Instruction conversion control method and system based on voice recognition

Abstract

The present application relates to the technical field of instruction conversion control, and particularly relates to an instruction conversion control method and system based on voice recognition, which analyzes the operation threat degree of a to-be-executed voice instruction to a current running operation, analyzes the content analysis abnormal state of the to-be-executed voice instruction, constructs an operation intention risk assessment model, imports the operation threat degree analysis result and the content analysis abnormal state analysis result of the to-be-executed voice instruction into the operation intention risk assessment model, assesses the operation intention risk of the to-be-executed voice instruction, and converts the to-be-executed voice instruction according to the operation intention risk assessment result, so as to realize safe prediction and intelligent decision of instruction conversion control, improve the safety and stability of system operation, and reduce the fault risk in the instruction conversion process.

Description

Technical Field

[0001] This invention relates to the field of command conversion and control technology, and in particular to a command conversion and control method and system based on speech recognition. Background Technology

[0002] In the process of intelligent system interaction and automated control, command conversion control driven by voice recognition technology is a crucial link in achieving efficient human-machine collaboration and intelligent equipment response. As the core carrier of human-machine interaction, the accuracy of voice command parsing, operational safety, and reliability of execution directly affect the safety and stability of the intelligent system. For example, in intelligent production line control, issues with command parsing accuracy may trigger the execution of unauthorized commands, or production line interruptions may occur due to a lack of assessment of the threat posed by command execution to the current production process. Similarly, in intelligent traffic control, incomplete command parsing or failure to consider execution threats can lead to scheduling errors. Traditional voice command control methods often focus on the accuracy of command recognition, simply converting the recognition results directly into control operations, lacking in-depth analysis of voice commands and a comprehensive assessment of the impact of command execution on current operations. Therefore, traditional voice command control methods struggle to address the risk control requirements of command execution in complex scenarios and cannot guarantee the reliable operation of the system in human-machine interaction and automated processes.

[0003] Meanwhile, existing technologies rely solely on accurate text matching of voice command content to determine whether to execute control commands, neglecting quantitative analysis of the semantic integrity and compliance of voice commands. They also fail to consider the various conflicting factors between the command to be executed and the current operation, which could lead to device interruptions. Furthermore, existing technologies do not incorporate command parsing anomalies and operational threats into the risk assessment of operational intent during command conversion, resulting in overly simplistic risk identification and blind command execution during command conversion control.

[0004] To address these issues, this application presents a command conversion control method and system based on speech recognition. Summary of the Invention

[0005] To overcome the shortcomings and deficiencies of existing technologies, this invention provides a speech recognition-based command conversion control method and system. The method involves comprehensively analyzing the operational threat level and content parsing anomalies posed by the currently running operation in relation to the voice command to be executed; constructing an operation intent risk assessment model to evaluate the operational intent risk of the voice command to be executed; and converting the voice command to be executed based on the operation intent risk assessment results. This improves the security and stability of system operation and reduces the risk of failures during command conversion.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, embodiments of the present invention provide a command conversion control method based on speech recognition, comprising the following steps:

[0008] S1. Obtain the voice command data to be executed, historical accident data, and current operation data;

[0009] S2. Import the voice command data to be executed and the current operation data into the operation threat level analysis model to analyze the degree of operation threat posed by the voice command to be executed to the currently running operation.

[0010] S3. Import the voice command data to be executed and historical accident data into the command content parsing abnormal state analysis model, and analyze the abnormal state of the content parsing of the voice command to be executed.

[0011] S4. Construct an operation intent risk assessment model. Import the analysis results of the degree of operation threat caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing into the operation intent risk assessment model to assess the operation intent risk of the voice command to be executed.

[0012] S5. Based on the risk assessment results of the operational intent, convert the voice command to be executed.

[0013] In a preferred embodiment of the present invention, step S2, which analyzes the degree of operational threat posed by the voice command to be executed to the currently running operation, includes the following specific steps:

[0014] S21. Extract the voice command data to be executed and the current running operation data;

[0015] S22. Based on the voice command data to be executed and the current operation data, construct an operation threat level analysis model, analyze the operation threat level caused by the voice command to be executed to the current operation, and obtain the operation threat level analysis results caused by the voice command to be executed to the current operation.

[0016] The formula for calculating the operational threat level is as follows:

[0017] ;

[0018] In the formula, Cz represents the degree of operational threat posed by the voice command to be executed to the currently running operation, ct represents the analysis result of the degree of compatibility conflict threat posed by the voice command to be executed to the currently running operation, and yc represents the analysis result of the degree of abnormal urgency threat posed by the voice command to be executed to the currently running operation.

[0019] In a preferred embodiment of the present invention, the process of constructing the operational threat level analysis model in step S22 includes the following specific steps:

[0020] S221. Based on the voice command data to be executed and the current operation data, analyze the degree of compatibility conflict threat caused by the voice command to be executed to the current operation, and obtain the analysis results of the degree of compatibility conflict threat caused by the voice command to be executed to the current operation.

[0021] S222. Based on the voice command data to be executed and the current operation data, analyze the degree of urgency threat posed by the voice command to be executed to the current operation, and obtain the analysis results of the degree of urgency threat posed by the voice command to be executed to the current operation.

[0022] In a preferred embodiment of the present invention, step S3, which involves analyzing the abnormal state of the parsing of the voice command to be executed, includes the following specific steps:

[0023] S31. Extract the voice command data to be executed and historical accident data;

[0024] S32. Based on the voice command data to be executed and historical accident data, construct an abnormal state analysis model for command content parsing, analyze the abnormal state of the content parsing of the voice command to be executed, and obtain the analysis result of the abnormal state of the content parsing of the voice command to be executed.

[0025] The formula for calculating content parsing anomalies is as follows:

[0026] ;

[0027] In the formula, Nr represents the abnormal state of the content parsing of the voice command to be executed, wz represents the result of the analysis of the abnormal state of the semantic integrity of the voice command to be executed, and hg represents the result of the analysis of the abnormal state of the semantic compliance of the voice command to be executed.

[0028] In a preferred embodiment of the present invention, the process of constructing the abnormal state analysis model for instruction content parsing in step S32 includes the following specific steps:

[0029] S321. Based on the voice command data to be executed and historical accident data, analyze the semantic integrity anomaly state of the voice command to be executed, and obtain the analysis result of the semantic integrity anomaly state of the voice command to be executed.

[0030] S322. Based on the voice command data to be executed and historical accident data, analyze the semantic compliance anomaly state of the voice command to be executed, and obtain the analysis results of the semantic compliance anomaly state of the voice command to be executed.

[0031] In a preferred embodiment of the present invention, step S4, which involves constructing an operational intent risk assessment model, includes the following specific steps:

[0032] S41. Extract the analysis results of the operational threat level caused by the voice command to be executed to the currently running operation, and the analysis results of the abnormal state of the content parsing of the voice command to be executed;

[0033] S42. Based on the analysis results of the degree of operational threat caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing, assess the operational intent risk of the voice command to be executed;

[0034] The formula for assessing operational intent risk is as follows:

[0035] ;

[0036] In the formula, ZY represents the operational intent risk of the voice command to be executed, Cz represents the operational threat level analysis result caused by the voice command to be executed to the currently running operation, and Nr represents the abnormal state analysis result of the content parsing of the voice command to be executed. These are the weights for the impact of operational threats and the weights for the impact of content parsing anomalies.

[0037] In a preferred embodiment of the present invention, step S5 involves converting the voice command to be executed based on the risk assessment result of the operational intent, including the following specific steps:

[0038] S51. Obtain the risk assessment results of the operational intent of the voice command to be executed;

[0039] S52. A preset operation intent risk threshold is set. When the operation intent risk assessment result of the voice command to be executed is greater than or equal to the operation intent risk threshold, the voice command to be executed is rejected. When the operation intent risk assessment result of the voice command to be executed is less than the operation intent risk threshold, the voice command to be executed is converted.

[0040] Secondly, embodiments of the present invention also provide a command conversion control system based on speech recognition, comprising:

[0041] The data acquisition module is used to acquire data of voice commands to be executed, historical accident data, and current operation data;

[0042] The operation threat level analysis module is used to import the voice command data to be executed and the current operation data into the operation threat level analysis model to analyze the operation threat level caused by the voice command to be executed to the currently running operation.

[0043] The instruction content parsing anomaly analysis module is used to import the voice instruction data to be executed and historical accident data into the instruction content parsing anomaly analysis model to analyze the anomaly status of the content parsing of the voice instruction to be executed.

[0044] The Operation Intent Risk Assessment Module is used to build an Operation Intent Risk Assessment Model. It imports the analysis results of the operational threat level caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing into the Operation Intent Risk Assessment Model to assess the operational intent risk of the voice command to be executed.

[0045] The instruction conversion control module is used to convert the voice instructions to be executed based on the risk assessment results of the operation intent.

[0046] The control module is used to control the operation of the data acquisition module, the operation threat level analysis module, the instruction content parsing abnormal state analysis module, the operation intent risk assessment module, and the instruction conversion control module.

[0047] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0048] This invention analyzes the operational threat level posed by the voice command to be executed to the currently running operation; analyzes the abnormal state of the content parsing of the voice command to be executed; constructs an operation intent risk assessment model, imports the operational threat level analysis results and the content parsing abnormal state analysis results into the operation intent risk assessment model, and assesses the operation intent risk of the voice command to be executed; based on the operation intent risk assessment results, the voice command to be executed is converted. This enables safe prediction and intelligent decision-making in command conversion control, improves the security and stability of system operation, and reduces the risk of failure during command conversion. Attached Figure Description

[0049] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0050] Figure 1 This is a schematic diagram of the overall process of the speech recognition-based instruction conversion control method of the present invention;

[0051] Figure 2 This is a flowchart of step S2 in the speech recognition-based instruction conversion control method of the present invention;

[0052] Figure 3 This is a flowchart of step S3 in the speech recognition-based instruction conversion control method of the present invention;

[0053] Figure 4This is a schematic diagram of the command conversion control system based on speech recognition of the present invention. Detailed Implementation

[0054] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0055] Example 1

[0056] like Figure 1 As shown, this embodiment provides a command conversion control method based on speech recognition, which specifically includes the following steps:

[0057] S1. Obtain the voice command data to be executed, historical accident data, and current operation data;

[0058] S2. Import the voice command data to be executed and the current operation data into the operation threat level analysis model to analyze the degree of operation threat posed by the voice command to be executed to the currently running operation.

[0059] S3. Import the voice command data to be executed and historical accident data into the command content parsing abnormal state analysis model, and analyze the abnormal state of the content parsing of the voice command to be executed.

[0060] S4. Construct an operation intent risk assessment model. Import the analysis results of the degree of operation threat caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing into the operation intent risk assessment model to assess the operation intent risk of the voice command to be executed.

[0061] S5. Based on the risk assessment results of the operational intent, convert the voice command to be executed.

[0062] In this embodiment, as Figure 2 As shown, step S2 analyzes the operational threat level posed by the voice command to be executed to the currently running operation, including the following specific steps:

[0063] S21. Extract the voice command data to be executed and the current running operation data;

[0064] S22. Based on the voice command data to be executed and the current operation data, construct an operation threat level analysis model, analyze the operation threat level caused by the voice command to be executed to the current operation, and obtain the operation threat level analysis results caused by the voice command to be executed to the current operation.

[0065] The formula for calculating the operational threat level is as follows:

[0066] ;

[0067] In the formula, Cz represents the degree of operational threat posed by the voice command to be executed to the currently running operation, ct represents the analysis result of the degree of compatibility conflict threat posed by the voice command to be executed to the currently running operation, and yc represents the analysis result of the degree of abnormal urgency threat posed by the voice command to be executed to the currently running operation.

[0068] For example, in this embodiment, the sum of the compatibility conflict threat level analysis results and the abnormal urgency threat level analysis results is used to reflect the total contribution of the two types of threats to the operational threat level caused by the voice command to be executed to the currently running operation, thereby reflecting the basic superposition of operational threat levels; the denominator middle, This is used to characterize the difference between the level of compatibility conflict threat and the level of abnormal urgency threat. The greater the difference, the larger the denominator, which can suppress the unreasonable amplification of the operational threat level caused by the extreme difference in the occurrence of the two types of threats; adding 2 avoids the occurrence of a denominator of 0. Further, in this embodiment... The value range of Cz is constrained to the interval between -1 and 1. After mapping through an exponential function, the superposition trend of the two types of threats is preserved, while avoiding the excessive dominance of a single threat in the final operational threat level caused by simple linear calculation. Furthermore, subtracting 1 ensures the alignment of the operational threat level benchmark; that is, when there are no compatible conflict threats or abnormally urgent threats, the value of Cz is 0. Specifically, this embodiment comprehensively calculates compatible conflict threats and abnormally urgent threats, thus ensuring the superposition effect of the two types of threats while also controlling the denominator. The system dynamically constrains the extreme interference caused by the differences between the two types of threats. By introducing an exponential function, the calculation formula provided in this embodiment can meet the actual evolutionary law of threat level from zero to a gradual accumulation, and can accurately reflect the comprehensive operational threat level dominated by both compatible conflict threats and unusually urgent threats.

[0069] In this embodiment, the process of constructing the operational threat level analysis model in step S22 includes the following specific steps:

[0070] S221. Based on the voice command data to be executed and the current operation data, analyze the degree of compatibility conflict threat caused by the voice command to be executed to the current operation, and obtain the analysis results of the degree of compatibility conflict threat caused by the voice command to be executed to the current operation.

[0071] The formula for calculating the compatibility conflict threat level is as follows:

[0072] ;

[0073] In the formula, ct represents the degree of compatibility conflict threat caused by the voice command to be executed to the currently running operation, vi represents the sequence of operation action feature vectors parsed from the voice command to be executed data, and vo represents the sequence of operation action feature vectors of the currently running operation in the currently running operation data. Let vi and vo be the squared Euclidean distance. Let be the covariance of vi and vo. The formulas for calculating the squared Euclidean distance and the covariance are existing technologies and will not be elaborated here.

[0074] For example, this embodiment analyzes the degree of compatibility conflict threat based on differences in operational actions and synergy. Using the feature vector sequences of the voice command to be executed and the currently running operation as the core, the difference in operational actions between the two is measured by calculating the squared Euclidean distance; the greater the difference, the higher the compatibility conflict potential. Covariance is also considered. This reflects the synergy between the feature vector sequences of the two operations. Low synergy amplifies the threat of compatibility conflicts. Furthermore, this embodiment utilizes an exponential function and reciprocal operation to linearly calculate the differences and synergies between the voice command to be executed and the currently running operation. The result of this linear calculation is then transformed into a value with a specific distribution pattern using an exponential function, thereby compressing the impact of action differences and synergies to a reasonable range. This ensures that the final compatibility conflict threat level smoothly and reasonably reflects the conflict relationship between the two, while enhancing the sensitivity of the compatibility conflict threat level calculation formula provided in this embodiment to changes in differences and synergies, highlighting the distinguishability of conflict levels under different action feature vector sequences. Specifically, in this embodiment, the acquisition of the action feature vector sequences of the voice command to be executed and the currently running operation involves semantically parsing the voice command, extracting key elements of the operation, such as action type, execution order, and target, and encoding them into feature vectors containing action type, target, and parameters. The action feature vector sequence of the currently running operation can also be obtained by extracting similar feature codes from the real-time monitoring data of the currently running operation.

[0075] S222. Based on the voice command data to be executed and the current operation data, analyze the degree of urgency threat posed by the voice command to be executed to the current operation, and obtain the analysis results of the degree of urgency threat posed by the voice command to be executed to the current operation.

[0076] The formula for calculating the degree of abnormal urgency threat is as follows:

[0077] ;

[0078] In the formula, yc represents the degree of urgency threat posed by the voice command to be executed to the currently running operation, Ci represents the number of operation steps parsed from the voice command to be executed in the voice command data, Ti represents the required execution time limit of the voice command to be executed in the voice command data, To represents the remaining execution time of the currently running operation in the current running operation data, and Ro represents the system resource utilization rate of the currently running operation in the current running operation data.

[0079] For example, this embodiment analyzes the degree of abnormal urgency threat based on the urgency of voice commands in terms of time and resources. Specifically, this embodiment uses... The impact of executing voice commands on the time resources of currently running operations was analyzed. The more steps and the greater the time difference in the operation, the more significant the impact on the time resources of the currently running operations. This is used to reflect the resource adaptability of the current operation within the required execution time limit of the pending voice command. The lower the system resource utilization rate of the current operation, the stronger its resource adaptability within the required execution time limit of the pending voice command, which can alleviate the urgency of resource utilization. Specifically, this embodiment also uses an exponential function to perform a nonlinear transformation on the calculation results of factors such as the number of operation steps and time difference. The time resource impact reflected by the number of operation steps, the required execution time limit, and the remaining execution time is transformed into a value that conforms to a specific decay law after being processed by the exponential function. This ensures that the calculation result of the abnormal urgency threat level can remain within a reasonable range, highlighting the characteristic of the abnormal urgency threat rising rapidly and then stabilizing when the time difference is large, which is more in line with the evolution law of abnormal urgency threat in the actual command conversion control process. Furthermore, in this embodiment, the number of operation steps is obtained by parsing the semantics of the voice command and decomposing the execution process according to the command semantics; the required execution time limit is extracted from the time requirement expression of the voice command and converted into a numerical value.

[0080] In this embodiment, as Figure 3 As shown, step S3 involves analyzing the abnormal state of the parsing of the voice command to be executed, including the following specific steps:

[0081] S31. Extract the voice command data to be executed and historical accident data;

[0082] S32. Based on the voice command data to be executed and historical accident data, construct an abnormal state analysis model for command content parsing, analyze the abnormal state of the content parsing of the voice command to be executed, and obtain the analysis result of the abnormal state of the content parsing of the voice command to be executed.

[0083] The formula for calculating content parsing anomalies is as follows:

[0084] ;

[0085] In the formula, Nr represents the abnormal state of the content parsing of the voice command to be executed, wz represents the result of the analysis of the abnormal state of the semantic integrity of the voice command to be executed, and hg represents the result of the analysis of the abnormal state of the semantic compliance of the voice command to be executed.

[0086] For example, this embodiment uses the logistic function as the carrier for calculating the abnormal state of content parsing, which can compress the analysis result of any input abnormal state to a value range of 0 to 1, thus adapting to the quantitative requirements of this embodiment for the abnormal state of content parsing. Specifically, this embodiment linearly accumulates the abnormal states of semantic integrity and compliance, which can reflect the changes in the abnormal state of content parsing under the combined influence of the two types of abnormal states; and through the exponential function, the linear accumulation result is transformed into non-linear growth, which can reflect that the higher the degree of the two types of abnormalities, the greater the damage to the content parsing result of the voice command to be executed. At the same time, this embodiment also... Furthermore, the calculation results of abnormal content parsing states are constrained within a reasonable value range of 0 to 1, so that the calculation results can intuitively reflect the gradual process of abnormal states from non-existent to present and from low to high. This achieves the scientific quantification of abnormal states in voice command content parsing, and can effectively reflect the comprehensive state of abnormal risks faced by the content parsing of the voice command to be executed under the dual constraints of semantic integrity and compliance. This provides a reasonable and scientific quantitative basis for subsequent risk assessment of command operation intent and command conversion.

[0087] In this embodiment, the construction process of the instruction content parsing abnormal state analysis model in step S32 includes the following specific steps:

[0088] S321. Based on the voice command data to be executed and historical accident data, analyze the semantic integrity anomaly state of the voice command to be executed, and obtain the analysis result of the semantic integrity anomaly state of the voice command to be executed.

[0089] The formula for calculating semantic integrity anomalies is as follows:

[0090] ;

[0091] In the formula, wz represents the semantic integrity anomaly state of the voice command to be executed, Vs represents the semantic feature vector of the voice command to be executed in the voice command data, Vst represents the semantic feature vector of a semantically complete voice command of the same type as the voice command to be executed in the historical accident data, fmi represents the missing i-th type semantic feature vector in a semantically incomplete voice command of the same type as the voice command to be executed in the historical accident data, k represents the number of missing semantic feature vectors in a semantically incomplete voice command of the same type as the voice command to be executed in the historical accident data, E(Vs,Vst) represents the edit distance between Vs and Vst, Len(Vst) represents the vector length of Vst, and I(Vs,fmi) represents the indicator function, where I(Vs,fmi)=1 when Vs is missing a missing i-th type semantic feature vector in a semantically incomplete voice command of the same type as the voice command to be executed, and I(Vs,fmi)=0 otherwise; the formula for calculating the edit distance is existing technology and will not be elaborated further.

[0092] For example, this embodiment analyzes the semantic integrity anomaly state of the voice command to be executed through semantic differences and historical missing features. Specifically, this embodiment uses semantic feature vectors as the basis for calculation, measures the overall semantic difference between the voice command to be executed and semantically complete voice commands of the same type, by measuring the edit distance between their semantic feature vectors, and normalizes the overall semantic difference by measuring the length of the semantic feature vector of the semantically complete voice commands of the same type. It quantifies the impact of semantic missing features in the semantic feature vector of the voice command to be executed on command execution by using historical missing semantic feature vectors and an indicator function, and processes the operation of historical missing semantic feature vectors using an exponential function. By transforming the semantic missing information of instructions using an exponential function, the impact of semantic missing information on the semantic integrity of instructions is amplified. The discrete judgment of semantic missing information is transformed into a continuous numerical influence, so that the abnormal state of semantic integrity can be comprehensively quantified by the nonlinear influence of overall semantic differences and semantic missing information on the abnormal state of semantic integrity, thus reflecting the semantic integrity of instructions more accurately. Furthermore, this embodiment also quantifies the impact of overall semantic differences and semantic missing information on the abnormal state of semantic integrity through multiplication operations. The semantic feature vector of the voice instruction to be executed is constructed by semantically encoding the voice instruction to be executed and extracting key semantic elements, such as operation verbs, objects, and necessary parameters. The semantic feature vector of semantically complete voice instructions of the same type as the voice instruction to be executed is obtained by selecting semantically complete similar instructions from historical accident data and performing the same encoding. The missing semantic feature vector is formed by statistically encoding the high-frequency missing semantic elements from similar instructions with incomplete semantics that caused accidents in the past. The classification of similar voice instructions is to classify historical instructions with similar functions and contexts into similar voice instructions based on the operation content and domain involved in the semantics of the voice instruction to be executed.

[0093] S322. Based on the voice command data to be executed and historical accident data, analyze the semantic compliance anomaly state of the voice command to be executed, and obtain the analysis results of the semantic compliance anomaly state of the voice command to be executed.

[0094] The formula for calculating semantic compliance anomalies is as follows:

[0095] ;

[0096] In the formula, hg represents the semantic compliance abnormal state of the voice command to be executed, nt represents the total number of semantic units in the voice command data to be executed, n represents the number of non-compliant semantic units in the voice command data to be executed, Tn represents the average execution time of semantically compliant voice commands of the same type as the voice command to be executed in the historical incident data, and Tv represents the average incident processing time of the historical incident data for incidents caused by semantically non-compliant voice commands of the same type as the voice command to be executed.

[0097] For example, this embodiment amplifies the impact of semantically non-compliant voice commands causing incidents on the analysis of semantic compliance anomalies by performing a logarithmic operation on the ratio of the average processing time of violation incidents to the average execution time of compliant commands. This transforms the potentially large variation in the time ratio into a relatively gradual change, avoiding imbalances in calculation results due to an excessively large time ratio. Furthermore, this embodiment integrates the impact of semantic compliance with the impact of incidents, thereby quantifying the actual harm caused by semantic violations to command conversion and execution. Specifically, semantic units are obtained by segmenting voice commands and parsing semantic elements; the number of non-compliant semantic units is obtained by comparing the semantic units of semantic commands that are semantically non-compliant and similar to the voice command to be executed in historical incident data with the semantic unit database of historical compliant voice commands extracted from similar compliant commands, and counting the number of semantic units that do not conform to the rules; the average execution time is obtained by filtering similar semantically compliant voice commands from historical incident data and calculating the average execution time; the average incident processing time is obtained by statistically analyzing similar semantically non-compliant voice commands that caused incidents and calculating the average time consumed from the anomaly to recovery during the incident processing.

[0098] In this embodiment, step S4, which involves constructing an operational intent risk assessment model, includes the following specific steps:

[0099] S41. Extract the analysis results of the operational threat level caused by the voice command to be executed to the currently running operation, and the analysis results of the abnormal state of the content parsing of the voice command to be executed;

[0100] S42. Based on the analysis results of the degree of operational threat caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing, assess the operational intent risk of the voice command to be executed;

[0101] The formula for assessing operational intent risk is as follows:

[0102] ;

[0103] In the formula, ZY represents the operational intent risk of the voice command to be executed, Cz represents the operational threat level analysis result caused by the voice command to be executed to the currently running operation, and Nr represents the abnormal state analysis result of the content parsing of the voice command to be executed. These are the weights for the impact of operational threats and the weights for the impact of content parsing anomalies.

[0104] For example, in the voice command operation intent risk assessment of this embodiment, the degree of operation threat reflects the potential interference of command execution on the current operation, and the abnormal state of content parsing reflects the reliability of the semantic understanding of the command itself. Both contribute to the operation intent risk from two dimensions: the impact of command execution and the quality of the command. Linear weighting allows for a clear decomposition and quantification of the contribution of different dimensions to the operation intent risk. By assigning different weights, the differences in attention to operation threat and content abnormality during voice command conversion can be flexibly adapted. When more attention is paid to the impact of command execution on existing operations during voice command conversion, the weight of the operation threat can be appropriately increased; conversely, if more emphasis is placed on the parsing quality of the command itself during voice command conversion, the weight of the content parsing abnormality can be appropriately increased.

[0105] In this embodiment, step S5 involves converting the voice command to be executed based on the risk assessment result of the operation intent, including the following specific steps:

[0106] S51. Obtain the risk assessment results of the operational intent of the voice command to be executed;

[0107] S52. A preset operation intent risk threshold is established. When the operation intent risk assessment result of the voice command to be executed is greater than or equal to the operation intent risk threshold, the voice command to be executed is rejected from conversion; when the operation intent risk assessment result of the voice command to be executed is less than the operation intent risk threshold, the voice command to be executed is converted. The parameters (e.g., weights and thresholds) in this embodiment are obtained experimentally by those skilled in the art. The specific experimental method is as follows: acquire historical data of multiple voice commands to be executed, historical accident data, and corresponding current operation data; substitute the voice command data to be executed, historical accident data, and corresponding current operation data into each step of this embodiment to obtain the operation intent risk assessment results of multiple historical voice commands to be executed; acquire the judgment results of whether there is a risk in the conversion of multiple historical voice commands to be executed; import the operation intent risk assessment results of multiple historical voice commands to be executed and the judgment results of whether there is a risk in the conversion of multiple historical voice commands to be executed obtained in each step of this embodiment into the fitting software, and output the values ​​of the parameters (e.g., weights and thresholds) that meet the highest accuracy of operation intent risk judgment.

[0108] Example 2

[0109] like Figure 4 As shown, this embodiment provides a voice recognition-based command conversion control system, including:

[0110] The data acquisition module is used to acquire data of voice commands to be executed, historical accident data, and current operation data;

[0111] The operation threat level analysis module is used to import the voice command data to be executed and the current operation data into the operation threat level analysis model to analyze the operation threat level caused by the voice command to be executed to the currently running operation.

[0112] The instruction content parsing anomaly analysis module is used to import the voice instruction data to be executed and historical accident data into the instruction content parsing anomaly analysis model to analyze the anomaly status of the content parsing of the voice instruction to be executed.

[0113] The Operation Intent Risk Assessment Module is used to build an Operation Intent Risk Assessment Model. It imports the analysis results of the operational threat level caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing into the Operation Intent Risk Assessment Model to assess the operational intent risk of the voice command to be executed.

[0114] The instruction conversion control module is used to convert the voice instructions to be executed based on the risk assessment results of the operation intent.

[0115] The control module is used to control the operation of the data acquisition module, the operation threat level analysis module, the instruction content parsing abnormal state analysis module, the operation intent risk assessment module, and the instruction conversion control module.

[0116] The parameters and steps of each unit module in the speech recognition-based command conversion control system of the present invention described above for implementing the corresponding functions can be referred to the parameters and steps in the embodiments of the speech recognition-based command conversion control method above, and will not be repeated here.

[0117] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. In particular, the embodiments for IoT devices and media are relatively simple in description because they are fundamentally similar to the method embodiments; relevant parts can be referred to the descriptions in the method embodiments.

[0118] The systems, media, and methods provided in the embodiments of the present invention are in one-to-one correspondence. Therefore, the systems and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be repeated here.

[0119] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0122] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0123] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0124] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0125] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0126] The above are merely embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A command conversion control method based on speech recognition, characterized in that, Includes the following steps: S1. Obtain the voice command data to be executed, historical accident data, and current operation data; S2. Import the voice command data to be executed and the current operation data into the operation threat level analysis model to analyze the degree of operation threat posed by the voice command to be executed to the currently running operation. S3. Import the voice command data to be executed and historical accident data into the command content parsing abnormal state analysis model, and analyze the abnormal state of the content parsing of the voice command to be executed. S4. Construct an operation intent risk assessment model. Import the analysis results of the degree of operation threat caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing into the operation intent risk assessment model to assess the operation intent risk of the voice command to be executed. S5. Based on the risk assessment results of the operational intent, convert the voice command to be executed; Step S2 involves analyzing the operational threat level posed by the voice command to be executed to the currently running operation, including the following specific steps: S21. Extract the voice command data to be executed and the current running operation data; S22. Based on the voice command data to be executed and the current operation data, construct an operation threat level analysis model, analyze the operation threat level caused by the voice command to be executed to the current operation, and obtain the operation threat level analysis results caused by the voice command to be executed to the current operation. The formula for calculating the operational threat level is as follows: ； In the formula, Cz represents the degree of operational threat posed by the voice command to be executed to the currently running operation, ct represents the analysis result of the degree of compatibility conflict threat posed by the voice command to be executed to the currently running operation, and yc represents the analysis result of the degree of abnormal urgency threat posed by the voice command to be executed to the currently running operation. The formula for calculating the degree of abnormal urgency threat is as follows: ； In the formula, yc represents the degree of urgency threat posed by the voice command to be executed to the currently running operation, Ci represents the number of operation steps parsed from the voice command to be executed in the voice command data, Ti represents the required execution time limit of the voice command to be executed in the voice command data, To represents the remaining execution time of the currently running operation in the current running operation data, and Ro represents the system resource utilization rate of the currently running operation in the current running operation data.

2. The command conversion control method based on speech recognition according to claim 1, characterized in that, The process of constructing the operational threat level analysis model in step S22 includes the following specific steps: S221. Based on the voice command data to be executed and the current operation data, analyze the degree of compatibility conflict threat caused by the voice command to be executed to the current operation, and obtain the analysis results of the degree of compatibility conflict threat caused by the voice command to be executed to the current operation. S222. Based on the voice command data to be executed and the current operation data, analyze the degree of urgency threat posed by the voice command to be executed to the current operation, and obtain the analysis results of the degree of urgency threat posed by the voice command to be executed to the current operation.

3. The command conversion control method based on speech recognition according to claim 2, characterized in that, The analysis of abnormal states in the parsing of the voice command to be executed in step S3 includes the following specific steps: S31. Extract the voice command data to be executed and historical accident data; S32. Based on the voice command data to be executed and historical accident data, construct an abnormal state analysis model for command content parsing, analyze the abnormal state of the content parsing of the voice command to be executed, and obtain the analysis results of the abnormal state of the content parsing of the voice command to be executed. The formula for calculating content parsing anomalies is as follows: ； In the formula, Nr represents the abnormal state of the content parsing of the voice command to be executed, wz represents the result of the analysis of the abnormal state of the semantic integrity of the voice command to be executed, and hg represents the result of the analysis of the abnormal state of the semantic compliance of the voice command to be executed.

4. The command conversion control method based on speech recognition according to claim 3, characterized in that, The construction process of the instruction content parsing abnormal state analysis model in step S32 includes the following specific steps: S321. Based on the voice command data to be executed and historical accident data, analyze the semantic integrity anomaly state of the voice command to be executed, and obtain the analysis result of the semantic integrity anomaly state of the voice command to be executed. S322. Based on the voice command data to be executed and historical accident data, analyze the semantic compliance anomaly state of the voice command to be executed, and obtain the analysis results of the semantic compliance anomaly state of the voice command to be executed.

5. The command conversion control method based on speech recognition according to claim 4, characterized in that, The step S4, which involves constructing an operational intent risk assessment model, includes the following specific steps: S41. Extract the analysis results of the operational threat level caused by the voice command to be executed to the currently running operation, and the analysis results of the abnormal state of the content parsing of the voice command to be executed; S42. Based on the analysis results of the degree of operational threat caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing, assess the operational intent risk of the voice command to be executed; The formula for assessing operational intent risk is as follows: ； In the formula, ZY represents the operational intent risk of the voice command to be executed, Cz represents the operational threat level analysis result caused by the voice command to be executed to the currently running operation, and Nr represents the abnormal state analysis result of the content parsing of the voice command to be executed. These are the weights for the impact of operational threats and the weights for the impact of content parsing anomalies.

6. The command conversion control method based on speech recognition according to claim 5, characterized in that, In step S5, the voice command to be executed is converted based on the risk assessment result of the operation intention, including the following specific steps: S51. Obtain the risk assessment results of the operational intent of the voice command to be executed; S52. A preset operation intent risk threshold is set. When the operation intent risk assessment result of the voice command to be executed is greater than or equal to the operation intent risk threshold, the voice command to be executed is rejected. When the operation intent risk assessment result of the voice command to be executed is less than the operation intent risk threshold, the voice command to be executed is converted.

7. A voice recognition-based command conversion control system, implemented based on any one of claims 1-6, characterized in that, The system includes: The data acquisition module is used to acquire data of voice commands to be executed, historical accident data, and current operation data; The operation threat level analysis module is used to import the voice command data to be executed and the current operation data into the operation threat level analysis model to analyze the operation threat level caused by the voice command to be executed to the currently running operation. The instruction content parsing anomaly analysis module is used to import the voice instruction data to be executed and historical accident data into the instruction content parsing anomaly analysis model to analyze the anomaly status of the content parsing of the voice instruction to be executed. The Operation Intent Risk Assessment Module is used to build an Operation Intent Risk Assessment Model. It imports the analysis results of the operational threat level caused by the voice command to be executed to the currently running operation and the analysis results of the abnormal state of content parsing into the Operation Intent Risk Assessment Model to assess the operational intent risk of the voice command to be executed. The instruction conversion control module is used to convert the voice instructions to be executed based on the risk assessment results of the operation intent. The control module is used to control the operation of the data acquisition module, the operation threat level analysis module, the instruction content parsing abnormal state analysis module, the operation intent risk assessment module, and the instruction conversion control module; The analysis of the operational threat level posed by the voice command to be executed to the currently running operation includes the following specific steps: S21. Extract the voice command data to be executed and the current running operation data; S22. Based on the voice command data to be executed and the current operation data, construct an operation threat level analysis model, analyze the operation threat level caused by the voice command to be executed to the current operation, and obtain the operation threat level analysis results caused by the voice command to be executed to the current operation. The formula for calculating the operational threat level is as follows: ； In the formula, Cz represents the degree of operational threat posed by the voice command to be executed to the currently running operation, ct represents the analysis result of the degree of compatibility conflict threat posed by the voice command to be executed to the currently running operation, and yc represents the analysis result of the degree of abnormal urgency threat posed by the voice command to be executed to the currently running operation. The formula for calculating the degree of abnormal urgency threat is as follows: ； In the formula, yc represents the degree of urgency threat posed by the voice command to be executed to the currently running operation, Ci represents the number of operation steps parsed from the voice command to be executed in the voice command data, Ti represents the required execution time limit of the voice command to be executed in the voice command data, To represents the remaining execution time of the currently running operation in the current running operation data, and Ro represents the system resource utilization rate of the currently running operation in the current running operation data.

Images