is input into a preset language recognition model, and the speech is transcribed into text using the language recognition model to form text data.

[0029] S20: performing text cleaning processing on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data.

[0030] In this embodiment, the text entity feature refers to the entity in the text data, such as the experimental parameter name, unit and other information.

[0031] Specifically, the obtained text data is cleaned to remove irrelevant characters, punctuation marks and stop words to improve the quality of the text data, thereby improving the accuracy of the experimental data recording results, and the text data is subjected to entity feature recognition to obtain entities in the text data, such as experimental parameter names, units and other information.

[0032] S30: Obtain corresponding specific experimental values ​​according to the text entity features, and obtain coating experimental results based on the specific experimental values.

[0033] Specifically, the corresponding specific experimental values ​​are obtained according to the text entity features. The specific experimental values ​​refer to the data corresponding to the text entity features. For example, if the reaction temperature is set to 100 degrees Celsius, the reaction temperature and Celsius are the text entity features, and 100 is the corresponding specific experimental value.

[0034] S40: Input the text entity features, specific experimental values ​​and coating experiment results into a preset experiment report template to obtain a coating experiment record report.

[0035] Specifically, the specific experimental values ​​obtained are used for analysis to obtain the paint experiment results, thereby realizing the analysis function of the paint experiment. The text entity features, specific experimental values ​​and paint experiment results are input into a preset experimental report template to form a paint experiment record report, thereby completing the detection and recording of the paint experiment data.

[0036] In this embodiment, in the coating laboratory, when the experimenter needs to measure and record the coating experiment data, the experimenter can directly verbally explain the specific process of the coating experiment in the laboratory, and can obtain the laboratory voice data through the sound collection equipment in the laboratory and the speech recognition technology. The laboratory voice data refers to the voice data of the experimenter explaining the coating experiment process in the laboratory. The experimental voice data is input into a preset language recognition model, and the language recognition model is used to transcribe the voice into text to form text data. The obtained text data is cleaned to remove irrelevant characters, punctuation marks and stop words to improve the quality of the text data, thereby improving the accuracy of the experimental data recording results. The text data is subjected to entity feature recognition to obtain text entity features. Text entity features refer to entities in the text data, such as experimental parameter names, units and other information. , according to the text entity features, the corresponding specific experimental values ​​are obtained. The specific experimental values ​​refer to the data corresponding to the text entity features. For example, the reaction temperature is set to 100 degrees Celsius, the reaction temperature and Celsius are the text entity features, and 100 is the corresponding specific experimental value. The specific experimental values ​​obtained are used for analysis to obtain the coating experiment results, and the analysis function of the coating experiment is realized. The text entity features, specific experimental values ​​and coating experiment results are input into the preset experimental report template to form a coating experiment record report, complete the detection and recording of the coating experiment data, and form a coating experiment report to facilitate the experimenter to analyze the coating experiment process. The measurement record of the entire coating experiment data is obtained through the audio signal recognition of the voice signal acquisition system, and there is no need for the experimenter to manually record the data, which is convenient for real-time recording of coating experiment data and improves the integrity and accuracy of coating experiment data.

[0037] In one embodiment, if Figure 2 As shown, before step S10, that is, before inputting the laboratory voice data into a preset language recognition model, the coating laboratory test data measurement method based on voice recognition further includes: S101: performing adaptive data processing on the laboratory voice data to obtain the laboratory voice data from which environmental noise outside a pickup range is filtered out.

[0038] Specifically, the original laboratory voice data is converted and processed, and the original laboratory voice data is amplified through a power amplifier, and adaptive data processing is performed on the amplified original laboratory voice data. The laboratory voice data picked up by the sound collection device is adaptively processed, and the environmental noise beyond the sound pickup range is preliminarily filtered out to obtain laboratory voice data with the environmental noise outside the pickup range filtered out, which helps to reduce the impact of external environmental noise.

[0039] S22: Performing time-frequency blind separation processing on the laboratory speech data after filtering out the environmental noise, filtering out the residual noise in the sound pickup range, and obtaining the target laboratory speech data.

[0040] Specifically, combined with the time-frequency analysis algorithm, the laboratory speech data within the sound pickup range is subjected to time-frequency blind separation processing, and the residual noise in the laboratory speech data is further filtered out, which helps to further improve the quality of the laboratory speech data.

[0041] Furthermore, when each sound collection device receives the original laboratory speech data emitted by the sound source, the collection time of the sound collection device is recorded, and then the discrete time series of the sound source reaching the sound collection device can be obtained. According to the discrete time series, the impulse response signal corresponding to the discrete time series is marked. As the discrete value of the collection time is larger, the impulse response value is also longer, thereby obtaining a positive proportional relationship between the impulse response and the collection time difference between the sound collection device and the sound source. As the impulse response relationships of the sound collection devices are aggregated, the impulse response relationship of the sound source and the collection time of the sound collection device is obtained. The obtained impulse response relationship is used to calculate the distortion adjustment factor between the sound source and the sound collection device, and the distortion adjustment factor is used to adjust the distortion of the laboratory speech data, and the residual noise in the laboratory speech data is further filtered out again.

[0042] In one embodiment, if Figure 3 As shown, in step S101, adaptive data processing is performed on the laboratory voice data to obtain the laboratory voice data from which the ambient noise outside the pickup range is filtered out, which specifically includes: S1011: Perform fitting processing on the laboratory speech data to obtain sound source position information of the laboratory speech data.

[0043] Specifically, MATLAB can be used to calculate the sound source position information, that is, the position where the experimenter speaks, from the laboratory speech data through a fitting formula.

[0044] S1012: Acquire a coupling relationship between the sound source position information and a sound pickup range of a sound pickup device, and construct a noise reduction optimization criterion that conforms to the sound source position information.

[0045] Specifically, by coupling the sound source position information with the sound pickup range of the sound pickup device, the coupling relationship between the sound source position and the sound pickup range of the sound pickup device is found, which helps to construct a noise reduction optimization criterion that conforms to the sound source position, and helps to perform targeted filtering processing on the laboratory voice data picked up by the sound pickup device according to the noise reduction optimization criterion.

[0046] S1013: According to the noise reduction optimization criterion, the laboratory speech data is filtered to obtain the laboratory speech data from which the environmental noise outside the pickup range is filtered out.

[0047] Specifically, all original laboratory speech data are filtered according to the noise reduction optimization criterion, the original laboratory speech data are error compensated, and according to the triangular sound pickup range, the laboratory speech data outside the triangular sound pickup range is filtered out as an environmental noise signal, thereby obtaining only the laboratory speech data within the triangular sound pickup range as the laboratory speech data for filtering out the environmental noise outside the pickup range.

[0048] In one embodiment, if Figure 4 As shown, in step S20, the text data is cleaned to obtain cleaned text data, and text entity features are extracted based on the cleaned text data, specifically including: S21: The text cleaning process includes an abnormality removal process and a repair process, and abnormal text features are obtained based on the text data.

[0049] Specifically, in the process of extracting text data to obtain corresponding experimental data, abnormal text features are extracted based on the converted text data. The abnormal text features refer to abnormal text content and redundant content identified in the process of converting into text content.

[0050] S22: acquiring the text to be deleted and the text to be repaired according to the abnormal text features, performing an abnormality removal processing on the text data based on the text to be deleted, and performing a repair processing on the text data based on the text to be repaired, to obtain cleaned text data.

[0051] Specifically, the corresponding text to be deleted and the text to be repaired are determined according to the abnormal text features, the text to be deleted is processed for abnormal removal, and the text to be repaired is processed for repair to form cleaned text data, thereby improving the quality of the text data.

[0052] S23: Inputting the cleaned text data into a preset named entity recognition model to obtain text entity features.

[0053] Specifically, the cleaned text data is input into a preset named entity recognition model, and the NER technology in the named entity model is used to extract text entity features from the cleaned text data. The NER technology is used to label these entities in the text, such as "temperature", "pressure", "yield", etc., for subsequent information extraction.

[0054] In one embodiment, if Figure 5As shown, in step S30, the corresponding specific experimental values ​​are obtained according to the text entity features, and the coating experimental results are obtained based on the specific experimental values, which specifically includes: S31: Based on a preset regular expression formula, a specific experimental value is obtained in the text data according to the text entity feature.

[0055] Specifically, in the process of acquiring experimental data, regular expression formulas are set up in the system, such as "the temperature is (\d+) degrees", "the pressure is ([\d.]+)MPa", "the reaction lasted (\d+) hours", etc., where "\d+" means matching one or more numbers, and the brackets indicate the part we want to extract, such as "under the conditions of temperature of 80 degrees and pressure of 1.5MPa, the reaction lasted for 3 hours". "Temperature" is used as a text entity feature. Using regular expressions, the corresponding value "80" can be extracted according to the text entity feature, and specific experimental values ​​can be quickly extracted from the text data, which can help us accurately extract specific values ​​related to parameter entities from complex text data, greatly simplifying the process of data extraction and analysis.

[0056] S32: Compare the specific experimental value with the preset experimental standard value to obtain the coating experimental result.

[0057] Specifically, the extracted specific experimental values ​​are compared with the preset experimental standard values ​​to analyze the coating experimental results and complete the coating experimental data recording process.

[0058] In one embodiment, if Figure 6 As shown, before step S40, that is, before inputting the text entity features, specific experimental values ​​and coating experimental results into a preset experimental report template to obtain a coating experimental record report, the coating laboratory test data measurement method based on speech recognition also includes: S401: Acquire paint experiment environment information, and acquire an experiment environment picture and experiment environment data based on the paint experiment environment information.

[0059] S402: Integrate the experimental environment image and the experimental environment data to form a coating experimental environment analysis result, and input it into a preset experimental report template.

[0060] Specifically, in the process of generating a paint experiment report, the paint experiment environment information can be obtained through the image acquisition equipment in the laboratory and the environmental monitoring equipment in the laboratory, and the experimental environment picture and experimental environment data can be extracted from the paint experiment environment information. The experimental environment picture refers to the picture information of the paint experiment process, and the experimental environment data refers to the laboratory environment conditions during the paint experiment process, such as laboratory temperature, humidity and other data. The experimental environment picture and the experimental environment data are integrated to form an experimental environment analysis result, and are input into the experimental report template, so that the formed paint experiment report can record real-time experimental process pictures and laboratory environmental conditions, forming a more comprehensive paint experiment report.

[0061] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0062] In one embodiment, a coating laboratory test data measurement system based on speech recognition is provided, and the coating laboratory test data measurement system based on speech recognition corresponds to the coating laboratory test data measurement method based on speech recognition in the above embodiment. Figure 7 As shown, the coating laboratory test data measurement system based on speech recognition includes a text data acquisition module, a text data analysis module, an experimental data extraction module and an experimental report generation module. The detailed description of each functional module is as follows: A text data acquisition module is used to acquire laboratory voice data, and input the laboratory voice data into a preset language recognition model to obtain text data; A text data analysis module, used for performing text cleaning processing on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; An experimental data extraction module, used to obtain corresponding specific experimental values ​​according to the text entity features, and obtain coating experimental results based on the specific experimental values; The experimental report generation module is used to input the text entity features, specific experimental values ​​and coating experimental results into a preset experimental report template to obtain a coating experiment record report.

[0063] Optionally, the coating laboratory test data measurement system based on speech recognition also includes: The voice data processing module is used to perform adaptive data processing on the laboratory voice data to obtain laboratory voice data with environmental noise outside the pickup range filtered out, perform blind time-frequency separation processing on the laboratory voice data after the environmental noise is filtered out, filter out residual noise within the sound pickup range, and obtain target laboratory voice data.

[0064] For the specific definition of the coating laboratory test data measurement system based on speech recognition, please refer to the definition of the coating laboratory test data measurement method based on speech recognition above, which will not be repeated here. Each module in the above-mentioned coating laboratory test data measurement system based on speech recognition can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0065] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store laboratory voice data and convert it into text data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a coating laboratory test data measurement method based on speech recognition is implemented.

[0066] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Acquire laboratory speech data, and input the laboratory speech data into a preset language recognition model to obtain text data; Performing text cleaning on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; Acquire corresponding specific experimental values ​​according to the text entity features, and acquire coating experimental results based on the specific experimental values; The text entity features, specific experimental values ​​and coating experiment results are input into a preset experimental report template to obtain a coating experiment record report.

[0067] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Acquire laboratory speech data, and input the laboratory speech data into a preset language recognition model to obtain text data; Performing text cleaning on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; Acquire corresponding specific experimental values ​​according to the text entity features, and acquire coating experimental results based on the specific experimental values; The text entity features, specific experimental values ​​and coating experiment results are input into a preset experimental report template to obtain a coating experiment record report.

[0068] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0069] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0070] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for measuring coating laboratory test data based on speech recognition, characterized in that: The coating laboratory test data measurement method based on speech recognition comprises the following steps: Acquire laboratory speech data, and input the laboratory speech data into a preset language recognition model to obtain text data; Performing text cleaning on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; Acquire corresponding specific experimental values ​​according to the text entity features, and acquire coating experimental results based on the specific experimental values; The text entity features, specific experimental values ​​and coating experiment results are input into a preset experimental report template to obtain a coating experiment record report.

2. The method for measuring coating laboratory test data based on speech recognition according to claim 1 is characterized in that: Before inputting the laboratory voice data into a preset language recognition model, the coating laboratory test data measurement method based on voice recognition further includes: Performing adaptive data processing on the laboratory voice data to obtain laboratory voice data from which environmental noise outside a pickup range is filtered out; The laboratory speech data from which the environmental noise outside the pickup range is filtered out is subjected to blind time-frequency separation processing to filter out the residual noise within the sound pickup range, thereby obtaining the target laboratory speech data.

3. The method for measuring coating laboratory test data based on speech recognition according to claim 2 is characterized in that: The performing of adaptive data processing on the laboratory voice data to obtain the laboratory voice data from which the environmental noise outside the pickup range is filtered out specifically includes: Performing fitting processing on the laboratory speech data to obtain sound source position information of the laboratory speech data; Acquire the coupling relationship between the sound source position information and the sound pickup range of the sound pickup device, and construct a noise reduction optimization criterion that conforms to the sound source position information; According to the noise reduction optimization criterion, the laboratory speech data is filtered to obtain the laboratory speech data from which the environmental noise outside the pickup range is filtered out.

4. The method for measuring coating laboratory test data based on speech recognition according to claim 1, characterized in that: The performing text cleaning on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data specifically includes: The text cleaning process includes an abnormality removal process and a repair process, and abnormal text features are obtained based on the text data; Acquire the text to be deleted and the text to be repaired according to the abnormal text features, perform abnormal removal processing on the text data based on the text to be deleted, and perform repair processing on the text data based on the text to be repaired, so as to obtain cleaned text data; The cleaned text data is input into a preset named entity recognition model to obtain text entity features.

5. The method for measuring coating laboratory test data based on speech recognition according to claim 1, characterized in that: The step of obtaining a corresponding specific experimental value according to the text entity feature and obtaining a coating experimental result based on the specific experimental value specifically includes: Based on a preset regular expression formula, a specific experimental value is obtained in the text data according to the text entity feature; The specific experimental values ​​are compared with the preset experimental standard values ​​to obtain the coating experimental results.

6. The method for measuring coating laboratory test data based on speech recognition according to claim 1, characterized in that: Before inputting the text entity features, specific experimental values ​​and coating experiment results into a preset experimental report template to obtain a coating experiment record report, the coating laboratory test data measurement method based on speech recognition also includes: Acquire coating experiment environment information, and acquire experiment environment pictures and experiment environment data based on the coating experiment environment information; The experimental environment picture and the experimental environment data are integrated to form a coating experimental environment analysis result, and input into a preset experimental report template.

7. A coating laboratory test data measurement system based on speech recognition, characterized in that: The coating laboratory test data measurement system based on speech recognition includes: A text data acquisition module is used to acquire laboratory voice data, and input the laboratory voice data into a preset language recognition model to obtain text data; A text data analysis module, used for performing text cleaning processing on the text data to obtain cleaned text data, and extracting text entity features based on the cleaned text data; An experimental data extraction module, used to obtain corresponding specific experimental values ​​according to the text entity features, and obtain coating experimental results based on the specific experimental values; The experimental report generation module is used to input the text entity features, specific experimental values ​​and coating experimental results into a preset experimental report template to obtain a coating experiment record report.

8. The coating laboratory test data measurement system based on speech recognition according to claim 7 is characterized in that: The coating laboratory test data measurement system based on speech recognition also includes: The voice data processing module is used to perform adaptive data processing on the laboratory voice data to obtain laboratory voice data with environmental noise outside the pickup range filtered out, perform blind time-frequency separation processing on the laboratory voice data with environmental noise outside the pickup range filtered out, filter out residual noise within the sound pickup range, and obtain target laboratory voice data.

Images