Practical training behavior monitoring and analysis method and device, equipment and storage medium

Abstract

The invention belongs to the technical field of practical training monitoring, and particularly relates to a practical training behavior monitoring and analysis method, device and equipment and a storage medium, and the method comprises the steps: S1, a first monitoring module carries out the collection and processing of the utterance data of a practical training person, and a second monitoring module carries out the collection and processing of the motion data of the practical training person; s2, the local analysis module compares and analyzes the features of the received utterance data with the features of different standard utterance data at the same time to determine a transmission moment, and transmits the received utterance data to a remote analysis module from the transmission moment; s3, the remote analysis module compares and analyzes the features of the utterance data sent by the local analysis module with the features of the partial standard utterance data to obtain a second analysis result, and determines an utterance analysis result based on the first analysis result and the second analysis result; and S4, the remote analysis module judges a final analysis result according to the utterance analysis result and the action analysis result. According to the invention, the recognition accuracy of the utterance data can be ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of practical training monitoring, and specifically relates to a practical training behavior monitoring and analysis method, device, equipment and storage medium. Background Art

[0002] The method of manually evaluating the performance of trainees at the aircraft emergency response training site results in a high cost of training. Therefore, a method of automatically collecting the speech data of trainees and automatically analyzing the collected speech data was developed, so that the performance of trainees can be automatically evaluated.

[0003] A similar prior art includes a Chinese patent application with publication number CN101622660A, which provides a speech recognition device, a speech recognition method and a speech recognition program, wherein a speech model storage unit (7) pre-stores speech models with multiple detail levels that represent speech feature properties, a detail level judgment unit (9) selects a detail level that is closest to the feature properties of an input signal from among the detail levels of the speech models stored in the speech model storage unit (7), and a parameter setting unit (10) controls speech recognition-related parameters according to the selected detail level. A similar prior art also includes a Chinese patent application with publication number CN104485106A, which proposes a speech recognition method, a speech recognition system and a speech recognition device, wherein the speech recognition method includes: obtaining a speech string; creating a grammar analysis tree according to the speech string, the grammar analysis tree being used to determine whether the speech string can be recognized for the first time and to parse the word-unit attribute information of at least one word-unit in the speech string; creating a semantic analysis tree according to the result of the first recognition, the semantic analysis tree obtaining pre-stored attribute information according to the parsed word-unit attribute information to create a speech recognition result. However, the recognition methods of the above two patent applications are relatively complex, and the recognition accuracy needs to be improved. Summary of the invention

[0004] A first monitoring module is set to collect speech data, and a second monitoring module is set to collect action data. The first collection module continuously sends speech data to the local analysis module, but only sends speech data to the remote analysis module when the transmission time arrives. The remote analysis module determines the speech analysis result according to the first analysis result and the second analysis result, and determines the final analysis result according to the speech analysis result and the action analysis result. The present invention aims to provide a method for identifying speech data that is simple, fast and highly accurate.

[0005] In order to achieve the above-mentioned purpose of the invention, the present invention provides a training behavior monitoring and analysis method as described below, which mainly includes the following steps:

[0006] S1. A first monitoring module and a second monitoring module are set at a training site, wherein the first monitoring module collects and processes speech data of trainees, and the second monitoring module collects and processes motion data of trainees, and the first monitoring module sends the collected speech data to a local analysis module set at the training site, and the second monitoring module transmits the collected motion data to a remote analysis module;

[0007] S2, the local analysis module compares and analyzes the features of the received speech data with the features of different pre-stored standard speech data to determine the transmission time, and the local analysis module transmits the received speech data to the remote analysis module starting from the transmission time, and the local analysis module also transmits the first analysis result of the received speech data to the remote analysis module;

[0008] S3, the remote analysis module compares and analyzes the features of the speech data sent by the local analysis module with the features of some pre-stored standard speech data to obtain a second analysis result, and the remote analysis module determines a speech analysis result based on the first analysis result and the second analysis result;

[0009] S4. The remote analysis module receives the action data sent by the second monitoring module, compares and analyzes the features of the received action data with the features of different pre-stored standard action data to obtain an action analysis result, and determines the final analysis result based on the speech analysis result and the action analysis result.

[0010] As a preferred technical solution of the present invention, the standard discourse data includes several words, and the process of generating partial standard discourse data corresponding to the standard discourse data includes dividing the several words constituting the standard discourse data into two groups, front and back, so that different words in the latter group constitute partial standard discourse data.

[0011] As a preferred technical solution of the present invention, S2 comprises the following steps:

[0012] S21, the local analysis module compares and analyzes the features of the standard speech data with the features of the received speech data for each standard speech data, calculates the difference degree value between the two, and the local analysis module determines whether there is a difference degree value that is changing from greater than the difference degree value threshold 1 to less than the difference degree value threshold 1. If not, repeat this step. If yes, take the moment of successful judgment as the transmission moment, start transmitting the received speech data to the remote analysis module from the transmission moment, and proceed to the next step;

[0013] S22, the local analysis module calculates the time interval from the transmission moment to the current moment, compares and analyzes the features of the received speech data with the features of the standard speech data corresponding to the difference degree value determined to exist, and calculates the difference degree value between the two. When the time interval does not reach the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, repeat this step. When the time interval does not reach the first time interval threshold and the difference degree value is less than or equal to the difference degree value threshold 2, generate the first analysis result of successful analysis, and transmit the first analysis result to the remote analysis module. When the time interval reaches the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, generate the first analysis result of failed analysis, and transmit the first analysis result to the remote analysis module.

[0014] As a preferred technical solution of the present invention, the difference degree value threshold 2 is less than the difference degree value threshold 1.

[0015] As a preferred technical solution of the present invention, S3 comprises the following steps:

[0016] S31, when the remote analysis module obtains the speech data sent by the local analysis module, it compares and analyzes the characteristics of the speech data with the characteristics of some pre-stored standard speech data, calculates the difference between the two, and calculates the time interval from the start time of the analysis process to the current time;

[0017] S32, when the time interval does not reach the second time interval threshold and the difference degree value is greater than the difference degree value threshold three, the remote analysis module repeatedly executes S31, when the time interval does not reach the second time interval threshold and the difference degree value is less than or equal to the difference degree value threshold three, the remote analysis module generates the second analysis result of successful analysis and ends the analysis process, when the time interval reaches the second time interval threshold and the difference degree value is greater than the difference degree value threshold three, the remote analysis module generates the second analysis result of failed analysis and ends the analysis process;

[0018] S33: When both the first analysis result and the second analysis result are successful, the remote analysis module sets the discourse analysis result as successful; otherwise, the remote analysis module sets the discourse analysis result as failed.

[0019] As a preferred technical solution of the present invention, the local analysis module transmits the received speech data to the remote analysis module, comprising the following steps:

[0020] S211, the local analysis module sets the count value p to 1, and the local analysis module regards the received speech data as initial data, divides the initial data into first speech data and second speech data, performs a specific operation on the first speech data and the second speech data to obtain p-th intermediate data, and increases the count value p by 1;

[0021] S212, the local analysis module regards the first utterance data as initial data, continues to divide the initial data into first utterance data and second utterance data, performs a specific operation on the first utterance data and the second utterance data to obtain pth intermediate data, and the local analysis module determines whether the end condition is met, if not, increments the count value p by 1, repeats this step, if yes, proceeds to the next step;

[0022] S213: The local analysis module performs secret processing on the last first speech data to obtain secret data, and the local analysis module sends the secret data and the 1st to pth intermediate data to the remote analysis module.

[0023] As a preferred technical solution of the present invention, the remote analysis module obtains the speech data sent by the local analysis module through the following steps:

[0024] S311, the local analysis module recovers the secret data to obtain the last first utterance data, performs a specific operation on the last first utterance data and the pth intermediate data to obtain the last second utterance data, recovers the last first utterance data from the last first utterance data and the last second utterance data, and decrements the count value p by 1;

[0025] S312, the local analysis module performs a specific operation on the previous first utterance data and the pth intermediate data to obtain the previous second utterance data, recovers the new previous first utterance data from the previous first utterance data and the previous second utterance data, decrements the count value p by 1, and determines whether the count value p is equal to 1. If not, repeat this step, and if yes, proceed to the next step;

[0026] S313: The local analysis module performs a specific operation on the previous first utterance data and the first intermediate data to obtain the previous second utterance data, and restores the initial data from the previous first utterance data and the previous second utterance data.

[0027] The present invention also provides a training behavior monitoring and analysis device, comprising the following modules:

[0028] A first monitoring module is used to be set up at the training site, collect and process the speech data of the trainees, and send the collected speech data to a local analysis module set up at the training site;

[0029] The second monitoring module is used to be set up at the training site, collect and process the motion data of the trainees, and transmit the collected motion data to the remote analysis module;

[0030] A local analysis module, configured to compare and analyze the features of the received speech data with the features of different pre-stored standard speech data to determine a transmission time, and to transmit the received speech data to the remote analysis module starting from the transmission time, and to transmit a first analysis result of the received speech data to the remote analysis module;

[0031] The remote analysis module is used to compare and analyze the characteristics of the speech data sent by the local analysis module with the characteristics of some pre-stored standard speech data to obtain a second analysis result, determine the speech analysis result based on the first analysis result and the second analysis result, and receive the action data sent by the second monitoring module, compare and analyze the characteristics of the received action data with the characteristics of different pre-stored standard action data to obtain the action analysis result, and determine the final analysis result based on the speech analysis result and the action analysis result.

[0032] The present invention also provides a device, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to implement any one of the above-mentioned methods when executing the computer program.

[0033] The present invention also provides a storage medium, wherein the storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute any one of the above methods.

[0034] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0035] In the present invention, first, the first monitoring module collects and processes the speech data of the trainees, and the second monitoring module collects and processes the action data of the trainees. The first monitoring module sends the collected speech data to the local analysis module set at the training site, and the second monitoring module transmits the collected action data to the remote analysis module. Secondly, the local analysis module compares and analyzes the features of the received speech data with the features of different standard speech data to determine the transmission time, transmits the received speech data to the remote analysis module from the transmission time, and also transmits the first analysis result of the received speech data to the remote analysis module. Thirdly, the remote analysis module compares and analyzes the features of the speech data sent by the local analysis module with the features of some standard speech data to obtain the second analysis result, and determines the speech analysis result based on the first analysis result and the second analysis result. Finally, the remote analysis module receives the action data sent by the second monitoring module, compares and analyzes the features of the received action data with the features of different standard action data to obtain the action analysis result, and determines the final analysis result according to the speech analysis result and the action analysis result. Through the present invention, not only can the speech data of the trainees be identified simply and quickly while ensuring the accuracy of the identification, but also the security of the speech data transmitted from the local analysis module to the remote analysis module can be guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flowchart of a training behavior monitoring and analysis method of the present invention;

[0037] Figure 2 The present invention is a composition structure diagram of a training behavior monitoring and analysis device. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0039] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0040] The present invention provides Figure 1 The training behavior monitoring and analysis method shown is mainly implemented by executing the following steps:

[0041] S1. A first monitoring module and a second monitoring module are set at the training site. The first monitoring module collects and processes speech data of trainees, and the second monitoring module collects and processes motion data of trainees. The first monitoring module sends the collected speech data to a local analysis module set at the training site, and the second monitoring module transmits the collected motion data to a remote analysis module.

[0042] S2, the local analysis module compares and analyzes the features of the received speech data with the features of different pre-stored standard speech data to determine the transmission time, and the local analysis module transmits the received speech data to the remote analysis module starting from the transmission time, and the local analysis module also transmits the first analysis result of the received speech data to the remote analysis module;

[0043] S3, the remote analysis module compares and analyzes the features of the speech data sent by the local analysis module with the features of some pre-stored standard speech data to obtain a second analysis result, and the remote analysis module determines the speech analysis result based on the first analysis result and the second analysis result;

[0044] S4. The remote analysis module receives the action data sent by the second monitoring module, compares and analyzes the features of the received action data with the features of different pre-stored standard action data to obtain the action analysis results, and determines the final analysis results based on the discourse analysis results and the action analysis results.

[0045] Specifically, in order to realize automatic analysis of the training performance of trainees at the aircraft emergency response training site, S1 to S4 are proposed. In actual application scenarios, S1 to S4 are repeatedly executed multiple times, so as to analyze whether multiple statements and multiple corresponding actions of the trainees are standardized. The correspondence between statements and actions means that the statements and actions occur in the same time period. For ease of understanding, this embodiment takes the automatic analysis of a statement and corresponding action of the trainee as an example for explanation.

[0046] In S1, a first monitoring module and a second monitoring module are set up at the training site. The first monitoring module is used to collect and process the speech data of the trainees, and the second monitoring module is used to collect and process the action data of the trainees. The action data is, for example, a picture containing the action performed by the trainees. Then, the first monitoring module sends the collected speech data to the local analysis module, wherein the local analysis module is also set up at the training site, and the second monitoring module transmits the collected action data to the remote analysis module. It should be noted that the first monitoring module collects speech data while sending the speech data to the local analysis module, and the second monitoring module transmits the typical pictures, i.e., action data, collected in the corresponding time period to the remote analysis module before the trainees finish speaking a sentence. In S2, the local analysis module compares and analyzes the features of the received speech data with the features of different standard speech data at the same time, thereby determining the transmission time, and transmits the received speech data to the remote analysis module from the transmission time, and then transmits the first analysis result of the received speech data to the remote analysis module. The detailed process of S2 will be described below. In S3, the remote analysis module compares and analyzes the features of the speech data sent by the local analysis module with the features of some standard speech data to obtain a second analysis result, and also determines the speech analysis result based on the first analysis result and the second analysis result. The detailed process of S3 will be described below. In S4, the remote analysis module receives the action data sent by the second monitoring module, and compares and analyzes the features of the received action data with the features of different standard action data at the same time, and then obtains the action analysis result. That is to say, the remote analysis module compares the received picture with multiple standard pictures at the same time, wherein the action in the standard picture is the action that the trainee should perform during the time period of speaking the corresponding standard speech data. The remote analysis module considers the action in the corresponding standard picture with the highest consistency score as the action in the received picture. If the consistency score is higher than the consistency score threshold, a successful action analysis result is generated, that is, the trainee's action is considered to be standard. After that, the remote analysis module determines the final analysis result based on the speech analysis result and the action analysis result. The specific method is that the final analysis result of successful analysis is generated only when both the speech analysis result and the action analysis result are successfully analyzed.

[0047] Further, the standard discourse data includes a number of words, and the process of generating partial standard discourse data corresponding to the standard discourse data includes dividing the several words constituting the standard discourse data into two groups, front and back groups, so that different words in the latter group constitute the partial standard discourse data.

[0048] Specifically, standard discourse data is introduced here. Standard discourse data refers to discourse data that trainees should say during the training process. For example, a certain standard discourse data is "arms crossed in front of the chest". In addition, in order to obtain partial standard discourse data corresponding to the standard discourse data, several words that constitute the standard discourse data are first divided into two groups, front and back groups, and then different words in the latter group constitute partial standard discourse data. For example, "arms crossed" corresponds to different words in the former group, and "holding in front of the chest" corresponds to different words in the latter group, that is, "holding in front of the chest" is partial standard discourse data corresponding to "arms crossed in front of the chest".

[0049] Further, S2 includes the following steps:

[0050] S21, the local analysis module compares and analyzes the features of the standard speech data with the features of the received speech data for each standard speech data, calculates the difference degree value between the two, and determines whether there is a difference degree value that is changing from greater than the difference degree value threshold 1 to less than the difference degree value threshold 1. If not, repeat this step. If yes, take the moment of successful judgment as the transmission moment, start transmitting the received speech data to the remote analysis module from the transmission moment, and proceed to the next step;

[0051] S22, the local analysis module calculates the time interval from the transmission time to the current time, compares and analyzes the features of the received speech data with the features of the standard speech data corresponding to the difference degree value determined to exist, and calculates the difference degree value between the two. When the time interval does not reach the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, repeat this step. When the time interval does not reach the first time interval threshold and the difference degree value is less than or equal to the difference degree value threshold 2, generate a first analysis result of successful analysis, and transmit the first analysis result to the remote analysis module. When the time interval reaches the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, generate a first analysis result of failed analysis, and transmit the first analysis result to the remote analysis module.

[0052] Further, the difference degree value threshold 2 is less than the difference degree value threshold 1.

[0053] Specifically, the detailed process of S2 is introduced here. In S21, for each standard speech data, the local analysis module compares and analyzes the characteristics of the standard speech data with the characteristics of the received speech data, for example, compares and analyzes the fluctuation pattern corresponding to the standard speech data and the fluctuation pattern of the received speech data, and calculates the difference degree value between the two. In the time period when the first monitoring module just starts to send speech data to the local analysis module, S21 is executed multiple times, and the difference degree value is always large, but has a trend of gradually decreasing. The local analysis module always determines whether there is a difference degree value that is changing from greater than the difference degree value threshold value one to less than the difference degree value threshold value one, that is, it always determines that the difference degree value is equal to the difference degree value Whether the moment of threshold one has arrived, if so, the moment of successful judgment, that is, the moment when the difference degree value is equal to the difference degree value threshold one is regarded as the transmission moment. At this time, it is considered that the characteristics of the received speech data match the characteristics of different words in the previous group corresponding to the corresponding standard speech data, and the received speech data is transmitted to the remote analysis module from the transmission moment. It should be noted that because the local analysis module compares and analyzes the characteristics of the speech data sent by the first monitoring module with the characteristics of multiple standard speech data at the same time, the moment corresponding to the earliest judgment of the difference degree value is used as the transmission moment. If not, the local analysis module continues to receive the speech data sent by the first monitoring module and repeats S21. In S22, the local analysis module continues to receive the speech data sent by the first monitoring module, calculates the time interval from the transmission time to the current time, continues to compare and analyze the features of the received speech data with the features of the standard speech data corresponding to the difference degree value determined to exist, and calculates the difference degree value between the two. If the time interval does not reach the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, S22 is repeated. If the time interval does not reach the first time interval threshold and the difference degree value is less than or equal to the difference degree value threshold 2, it is considered that the features of the received speech data match the features of the corresponding standard speech data, and a first analysis result of successful analysis is generated, and the first analysis result is transmitted to the remote analysis module. If the time interval reaches the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, it is considered that the features of the received speech data do not match the features of the corresponding standard speech data, and a first analysis result of failed analysis is generated, and the first analysis result is transmitted to the remote analysis module. It should also be noted that the difference degree value threshold 2 is less than the difference degree value threshold 1. Through the above method, the local analysis module can quickly obtain the first analysis result of the speech data sent by the first monitoring module.

[0054] Furthermore, S3 includes the following steps:

[0055] S31, when the remote analysis module obtains the speech data sent by the local analysis module, it compares and analyzes the characteristics of the speech data with the characteristics of some pre-stored standard speech data, calculates the difference between the two, and calculates the time interval from the start time of the analysis process to the current time;

[0056] S32, when the time interval does not reach the second time interval threshold and the difference degree value is greater than the difference degree value threshold three, the remote analysis module repeatedly executes S31, when the time interval does not reach the second time interval threshold and the difference degree value is less than or equal to the difference degree value threshold three, the remote analysis module generates a second analysis result of successful analysis and ends the analysis process, when the time interval reaches the second time interval threshold and the difference degree value is greater than the difference degree value threshold three, the remote analysis module generates a second analysis result of failed analysis and ends the analysis process;

[0057] S33: When both the first analysis result and the second analysis result are successful, the remote analysis module sets the discourse analysis result as successful; otherwise, the remote analysis module sets the discourse analysis result as failed.

[0058] Specifically, the detailed process of S3 is introduced here. In S31, when the remote analysis module obtains the speech data sent by the local analysis module, it compares and analyzes the characteristics of the speech data with the characteristics of some pre-stored standard speech data. The partial standard speech data refers to the partial standard speech data corresponding to the standard speech data corresponding to the judged difference degree value, thereby calculating the difference degree value between the two. At the same time, the remote analysis module calculates the time interval from the start time of the analysis processing to the current time. In S32, if the time interval does not reach the second time interval threshold and the difference value is greater than the difference value threshold three, the remote analysis module repeats S31. If the time interval does not reach the second time interval threshold and the difference value is less than or equal to the difference value threshold three, the remote analysis module considers that the characteristics of the speech data sent by the local analysis module match the characteristics of part of the standard speech data, generates a second analysis result of successful analysis, and ends the analysis process. That is, when the speech data sent by the local analysis module is not received, the remote analysis module stops the analysis process from S31 to S33, which can reduce the processing burden. If the time interval reaches the second time interval threshold and the difference value is greater than the difference value threshold three, the remote analysis module considers that the characteristics of the speech data sent by the local analysis module do not match the characteristics of part of the standard speech data, generates a second analysis result of failed analysis, and ends the analysis process. In S33, if both the first analysis result and the second analysis result are successful, the remote analysis module sets the speech analysis result to success, that is, the speech spoken by the trainee during the training process is standard. Otherwise, the remote analysis module sets the speech analysis result to failure. Through the above method, the remote analysis module can quickly generate a second analysis result on the speech data sent by the local analysis module, and comprehensively consider the second analysis result and the first analysis result generated by the local analysis module to determine the speech analysis result, thereby ensuring the accuracy of the speech analysis result.

[0059] Furthermore, the local analysis module transmits the received speech data to the remote analysis module, including the following steps:

[0060] S211, the local analysis module sets the count value p to 1, and the local analysis module regards the received speech data as initial data, divides the initial data into first speech data and second speech data, performs a specific operation on the first speech data and the second speech data to obtain p-th intermediate data, and increases the count value p by 1;

[0061] S212, the local analysis module regards the first utterance data as initial data, continues to divide the initial data into first utterance data and second utterance data, performs a specific operation on the first utterance data and the second utterance data to obtain pth intermediate data, and the local analysis module determines whether the end condition is met. If not, the count value p is incremented by 1 and this step is repeated. If yes, proceed to the next step.

[0062] S213. The local analysis module performs secret processing on the last first utterance data to obtain secret data, and the local analysis module sends the secret data and the 1st to pth intermediate data to the remote analysis module.

[0063] Specifically, the detailed process of transmitting the received speech data from the local analysis module to the remote analysis module is introduced here. In S211, the local analysis module sets the count value p to 1, and regards the received speech data as the initial data, and divides the initial data into the first speech data and the second speech data. The data scale of the first speech data and the second speech data is the same. The first speech data and the second speech data are further operated to obtain the pth intermediate data. For the convenience of understanding the specific operation, for example, performing a specific operation on 0 and 1 can obtain 1, and performing a specific operation on 0 and 0, and 1 and 1 can obtain 0, and then the count value p is incremented by 1. In S212, the local analysis module regards the first speech data as the initial data, and continues to divide the initial data into the first speech data and the second speech data. The first speech data and the second speech data are further operated to obtain the pth intermediate data. The local analysis module then determines whether the end condition is met. Specifically, it can be determined whether p is greater than or equal to the preset threshold value P. If not, the count value p is further incremented by 1, and then S212 is repeatedly executed. If yes, S213 is continued. In S213, the local analysis module performs secret processing on the last first speech data to obtain secret data, that is, generates the first speech data of the pth intermediate data, where p is the same as P. The secret processing may be, for example, processing using an encryption algorithm in the prior art. The local analysis module also sends the secret data, the first intermediate data, the second intermediate data, and up to the pth intermediate data to the remote analysis module, where p is the same as P. Through the above method, even if the data sent by the local analysis module to the remote analysis module is illegally intercepted, the speech data sent by the local analysis module cannot be obtained, thereby ensuring the security of the speech data.

[0064] Furthermore, the remote analysis module obtains the speech data sent by the local analysis module through the following steps:

[0065] S311, the local analysis module recovers the secret data to obtain the last first utterance data, performs a specific operation on the last first utterance data and the pth intermediate data to obtain the last second utterance data, recovers the last first utterance data from the last first utterance data and the last second utterance data, and decrements the count value p by 1;

[0066] S312, the local analysis module performs a specific operation on the previous first utterance data and the pth intermediate data to obtain the previous second utterance data, recovers the new previous first utterance data from the previous first utterance data and the previous second utterance data, decrements the count value p by 1, and determines whether the count value p is equal to 1. If not, repeat this step, and if yes, proceed to the next step;

[0067] S313: The local analysis module performs a specific operation on the previous first utterance data and the first intermediate data to obtain the previous second utterance data, and restores the initial data from the previous first utterance data and the previous second utterance data.

[0068] Specifically, here is introduced how the remote analysis module obtains the speech data sent by the local analysis module. In S311, the local analysis module recovers the secret data to obtain the last first speech data, that is, to generate the first speech data of the pth intermediate data, where p and P are the same, and performs a specific operation on the last first speech data and the pth intermediate data to obtain the last second speech data, and also recovers the previous first speech data from the last first speech data and the last second speech data, specifically by connecting the last first speech data and the last second speech data, and then decrements the count value p by 1. In S312, the local analysis module performs a specific operation on the previous first speech data and the pth intermediate data to obtain the previous second speech data, and recovers the new previous first speech data from the previous first speech data and the previous second speech data, where the new previous first speech data actually refers to the first speech data before the current previous first speech data, and then decrements the count value p by 1, and determines whether the count value p is equal to 1. If not, S312 is repeated, and if yes, S313 is continued. In S313, the local analysis module performs a specific operation on the last first utterance data and the first intermediate data. The last first utterance data actually refers to the last first utterance data finally obtained after executing S312 multiple times, to obtain the last second utterance data, and restore the initial data from the last first utterance data and the last second utterance data. Through the above method, the remote analysis module can restore the utterance data from the data sent by the local analysis module.

[0069] According to another aspect of the embodiment of the present invention, Figure 2 As shown, a training behavior monitoring and analysis device is also provided, including a first monitoring module, a second monitoring module, a local analysis module, and a remote analysis module, to implement a training behavior monitoring and analysis method described above, wherein the functions of each module are as follows:

[0070] A first monitoring module is used to be set up at the training site, collect and process the speech data of the trainees, and send the collected speech data to a local analysis module set up at the training site;

[0071] The second monitoring module is used to be set up at the training site, collect and process the motion data of the trainees, and transmit the collected motion data to the remote analysis module;

[0072] A local analysis module, configured to compare and analyze the features of the received speech data with the features of different pre-stored standard speech data to determine a transmission time, and to transmit the received speech data to the remote analysis module starting from the transmission time, and to transmit a first analysis result of the received speech data to the remote analysis module;

[0073] The remote analysis module is used to compare and analyze the characteristics of the speech data sent by the local analysis module with the characteristics of some pre-stored standard speech data to obtain a second analysis result, determine the speech analysis result based on the first analysis result and the second analysis result, and receive the action data sent by the second monitoring module, compare and analyze the characteristics of the received action data with the characteristics of different pre-stored standard action data to obtain the action analysis result, and determine the final analysis result based on the speech analysis result and the action analysis result.

[0074] According to another aspect of an embodiment of the present invention, there is further provided a device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to implement any one of the above methods when executing the computer program.

[0075] According to another aspect of an embodiment of the present invention, a storage medium is provided, wherein the storage medium stores program instructions, wherein when the program instructions are executed, a device where the storage medium is located is controlled to execute any one of the above methods.

[0076] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0077] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the 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 can 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. As an illustration and not limitation, RAM is available in various 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).

[0078] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0079] The above embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

[0080] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A training behavior monitoring and analysis method, characterized in that: The steps include: S1. A first monitoring module and a second monitoring module are set at a training site, wherein the first monitoring module collects and processes speech data of trainees, and the second monitoring module collects and processes motion data of trainees, and the first monitoring module sends the collected speech data to a local analysis module set at the training site, and the second monitoring module transmits the collected motion data to a remote analysis module; S2, the local analysis module compares and analyzes the features of the received speech data with the features of different pre-stored standard speech data to determine the transmission time, and the local analysis module transmits the received speech data to the remote analysis module starting from the transmission time, and the local analysis module also transmits the first analysis result of the received speech data to the remote analysis module; S3, the remote analysis module compares and analyzes the features of the speech data sent by the local analysis module with the features of some pre-stored standard speech data to obtain a second analysis result, and the remote analysis module determines a speech analysis result based on the first analysis result and the second analysis result; S4. The remote analysis module receives the action data sent by the second monitoring module, compares and analyzes the features of the received action data with the features of different pre-stored standard action data to obtain an action analysis result, and determines the final analysis result based on the speech analysis result and the action analysis result.

2. The method according to claim 1, characterized in that The standard discourse data includes a number of words, and the process of generating partial standard discourse data corresponding to the standard discourse data includes dividing the several words constituting the standard discourse data into two groups, front and back groups, so that different words in the latter group constitute the partial standard discourse data.

3. The method according to claim 2, characterized in that The S2 comprises the following steps: S21, the local analysis module compares and analyzes the features of the standard speech data with the features of the received speech data for each standard speech data, calculates the difference degree value between the two, and the local analysis module determines whether there is a difference degree value that is changing from greater than the difference degree value threshold 1 to less than the difference degree value threshold 1. If not, repeat this step. If yes, take the moment of successful judgment as the transmission moment, start transmitting the received speech data to the remote analysis module from the transmission moment, and proceed to the next step; S22, the local analysis module calculates the time interval from the transmission moment to the current moment, compares and analyzes the features of the received speech data with the features of the standard speech data corresponding to the difference degree value determined to exist, and calculates the difference degree value between the two. When the time interval does not reach the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, repeat this step. When the time interval does not reach the first time interval threshold and the difference degree value is less than or equal to the difference degree value threshold 2, generate the first analysis result of successful analysis, and transmit the first analysis result to the remote analysis module. When the time interval reaches the first time interval threshold and the difference degree value is greater than the difference degree value threshold 2, generate the first analysis result of failed analysis, and transmit the first analysis result to the remote analysis module.

4. The method according to claim 3, characterized in that The difference degree value threshold 2 is less than the difference degree value threshold 1.

5. The method according to claim 3, characterized in that: The S3 comprises the following steps: S31, when the remote analysis module obtains the speech data sent by the local analysis module, it compares and analyzes the characteristics of the speech data with the characteristics of some pre-stored standard speech data, calculates the difference between the two, and calculates the time interval from the start time of the analysis process to the current time; S32, when the time interval does not reach the second time interval threshold and the difference degree value is greater than the difference degree value threshold three, the remote analysis module repeatedly executes S31, when the time interval does not reach the second time interval threshold and the difference degree value is less than or equal to the difference degree value threshold three, the remote analysis module generates the second analysis result of successful analysis and ends the analysis process, when the time interval reaches the second time interval threshold and the difference degree value is greater than the difference degree value threshold three, the remote analysis module generates the second analysis result of failed analysis and ends the analysis process; S33: When both the first analysis result and the second analysis result are successful, the remote analysis module sets the discourse analysis result as successful; otherwise, the remote analysis module sets the discourse analysis result as failed.

6. The method according to claim 5, characterized in that The local analysis module transmits the received speech data to the remote analysis module, comprising the following steps: S211, the local analysis module sets the count value p to 1, and the local analysis module regards the received speech data as initial data, divides the initial data into first speech data and second speech data, performs a specific operation on the first speech data and the second speech data to obtain p-th intermediate data, and increases the count value p by 1; S212, the local analysis module regards the first utterance data as initial data, continues to divide the initial data into first utterance data and second utterance data, performs a specific operation on the first utterance data and the second utterance data to obtain pth intermediate data, and the local analysis module determines whether the end condition is met, if not, increments the count value p by 1, repeats this step, if yes, proceeds to the next step; S213: The local analysis module performs secret processing on the last first speech data to obtain secret data, and the local analysis module sends the secret data and the 1st to pth intermediate data to the remote analysis module.

7. The method according to claim 6, characterized in that The remote analysis module obtains the speech data sent by the local analysis module through the following steps: S311, the local analysis module recovers the secret data to obtain the last first utterance data, performs a specific operation on the last first utterance data and the pth intermediate data to obtain the last second utterance data, recovers the last first utterance data from the last first utterance data and the last second utterance data, and decrements the count value p by 1; S312, the local analysis module performs a specific operation on the previous first utterance data and the pth intermediate data to obtain the previous second utterance data, recovers the new previous first utterance data from the previous first utterance data and the previous second utterance data, decrements the count value p by 1, and determines whether the count value p is equal to 1. If not, repeat this step, and if yes, proceed to the next step; S313: The local analysis module performs a specific operation on the previous first utterance data and the first intermediate data to obtain the previous second utterance data, and restores the initial data from the previous first utterance data and the previous second utterance data.

8. A training behavior monitoring and analysis device, used to implement the method according to any one of claims 1 to 7, characterized in that: Includes the following modules: A first monitoring module is used to be set up at the training site, collect and process the speech data of the trainees, and send the collected speech data to a local analysis module set up at the training site; The second monitoring module is used to be set up at the training site, collect and process the motion data of the trainees, and transmit the collected motion data to the remote analysis module; A local analysis module, configured to compare and analyze the features of the received speech data with the features of different pre-stored standard speech data to determine a transmission time, and to transmit the received speech data to the remote analysis module starting from the transmission time, and to transmit a first analysis result of the received speech data to the remote analysis module; The remote analysis module is used to compare and analyze the characteristics of the speech data sent by the local analysis module with the characteristics of some pre-stored standard speech data to obtain a second analysis result, determine the speech analysis result based on the first analysis result and the second analysis result, and receive the action data sent by the second monitoring module, compare and analyze the characteristics of the received action data with the characteristics of different pre-stored standard action data to obtain the action analysis result, and determine the final analysis result based on the speech analysis result and the action analysis result.

9. A device, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to implement the method according to any one of claims 1 to 7 when executing the computer program.

10. A storage medium, characterized in that: The storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

Images