Property management conference quality evaluation method, electronic equipment and storage medium

Abstract

The invention discloses a property management conference quality evaluation method, electronic equipment and a storage medium, and relates to the technical field of conference evaluation. The method comprises the following steps: identifying and evaluating conference voice data to obtain participant identity information and a conference participation rate score; identifying and evaluating the conference voice data, and obtaining a conference emotion score and an employee participation score based on first text data obtained by converting the conference voice data; performing identification and evaluation according to the first text data and the determined quality evaluation standard table to obtain a conference completion degree score; distributing weights according to weight priorities of the quality evaluation standard table to obtain a weight coefficient set; according to a score adjustment formula, the weight coefficient set, the conference emotion score, the conference completion degree score, the employee participation degree score and the conference participation rate score, performing calculation to obtain a conference quality score; and generating a conference quality report. The conference quality can be automatically evaluated, an objective and reliable conference quality evaluation report is formed, and the efficiency and reliability of conference quality evaluation are improved.

Description

Technical Field

[0001] The present application relates to the technical field of conference evaluation, and in particular to a property management conference quality evaluation method, electronic equipment, and storage medium. Background Art

[0002] Holding meetings in property management is an important way for team collaboration and information transfer. After the meeting, it is usually necessary to conduct a quality assessment of the meeting so that the meeting can be effectively managed and supervised, providing a reliable reference for subsequent improvements in meeting formats, enrichment of meeting content, and improvement of meeting processes.

[0003] There are several methods of meeting quality assessment generally used in current enterprise management. First, after the meeting, participants are required to fill out a questionnaire to evaluate the quality of the meeting, and the quality of the meeting is evaluated through the results of the collected questionnaires. Second, the meeting host constantly observes the reactions of the participants during the meeting and evaluates the quality of the meeting through feedback from the meeting site. Third, during the meeting, the recorder records the organization and specific content of the meeting to form a meeting record. Therefore, after the meeting, the meeting can be reviewed based on the meeting record and the quality of the meeting can be evaluated. The current meeting quality assessment usually relies on manual subjective judgment. For different types of meetings, there is a lack of quantitative and standard meeting quality assessment methods, making it difficult to objectively evaluate the quality and effectiveness of the meeting; and the efficiency of meeting quality assessment is low. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a property management meeting quality assessment method and electronic equipment and storage medium, which can automatically assess the quality of the meeting, form an objective and reliable meeting quality assessment report, and improve the efficiency and reliability of the meeting quality assessment.

[0005] In a first aspect, an embodiment of the present application provides a method for evaluating the quality of a property management meeting, comprising:

[0006] Performing a first recognition and evaluation process on the conference voice data obtained from the recording of the property management meeting to obtain the participant identity information and the attendance rate score of each participant;

[0007] The conference voice data is converted into text to obtain first text data including speech timing information and identity information of the participant;

[0008] Performing a second recognition and evaluation process on the first text data and the conference voice data to obtain a conference emotion score of the target audio segment;

[0009] Performing a third identification and evaluation process based on the conference voice data, the speech timing information in the first text data, and the participant identity information to obtain an employee participation score;

[0010] Determining a preset quality assessment standard table corresponding to the conference theme according to the first text data, and performing a fourth identification and assessment process according to the first text data and the quality assessment standard table to obtain a conference completion score;

[0011] Performing weight allocation processing according to the weight priority determined from the quality assessment standard table to obtain a weight coefficient set;

[0012] The meeting quality score is obtained by calculating according to the score adjustment formula, the weight coefficient set, the meeting sentiment score, the meeting completion score, the employee participation score, and the attendance rate score; and a meeting quality report is generated according to the first text data and the meeting quality score.

[0013] According to some embodiments of the present application, the first recognition and evaluation processing is performed on the conference voice data recorded at the property management meeting to obtain the participant identity information and the attendance rate score of each participant, including:

[0014] Identify the conference voice data to obtain the voiceprint information of each participant and determine the actual total number of participants;

[0015] According to the matching of the personnel voiceprint information with the preset voiceprint identity information, determining the participant identity information of each participant;

[0016] The attendance rate score is obtained by multiplying the ratio of the actual total number of participants to the planned total number of participants by 100.

[0017] According to some embodiments of the present application, the converting the conference voice data into text to obtain the first text data including the speech timing information and the identity information of the participant includes:

[0018] Performing voice conversion processing on the conference voice data to obtain second text data;

[0019] The second text data is annotated according to the speech timing information and the participant identity information to obtain the first text data.

[0020] According to some embodiments of the present application, performing a second recognition evaluation process on the first text data and the conference voice data to obtain a conference emotion score of the target audio segment includes:

[0021] Identify emotional keywords;

[0022] Performing a first text recognition process on the first text data according to the emotion keyword to determine an emotion score adjustment coefficient;

[0023] Determining a target audio segment including the emotion keyword from the conference voice data;

[0024] Performing emotion recognition processing on the target audio segment to obtain an initial emotion score;

[0025] The meeting emotion score is determined according to the emotion score adjustment coefficient and the initial emotion score.

[0026] According to some embodiments of the present application, the participant identity information includes: manager, employee; the third identification evaluation process is performed based on the conference voice data, the speech timing information in the first text data, and the participant identity information to obtain an employee participation score, including:

[0027] Counting the total duration of the conference and the silence duration in the conference voice data, and subtracting the silence duration from the total duration of the conference to obtain the effective duration of the conference;

[0028] Determine the speaking time of each employee according to the speaking timing information of the first text data and the identity information of the participant;

[0029] Add up the speaking time of each employee to get the total speaking time of the employee;

[0030] The ratio of the total speaking time of the employee to the effective time of the meeting is multiplied by 100 to obtain the employee participation score.

[0031] According to some embodiments of the present application, determining a preset quality assessment standard table corresponding to a conference theme according to the first text data, and performing a fourth identification and assessment process according to the first text data and the quality assessment standard table to obtain a conference completion score includes:

[0032] Performing a second text recognition process on the first text data to determine conference topic keywords;

[0033] A first flag is identified for the conference theme keyword; wherein different values ​​of the first flag are used to indicate different types of conference themes;

[0034] Determine the quality assessment standard table corresponding to the conference theme in the historical database according to the first mark; the quality assessment standard table includes: a weight coefficient priority table, an agenda level mapping table, and a mark scoring table;

[0035] Performing a third text recognition process on the first text data to obtain an agenda keyword, determining an agenda event according to the agenda keyword, and adding an agenda level mark to the agenda event; the agenda level mark is used to indicate the importance of the agenda event;

[0036] The meeting completion score is calculated based on the statistical number of agenda level marks at each level and the mark scoring table.

[0037] According to some embodiments of the present application, the weight coefficient set includes: an emotion weight coefficient, a completion weight coefficient, a participation weight coefficient, and an attendance rate weight coefficient; the weight allocation process is performed according to the weight priority determined from the quality assessment standard table to obtain a weight coefficient set, including:

[0038] Determining the weight coefficient priority table from the quality assessment standard table;

[0039] The emotion weight coefficient, the completion weight coefficient, the participation weight coefficient, and the attendance rate weight coefficient are calculated according to the weight coefficient priority table, the preset tolerance, the weight sum value of 1, and the arithmetic progression sum formula;

[0040] The weight coefficient set is generated according to the emotion weight coefficient, the completion weight coefficient, the participation weight coefficient, and the attendance rate weight coefficient.

[0041] According to some embodiments of the present application, the score adjustment formula is:

[0042] S=W1*S1+W2*S2+W3*S3+W4*S4; where W1+W2+W3+W4=1; W1, W2, W3, W4 are all natural numbers;

[0043] Among them, S is the meeting quality score, W1 is the emotion weight coefficient, W2 is the completion weight coefficient, W3 is the participation weight coefficient, and W4 is the attendance rate weight coefficient; S1 is the meeting emotion score, S2 is the meeting completion score, S3 is the employee participation score, and S4 is the attendance rate score.

[0044] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the property management meeting quality assessment method as described in the first aspect is implemented.

[0045] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to implement the property management meeting quality assessment method as described in the first aspect when executed by a processor.

[0046] The embodiment of the present application includes: in the process of evaluating the quality of property management meetings, first, performing a first recognition and evaluation process on the conference voice data obtained by recording at the property management meeting to obtain the participant identity information and the attendance rate score of each participant; secondly, performing a text conversion process on the conference voice data to obtain first text data including speech timing information and participant identity information; then, performing a second recognition and evaluation process on the first text data and the conference voice data to obtain the meeting emotion score of the target audio segment; then, performing a third recognition and evaluation process based on the conference voice data, the speech timing information in the first text data and the participant identity information to obtain the employee emotion score. Participation score; then, determine the quality assessment standard table corresponding to the preset meeting theme according to the first text data, and perform the fourth identification and assessment processing according to the first text data and the quality assessment standard table to obtain the meeting completion score; then, perform weight allocation processing according to the weight priority determined from the quality assessment standard table to obtain a weight coefficient set; then, calculate according to the score adjustment formula, weight coefficient set, meeting sentiment score, meeting completion score, employee participation score, and attendance rate score to obtain a meeting quality score; can automatically complete the property meeting quality assessment; finally, generate a meeting quality report according to the first text data and the meeting quality score. Based on this, the embodiment of the present application can automatically evaluate the quality of the meeting, form an objective and reliable meeting quality assessment report, and improve the efficiency and reliability of the meeting quality assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the steps of a property management meeting quality assessment method provided by an embodiment of the present application;

[0048] Figure 2 It is a schematic diagram of a table of different quality assessment standards corresponding to different conference themes provided in one embodiment of the present application;

[0049] Figure 3 It is a schematic diagram of a weight coefficient priority table corresponding to a summary meeting provided by an embodiment of the present application;

[0050] Figure 4 It is a schematic diagram of an agenda level mapping table corresponding to a summary meeting provided by an embodiment of the present application;

[0051] Figure 5 is a schematic diagram of a sign scoring table provided in one embodiment of the present application;

[0052] Figure 6It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0054] It should be understood that in the description of the present application, the orientation descriptions, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0055] It should be noted that although a logical order is shown in the flowchart in the description of the present application, in some cases, the steps shown or described may be performed in an order different from that in the flowchart. In the description of the present application, a number of means one or more, and a plurality of means two or more. The description of "first" and "second" is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0057] The present application provides a property management meeting quality assessment method, electronic device and computer-readable storage medium, and relates to the technical field of meeting assessment. The method includes: identifying and evaluating meeting voice data to obtain participant identity information and attendance rate score; identifying and evaluating meeting voice data, and obtaining a meeting sentiment score and an employee participation score based on the first text data obtained by converting the meeting voice data; performing identification and assessment based on the first text data and the quality assessment standard table determined by the first text data to obtain a meeting completion score; assigning weights based on the weight priority of the quality assessment standard table to obtain a weight coefficient set; calculating based on the score adjustment formula, the weight coefficient set, the meeting sentiment score, the meeting completion score, the employee participation score, and the attendance rate score to obtain a meeting quality score; and generating a meeting quality report. It can automatically evaluate the quality of the meeting, form an objective and reliable meeting quality assessment report, and improve the efficiency and reliability of the meeting quality assessment.

[0058] The embodiments of the present application are further described below in conjunction with the accompanying drawings.

[0059] First, as Figure 1 As shown, the property management meeting quality assessment method may include but is not limited to steps S110 to S170.

[0060] Step S110: Perform a first recognition and evaluation process on the conference voice data recorded at the property management meeting to obtain the participant identity information and attendance rate score of each participant.

[0061] Step S120: converting the conference voice data into text to obtain first text data including speech timing information and participant identity information.

[0062] Step S130: performing a second recognition evaluation process on the first text data and the conference voice data to obtain a conference emotion score of the target audio segment.

[0063] Step S140: performing a third identification and evaluation process based on the conference voice data, the speech timing information in the first text data, and the identity information of the participants to obtain an employee participation score.

[0064] Step S150: determining a quality assessment standard table corresponding to a preset conference theme according to the first text data, and performing a fourth recognition and assessment process according to the first text data and the quality assessment standard table to obtain a conference completion score.

[0065] Step S160: Perform weight allocation processing according to the weight priority determined from the quality evaluation standard table to obtain a weight coefficient set.

[0066] Step S170: Calculate the meeting quality score according to the score adjustment formula, the weight coefficient set, the meeting sentiment score, the meeting completion score, the employee participation score, and the attendance rate score; and generate a meeting quality report according to the first text data and the meeting quality score.

[0067] Specifically, the identity information of participants includes: managers and employees.

[0068] Specifically, the weight coefficient set includes: emotion weight coefficient, completion weight coefficient, participation weight coefficient, and attendance rate weight coefficient;

[0069] According to some embodiments of the present application, the score adjustment formula is: S=W1*S1+W2*S2+W3*S3+W4*S4; wherein, W1+W2+W3+W4=1; W1, W2, W3, W4 are all natural numbers.

[0070] Among them, S is the meeting quality score, W1 is the emotion weight coefficient, W2 is the completion weight coefficient, W3 is the participation weight coefficient, and W4 is the attendance rate weight coefficient; S1 is the meeting emotion score, S2 is the meeting completion score, S3 is the employee participation score, and S4 is the attendance rate score.

[0071] Specifically, the quality assessment standard table includes: weight coefficient priority table, agenda level mapping table, and mark scoring table; providing a reliable reference for obtaining multiple meeting indicator scores. The quality assessment standard table can be customized, pre-configured, and can be flexibly applied.

[0072] The conference quality report generated by step S170 provides a reliable reference for subsequent improvement of the conference and improvement of the conference quality.

[0073] Through steps S110 to S170, in the process of evaluating the quality of the property management meeting, first, the conference voice data obtained by recording at the property management meeting is subjected to a first recognition and evaluation process to obtain the participant identity information and the attendance rate score of each participant; secondly, the conference voice data is converted into text to obtain first text data including speech timing information and participant identity information; then, the first text data and the conference voice data are subjected to a second recognition and evaluation process to obtain the conference emotion score of the target audio segment; then, a third recognition and evaluation process is performed based on the conference voice data, the speech timing information in the first text data, and the participant identity information. Obtain an employee participation score; then, determine the quality assessment standard table corresponding to the preset meeting theme according to the first text data, and perform a fourth identification and assessment process according to the first text data and the quality assessment standard table to obtain a meeting completion score; then, perform a weight allocation process according to the weight priority determined from the quality assessment standard table to obtain a weight coefficient set; then, calculate according to the score adjustment formula, the weight coefficient set, the meeting sentiment score, the meeting completion score, the employee participation score, and the attendance rate score to obtain a meeting quality score; the property meeting quality assessment can be automatically completed; finally, a meeting quality report is generated according to the first text data and the meeting quality score. Therefore, the embodiment of the present application can automatically evaluate the quality of the meeting, form an objective and reliable meeting quality assessment report, and improve the efficiency and reliability of the meeting quality assessment.

[0074] According to some embodiments of the present application, step S110 is further described, and step S110 includes but is not limited to steps S111 to S113.

[0075] Step S111: Identify the conference voice data to obtain the voiceprint information of each participant and determine the actual total number of participants.

[0076] Step S112: Determine the participant identity information of each participant based on matching the personnel voiceprint information with the preset voiceprint identity information.

[0077] Step S113: multiply the ratio of the actual total number of participants to the planned total number of participants by 100 to obtain the attendance rate score.

[0078] Specifically, step S111 and step S112 are further explained. First, the conference voice data is preprocessed to obtain preprocessed conference voice data; the preprocessing includes: sampling, windowing, filtering, pre-emphasis and other processing, so as to better facilitate subsequent feature extraction and pattern matching. Secondly, voiceprint feature extraction is performed on the preprocessed conference voice data to obtain different voiceprint features to be identified. Each voiceprint feature to be identified is the voiceprint information of a participant; the actual total number of participants is determined based on the number of voiceprint features. Then, the voiceprint features to be identified are compared with the preset voiceprint identity information stored in the database to calculate the voiceprint similarity; based on the voiceprint similarity, the employee name information corresponding to the voiceprint features to be identified and the participant identity information corresponding to the employee name information are determined.

[0079] Specifically, the participant identity information includes: managers and / or employees.

[0080] Specifically, to further explain step S113, the methods for determining the total number of planned participants include: when holding multiple or single department meetings, determining the corresponding departments that should attend the property management meeting based on the number of managers in the participant identity information, and determining the total number of planned participants based on the actual total number of multiple or single departments; or, when holding a management meeting attended only by managers, the total number of planned participants can be directly configured; or, when holding a company-wide meeting, the number of people in the entire company can be directly obtained as the total number of planned participants. Therefore, this application does not impose specific restrictions on the method for determining the total number of planned participants.

[0081] The formula for calculating the attendance rate score in step S113 is: attendance rate score S4 = total number of actual attendees / total number of planned attendees*100.

[0082] Through steps S111 to S113, voiceprint recognition technology and voiceprint matching technology can be used to quickly determine the total number of actual participants and the identity information of each participant from the conference voice data; and the attendance rate score, a conference quality evaluation indicator, can be calculated relatively quickly to provide a reliable reference for subsequent evaluation of the overall property management meeting quality.

[0083] According to some embodiments of the present application, step S120 is further described, and step S120 includes but is not limited to steps S121 to S122.

[0084] Step S121: Perform voice conversion processing on the conference voice data to obtain second text data.

[0085] Step S122: annotate the second text data according to the speech timing information and the participant identity information to obtain the first text data.

[0086] Specifically, in step S121, a preset speech recognition engine is used in the speech conversion process to recognize the conference voice data and convert it into second text data; wherein, the speech recognition engine is configured with a deep learning algorithm, and the acquired conference voice data is converted into second text data using the deep learning algorithm; it should be noted that the second text data is text content information converted from pure speech.

[0087] Specifically, in step S122, text annotation processing is performed on the second text data, and speech timing information and participant identity information are annotated in the second text data to obtain a text data.

[0088] It should be noted that the speech conversion process and the text annotation process can be performed synchronously or asynchronously (ie, the text annotation process is performed after a preset period of time).

[0089] Through steps S121 to S122, the conference voice data is voice-converted and text-annotated to obtain first text data including speech timing information and participant identity information, thereby laying a data foundation for further evaluation and identification based on the text content information, speech timing information, and participant identity information of the first text data.

[0090] According to some embodiments of the present application, step S130 is further described. Step S130: perform a second recognition evaluation process on the first text data and the conference voice data to obtain a conference emotion score of the target audio segment, including but not limited to steps S131 to S135.

[0091] Step S131: Determine emotional keywords.

[0092] Step S132: performing a first text recognition process on the first text data according to the emotion keywords to determine the emotion score adjustment coefficient.

[0093] Step S133: Determine a target audio segment including emotion keywords from the conference voice data.

[0094] Step S134: performing emotion recognition processing on the target audio segment to obtain an initial emotion score.

[0095] Step S135: Determine the meeting emotion score according to the emotion score adjustment coefficient and the initial emotion score.

[0096] Specifically, the emotional keywords in step S131 include: positive keywords and negative keywords. Positive keywords include: good, very good, excellent, etc. Negative keywords include: bad, poor, poor, backward, unqualified, etc.

[0097] Specifically, step S132 includes: first, performing a first text recognition process on the first text data according to the emotional keywords to obtain the recognition situation; specifically, there are four recognition situations: the first recognition result is that only positive keywords are recognized, the second recognition result is that only negative keywords are recognized, the third recognition result is that both positive keywords and negative keywords are recognized, and the fourth recognition result is that neither positive keywords nor negative keywords are recognized. Then, the emotional score adjustment coefficient is determined according to the recognition situation; the emotional score adjustment coefficient A includes the first adjustment coefficient A1 and the second adjustment coefficient A2. Specifically, when only positive keywords are recognized, the first adjustment coefficient A1=1 and the second adjustment coefficient A2=0 are determined; when only negative keywords are recognized, the first adjustment coefficient A1=0 and the second adjustment coefficient A2=-1 are determined; when both positive keywords and negative keywords are recognized, the first adjustment coefficient A1=0.5 and the second adjustment coefficient A2=-0.5 are determined; when neither positive keywords nor negative keywords are recognized, the first adjustment coefficient A1=0 and the second adjustment coefficient A2=0 are determined.

[0098] Specifically, in step S133, the target sentence containing the emotional keywords is determined in the first text data, and the start time and end time of the target sentence are determined according to the speech timing information; the conference voice data is edited based on the start time and end time of the target paragraph to obtain the target audio segment including the emotional keywords.

[0099] It should be noted that when there are multiple emotion keywords, multiple target audio segments are obtained accordingly, which are denoted as Vi, where i represents the sequence number of the target audio segment.

[0100] Understandably, in some cases, although employees may verbally say positive slogans, their actual attitudes are not positive. Therefore, it is inaccurate to judge whether an employee is positive or negative only by identifying emotional keywords through text. Therefore, it is necessary to further identify the target audio segment to judge the mood of the participants and the atmosphere of the meeting.

[0101] Specifically, in step S134, first, the target audio segment is converted into a sound waveform image representing the sound wave of the voice information; then, the sound waveform image is input into the pre-trained emotion recognition scoring model; the emotion recognition scoring model is a neural network model trained to a convergence state for identifying the speech emotion information represented by the sound wave image and evaluating the initial emotion score; the emotion classification result and the initial emotion score are output through the emotion recognition model. The initial emotion score is a positive value, and the larger the initial emotion score, the higher the positivity or negativity. The initial emotion score is recorded as F, and Fi is used to represent the initial emotion score of the i-th target audio segment Vi.

[0102] It is understood that components of a speaker’s voice, such as frequency or pitch, reflect underlying emotions. For example, when a speaker is emotional, their voice may be louder or have a higher pitch. Neural network models are able to perform emotion recognition by learning the association between voice features and emotions.

[0103] Specifically, in step S135 , the formula used to determine the conference emotion score according to the emotion score adjustment coefficient and the initial emotion score is: S1=AF.

[0104] For example, when only positive keywords are identified, the first adjustment coefficient A1=1 and the second adjustment coefficient A2=0 are determined; and there are three target audio segments including positive keywords: V1, V2, and V3; the initial emotion scores of the three target audio segments are: F1=5, F2=7, and F3=9; then the meeting emotion score S1=A1*(F1+F2+F3)=21.

[0105] For example, when only negative keywords are identified, the first adjustment coefficient A1=0 and the second adjustment coefficient A2=-1 are determined; there are three target audio segments including negative keywords: V1, V2, and V3; the initial emotion scores of the three target audio segments are: F1=5, F2=7, and F3=9; then the meeting emotion score S1=A2*(F1+F2+F3)=-21.

[0106] For example, when both positive and negative keywords are identified, the first adjustment coefficient A1=0.5 and the second adjustment coefficient A2=-0.5 are determined. There are two target audio segments including positive keywords: V1 and V2; there is one target audio segment including negative keywords: V3; the initial emotion scores of the three target audio segments are: F1=5, F2=7, F3=9; then the meeting emotion score S1=A1*(F1+F2)+A2*F3=1.5.

[0107] For example, when neither positive keywords nor negative keywords are identified, the first adjustment coefficient A1 is determined to be 0, and the second adjustment coefficient A2 is determined to be 0. Then the meeting emotion score S1 is 0.

[0108] Through steps S131 to S135, the two factors of speech text content and voice emotion can be comprehensively considered to automatically determine the meeting emotion score, which is conducive to improving the efficiency of property management meeting quality assessment, and at the same time, improving the objectivity and reliability of property management meeting quality assessment.

[0109] According to some embodiments of the present application, the identity information of the participants includes: managers, employees. Further explanation of step S140, step S140: performing a third identification evaluation process according to the conference voice data, the speech timing information in the first text data and the identity information of the participants to obtain an employee participation score, including but not limited to steps S141 to S144.

[0110] Step S141: Count the total conference duration and silence duration in the conference voice data, and subtract the silence duration from the total conference duration to obtain the effective conference duration.

[0111] Step S142: Determine the speaking time of each employee according to the speaking timing information of the first text data and the identity information of the participant.

[0112] Step S143: Add up the speaking time of each employee to obtain the total speaking time of the employees.

[0113] Step S145: multiply the ratio of the total speaking time of employees to the effective meeting time by 100 to obtain the employee participation score.

[0114] Specifically, when a recording device is used to record a meeting, the recording device is usually turned on in advance, which will collect some useless audio. In addition, there may be situations in the meeting where no one is talking or filling out questionnaires, and silent audio with no one talking will be collected. Clips with a volume less than a certain threshold can be determined as silent audio. Therefore, the effective duration of the meeting obtained through the processing of step S141 can provide a more reliable data basis for the subsequent calculation of the employee engagement score.

[0115] Specifically, through step S142 and step S145, the speaking time of each employee is identified from the first to the last in sequence from the speech timing information of the first text data, and the employee speaking time is accumulated to obtain the total speaking time of the employee; finally, based on the employee participation score formula: employee participation score S3 = total speaking time of the employee / effective meeting time * 100. The employee participation score is used to reflect the degree of employee participation in the meeting. The higher the employee participation score, the higher the degree of employee participation; conversely, the lower the employee participation score, the lower the degree of employee participation.

[0116] Through steps S141 to S144, it is possible to quickly perform recognition and calculation based on the conference voice data and the first text data obtained by recognition to obtain an employee participation score, which objectively reflects the degree of employee participation in the meeting; this is conducive to improving the efficiency of property management meeting quality assessment, and at the same time, improving the objectivity and reliability of property management meeting quality assessment.

[0117] According to some embodiments of the present application, step S150 is further described, and step S150 includes but is not limited to steps S151 to S155.

[0118] Step S151: Perform a second text recognition process on the first text data to determine conference topic keywords.

[0119] Step S152: A first flag is identified for the conference theme keyword; wherein different values ​​of the first flag are used to indicate different types of conference themes.

[0120] Step S153: Determine the quality assessment standard table corresponding to the conference theme in the historical database according to the first mark; the quality assessment standard table includes: a weight coefficient priority table, an agenda level mapping table, and a mark scoring table.

[0121] Step S154: Perform third text recognition processing on the first text data to obtain agenda keywords, determine agenda events based on the agenda keywords, and add agenda level marks to the agenda events; the agenda level marks are used to indicate the importance of the agenda events.

[0122] Step S155: Calculate based on the number of agenda level marks at each level and the mark scoring table to obtain the meeting completion score.

[0123] It is understandable that the second text recognition process uses a keyword recognition algorithm. The keyword recognition algorithm used can be: a keyword recognition algorithm based on deep learning, a recognition algorithm that directly matches preset keywords to determine the target keyword. This application does not specifically limit the type of keyword recognition algorithm used.

[0124] Specifically, there are many key words for conference themes, including but not limited to: mobilization, summary, learning, sharing, etc.

[0125] It can be understood that step S152 marks the first mark for the conference theme keyword, so as to quickly determine the quality assessment standard table to be adopted based on the first mark. Specifically, when the conference theme keyword is "mobilization", the first mark with a value of 0 is used to identify the conference theme keyword; when the conference theme keyword is "summary", the first mark with a value of 1 is used to identify the conference theme keyword; when the conference theme keyword is "experience sharing", the first mark with a value of 2 is used to identify the conference theme keyword.

[0126] It is understandable that different conference themes have different quality assessment standard tables; different quality assessment standard tables are stored in the database such as Figure 2 As shown, the controller that executes the property management meeting quality assessment method can match and obtain the required quality assessment standard table from the database.

[0127] like Figure 2As shown, the quality assessment standard table also has a type number configured, which is used to indicate the type of meeting to which the quality assessment standard table applies. For example, if the type number is 0, the quality assessment standard table applies to meetings with the meeting theme keyword "mobilization"; if the type number is 1, the quality assessment standard table applies to meetings with the meeting theme keyword "summary". If the type number is 2, the quality assessment standard table applies to meetings with the meeting theme keyword "experience sharing". By analogy, the quality assessment standard table can be customized.

[0128] The weight coefficient priority table included in the quality assessment standard table is further explained. It can be understood that the evaluation indicators involved in this solution are: meeting sentiment score S1, meeting completion score S2, employee participation score S3, and attendance rate score S4. However, when evaluating the quality of different types of meetings, the importance and priority of each evaluation indicator are different. Therefore, after obtaining the score of each indicator, it is necessary to determine the weight coefficient of each evaluation indicator and the priority of the weight coefficient. For example, Figure 3 As shown, when the meeting is an annual summary meeting, the most important and top-priority evaluation indicator is the completion weight coefficient W2, which measures the process completion of the annual summary meeting; followed by the attendance rate weight coefficient W4, followed by the emotion weight coefficient W1, and finally the participation weight coefficient W3. In addition, when the meeting is a mobilization meeting, the most important and top-priority evaluation indicator is the attendance rate weight coefficient W4, followed by the emotion weight coefficient W1, followed by the completion weight coefficient W2, and finally the participation weight coefficient W3. It is understandable that two or three of the coefficients can also be set to the same priority.

[0129] Further explanation of the agenda level mapping table included in the quality assessment criteria table. Figure 4 As shown, the agenda level mapping table includes: various agenda keywords, agenda events corresponding to various agenda keywords, and corresponding level marks of various types of agenda events; among the corresponding level marks, the agenda indicated by the first-level mark is more important, the agenda indicated by the second-level mark is less important, and so on. Specifically, the agenda event corresponding to multiple keywords such as "plan" and "planning" is the announcement of the plan, corresponding to the first-level mark; the agenda event corresponding to keywords such as "in short", "in general", and "summarize as follows" is the work summary, corresponding to the first-level mark; the agenda event corresponding to keywords such as "commendation" and "reward" is commendation, corresponding to the second-level mark; the agenda event corresponding to keywords such as "hope" and "suggestion" is the exchange of opinions, corresponding to the third-level mark.

[0130] To further illustrate step S154, the third text recognition process uses a keyword recognition algorithm, and the present application does not impose specific restrictions on the keyword recognition algorithm. When an agenda keyword is recognized, the agenda event is determined based on the agenda keyword, and an agenda level mark is added to the agenda event. It is understandable that there can be agenda events of the same importance; in general meetings, each type of agenda only appears once; therefore, even if multiple keywords such as "plan" and "planning" are recognized, only one agenda event is recorded; at this time, there is 1 first-level mark.

[0131] The mark scoring table included in step S155 and the quality evaluation standard table is further described. Figure 5 As shown, the mark scoring table is used to indicate the mapping relationship between the level mark and the score, wherein the score shows a small change trend according to the order of importance of the level mark from high to low. Specifically, the score of the first level mark is 5 points, the score of the second level mark is 3 points, and the score of the third level mark is 1 point. It is understandable that the score can be adjusted, and this application does not make specific restrictions on this.

[0132] Through step S155, a more objective meeting completion score can be obtained by calculating based on the number of agenda level marks at each level and the mark scoring table.

[0133] For example, when the three agenda events of plan announcement, work summary, and opinion exchange are identified and the marks are completed, two first-level marks and one third-level mark can be detected accordingly, and the meeting completion score S2=2*5+1*1=11 points.

[0134] According to some embodiments of the present application, the weight coefficient set includes: emotion weight coefficient, completion weight coefficient, participation weight coefficient, attendance rate weight coefficient; further illustrating step S160, step S160 includes but is not limited to steps S161 to S163.

[0135] Step S161: determining a weight coefficient priority table from a quality assessment standard table;

[0136] Step S162: Calculate according to the weight coefficient priority table, the preset tolerance, the weight sum value of 1, and the arithmetic progression sum formula to obtain the emotion weight coefficient, the completion weight coefficient, the participation weight coefficient, and the attendance rate weight coefficient;

[0137] Step S163: Generate a weight coefficient set according to the emotion weight coefficient, the completion weight coefficient, the participation weight coefficient, and the attendance rate weight coefficient.

[0138] Specifically, the formula for summing an arithmetic sequence is: SUM=na1+n(n-1)d / 2; wherein a1 is the first term of the arithmetic sequence, d is the common difference, and n is the number of terms.

[0139] Take an example to illustrate step S162. First, determine a sequence of unknown values ​​according to the weight coefficient priority table; Figure 3 For example, the obtained sequence is W2, W4, W1, W3. Set the sequence: W2, W4, W1, W3 as an arithmetic sequence, and W2 is the first term a1. Set the preset tolerance d = 0.1. The weight sum is 1, that is, SUM = 1. Substitute a1 = W2, d = 0.1, SUM = 1 into SUM = na1 + n (n-1) d / 2; we can get a1 = W2 = 0.4, and so on to get: W4 = 0.3, W1 = 0.2, W3 = 0.1. In this way, the values ​​of each weight coefficient are distributed more reasonably.

[0140] Through step S161 to step S163, combined with the weight coefficient priority table and the arithmetic progression sum formula, the values ​​of each weight coefficient are allocated more reasonably, further improving the reliability of conference quality assessment.

[0141] It can be understood that the above-mentioned embodiment can determine the specific values ​​of eight parameters: emotion weight coefficient W1, completion weight coefficient W2, participation weight coefficient W3, attendance rate weight coefficient W4, meeting emotion score S1, meeting completion score S2, employee participation score S3, and attendance rate score S4, and substitute these eight parameters into the score adjustment formula: S=W1*S1+W2*S2+W3*S3+W4*S4; the final meeting quality score S can be obtained.

[0142] like Figure 6 As shown, the present application also provides an electronic device, including:

[0143] The processor 601 may be implemented by a general-purpose central processing unit, a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0144] The memory 602 may be implemented in the form of a read-only memory, a static storage device, a dynamic storage device, or a random access memory. The memory 602 may store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 602, and the processor 601 calls and executes the property management meeting quality assessment method of the embodiment of this application;

[0145] Input / output interface 603, used to implement information input and output;

[0146] Communication interface 604, used to realize communication interaction between the present apparatus and other devices, which can be realized by wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WI FI, Bluetooth, etc.);

[0147] Bus 605 , which transmits information between various components of the device (e.g., processor 601 , memory 602 , input / output interface 603 , and communication interface 604 );

[0148] The processor 601 , the memory 602 , the input / output interface 603 and the communication interface 604 are connected to each other in communication within the device via a bus 605 .

[0149] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned property management meeting quality assessment method is implemented.

[0150] As a non-transient computer-readable storage medium, the memory can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are implemented to be located in one place, or may also be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0151] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically include computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0152] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions without violating the spirit of the present application. These equivalent deformations or substitutions are all included in the scope defined by the present application.

Claims

1. A method for evaluating the quality of a property management meeting, characterized in that: include: Performing a first recognition and evaluation process on the conference voice data obtained from the recording of the property management meeting to obtain the participant identity information and the attendance rate score of each participant; The conference voice data is converted into text to obtain first text data including speech timing information and identity information of the participant; Performing a second recognition and evaluation process on the first text data and the conference voice data to obtain a conference emotion score of the target audio segment; Performing a third identification and evaluation process based on the conference voice data, the speech timing information in the first text data, and the participant identity information to obtain an employee participation score; Determining a preset quality assessment standard table corresponding to the conference theme according to the first text data, and performing a fourth identification and assessment process according to the first text data and the quality assessment standard table to obtain a conference completion score; Performing weight allocation processing according to the weight priority determined from the quality assessment standard table to obtain a weight coefficient set; The meeting quality score is obtained by calculating according to the score adjustment formula, the weight coefficient set, the meeting sentiment score, the meeting completion score, the employee participation score, and the attendance rate score; and a meeting quality report is generated according to the first text data and the meeting quality score.

2. The property management meeting quality assessment method according to claim 1, characterized in that: The first recognition and evaluation process is performed on the conference voice data recorded at the property management meeting to obtain the participant identity information and the attendance rate score of each participant, including: Identify the conference voice data to obtain the voiceprint information of each participant and determine the actual total number of participants; According to the matching of the personnel voiceprint information with the preset voiceprint identity information, determining the participant identity information of each participant; The attendance rate score is obtained by multiplying the ratio of the actual total number of participants to the planned total number of participants by 100.

3. The property management meeting quality assessment method according to claim 1, characterized in that: The converting the conference voice data into text to obtain first text data including speech timing information and the identity information of the participant includes: Performing voice conversion processing on the conference voice data to obtain second text data; The second text data is annotated according to the speech timing information and the participant identity information to obtain the first text data.

4. The property management meeting quality assessment method according to claim 1, characterized in that: The performing a second recognition and evaluation process on the first text data and the conference voice data to obtain a conference emotion score of the target audio segment includes: Identify emotional keywords; Performing a first text recognition process on the first text data according to the emotion keyword to determine an emotion score adjustment coefficient; Determining a target audio segment including the emotion keyword from the conference voice data; Performing emotion recognition processing on the target audio segment to obtain an initial emotion score; The meeting emotion score is determined according to the emotion score adjustment coefficient and the initial emotion score.

5. The property management meeting quality assessment method according to claim 4, characterized in that: The participant identity information includes: managers, employees; the third identification evaluation process is performed based on the conference voice data, the speech timing information in the first text data, and the participant identity information to obtain an employee participation score, including: Counting the total duration of the conference and the silence duration in the conference voice data, and subtracting the silence duration from the total duration of the conference to obtain the effective duration of the conference; Determine the speaking time of each employee according to the speaking timing information of the first text data and the identity information of the participant; Add up the speaking time of each employee to get the total speaking time of the employee; The ratio of the total speaking time of the employee to the effective time of the meeting is multiplied by 100 to obtain the employee participation score.

6. The property management meeting quality assessment method according to claim 1, characterized in that: The step of determining a preset quality assessment standard table corresponding to a conference theme according to the first text data, and performing a fourth identification and assessment process according to the first text data and the quality assessment standard table to obtain a conference completion score includes: Performing a second text recognition process on the first text data to determine conference topic keywords; A first flag is identified for the conference theme keyword; wherein different values ​​of the first flag are used to indicate different types of conference themes; Determine the quality assessment standard table corresponding to the conference theme in the historical database according to the first mark; the quality assessment standard table includes: a weight coefficient priority table, an agenda level mapping table, and a mark scoring table; Performing a third text recognition process on the first text data to obtain an agenda keyword, determining an agenda event according to the agenda keyword, and adding an agenda level mark to the agenda event; the agenda level mark is used to indicate the importance of the agenda event; The meeting completion score is calculated based on the statistical number of agenda level marks at each level and the mark scoring table.

7. The property management meeting quality assessment method according to claim 6, characterized in that: The weight coefficient set includes: emotion weight coefficient, completion weight coefficient, participation weight coefficient, and attendance rate weight coefficient; the weight allocation process is performed according to the weight priority determined from the quality assessment standard table to obtain a weight coefficient set, including: Determining the weight coefficient priority table from the quality assessment standard table; The emotion weight coefficient, the completion weight coefficient, the participation weight coefficient, and the attendance rate weight coefficient are calculated according to the weight coefficient priority table, the preset tolerance, the weight sum value of 1, and the arithmetic progression sum formula; The weight coefficient set is generated according to the emotion weight coefficient, the completion weight coefficient, the participation weight coefficient, and the attendance rate weight coefficient.

8. The method for evaluating the quality of property management meetings according to claim 7, characterized in that: The score adjustment formula is: S=W1*S1+W2*S2+W3*S3+W4*S4; where W1+W2+W3+W4=1; W1, W2, W3, W4 are all natural numbers; Among them, S is the meeting quality score, W1 is the emotion weight coefficient, W2 is the completion weight coefficient, W3 is the participation weight coefficient, and W4 is the attendance rate weight coefficient; S1 is the meeting emotion score, S2 is the meeting completion score, S3 is the employee participation score, and S4 is the attendance rate score.

9. An electronic device, characterized in that: It includes at least one processor and a memory for communicating with the at least one processor; the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the property management meeting quality assessment method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the property management meeting quality assessment method as described in any one of claims 1 to 8.

Images