Voice diagnosis method and device, storage medium and computer program product

Abstract

The invention discloses a voice diagnosis method and device, a storage medium and a computer program product, and relates to the technical field of vehicle software development, and the method comprises the steps: detecting whether a mobile storage device is successfully mounted or not; when the mobile storage device is successfully mounted, receiving a voice instruction; inputting the voice instruction into a pre-constructed voice instruction recognition model for recognition and analysis to obtain an instruction analysis result; and copying diagnosis data obtained according to the instruction analysis result into a corresponding directory of the mobile storage device. Through the voice instruction recognition model, automatic recognition of the voice instruction is achieved, manual operation is reduced, the operation efficiency of diagnosing vehicle software in the vehicle software development process is improved, and dependence on manpower is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle software development, and in particular to a speech diagnosis method, device, storage medium and computer program product. Background Art

[0002] In the process of vehicle software development, it is a common method to diagnose vehicle software by exporting system logs or reading fault causes based on fault codes to analyze technical problems in the software development process. However, currently exporting system logs or reading fault causes is overly dependent on manual operations, requiring vehicle software developers to mount mobile storage devices to vehicle software and enter engineering mode through key verification to export the log system or read fault causes. This process is cumbersome and over-reliance on manual operations leads to inefficiency in exporting system logs and reading fault causes, resulting in waste of resources and increased error costs.

[0003] Therefore, how to improve the operational efficiency of diagnostic software during vehicle software development and reduce reliance on manual labor has become an urgent problem to be solved in this application.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a voice diagnosis method, device, storage medium and computer program product, aiming to solve the technical problem of how to improve the operating efficiency of diagnostic software in the vehicle software development process and reduce dependence on manual labor.

[0006] To achieve the above object, the present application proposes a voice diagnosis method, which is applied to a diagnostic device, and the method comprises:

[0007] Check whether the mobile storage device is mounted successfully;

[0008] When the mobile storage device is successfully mounted, receiving a voice command;

[0009] Input the voice command into a pre-built voice command recognition model for recognition and analysis to obtain a command analysis result;

[0010] The diagnostic data obtained according to the instruction parsing result is copied into the corresponding directory of the mobile storage device.

[0011] In one embodiment, before the step of inputting the voice command into a pre-built voice command recognition model for recognition and analysis to obtain the command analysis result, the step further includes:

[0012] The deep convolutional neural network framework is trained according to the preset sample data set to obtain the voice command recognition model.

[0013] In one embodiment, the step of training a deep convolutional neural network framework according to a preset sample data set to obtain a voice command recognition model includes:

[0014] Acquire a sample data set and preprocess the sample data set;

[0015] Build a deep convolutional neural network framework by designing input layer, convolution layer, pooling layer, and fully connected layer;

[0016] Initializing the weights and biases of the deep convolutional neural network framework;

[0017] Select the cross entropy loss function as the loss function for initializing the deep convolutional neural network framework;

[0018] Based on the preprocessed sample data set, a deep convolutional neural network framework with a loss function is trained to obtain a voice command recognition model.

[0019] In one embodiment, the step of training the deep convolutional neural network framework based on the preprocessed sample data set to obtain the voice command recognition model further includes:

[0020] Optimizing the voice command recognition model using regularization technology;

[0021] Evaluate the performance of the instruction recognition model according to the preprocessed sample data set to obtain a performance evaluation result;

[0022] The structure and parameters of the instruction recognition model are regularly adjusted according to the performance evaluation results to maintain the operation performance.

[0023] In one embodiment, after the step of detecting whether the mobile storage device is successfully mounted, the step further includes:

[0024] When the mobile storage device fails to mount, the user is notified of the mounting failure through a visual mark.

[0025] In one embodiment, the step of copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device includes:

[0026] When the instruction parsing result is a fault code reading, a corresponding fault summary is retrieved according to the instruction parsing result;

[0027] The corresponding fault summary is copied into the corresponding directory of the mobile storage device.

[0028] In one embodiment, the step of copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device includes:

[0029] When the instruction parsing result is system log export, packaging the system log according to the instruction parsing result;

[0030] The system log is copied into the corresponding directory of the mobile storage device.

[0031] In addition, to achieve the above objectives, the present application also proposes a speech diagnosis device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the speech diagnosis method described above.

[0032] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the speech diagnosis method described above are implemented.

[0033] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the speech diagnosis method described above are implemented.

[0034] One or more technical solutions proposed in this application have at least the following technical effects:

[0035] When the mobile storage device successfully mounts the diagnostic device, it receives the voice command from the vehicle software development staff, and the voice command recognition model deployed in the diagnostic device recognizes the voice command and obtains the command parsing result. The diagnostic device performs corresponding operations according to the command parsing result, obtains the corresponding diagnostic data, and copies the diagnostic data into the mobile storage device. Through the voice command recognition model, automatic recognition of voice commands is achieved, thereby reducing the need for manual operation. The diagnostic device automatically obtains diagnostic data and copies it to the mobile storage device, avoiding data acquisition errors that may be caused by manual operation. Through the automated process, the time from receiving commands to obtaining diagnostic data is shortened, the operating efficiency of the diagnostic software in the vehicle software development process is improved, and the dependence on manual labor is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0038] Figure 1 A flowchart of the first embodiment of the speech diagnosis method of the present application is provided;

[0039] Figure 2 A flowchart of the second embodiment of the speech diagnosis method of the present application is provided;

[0040] Figure 3 A flowchart of the third embodiment of the speech diagnosis method of the present application is provided;

[0041] Figure 4 A simplified flowchart of a speech diagnosis method is shown;

[0042] Figure 5 This is a schematic diagram of the module structure of the speech diagnosis device according to an embodiment of the present application;

[0043] Figure 6 Schematic diagram of the device structure of the hardware operating environment involved in the voice diagnosis method in the embodiment of the present application.

[0044] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0045] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0046] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0047] The main solution of the embodiment of the present application is: detect whether the mobile storage device is successfully mounted, and when the mobile storage device is successfully mounted, receive the voice command. Input the voice command into the pre-built voice command recognition model for recognition and analysis to obtain the command analysis result. The diagnostic device automatically obtains the diagnostic data according to the command analysis result, and copies the diagnostic data into the corresponding directory of the mobile storage device for further analysis.

[0048] The embodiment of the present application takes into account that, since the current export of system logs or reading of fault causes is overly dependent on manual operation, the vehicle software developer is required to mount the mobile storage device to the vehicle software and enter the engineering mode through key verification. When the diagnostic device is in the engineering mode, the technician operates the diagnostic device to enter the corresponding interface to obtain the corresponding diagnostic data, and manually imports the diagnostic data into the mobile storage device. Each step of this vehicle software diagnostic process relies on manual operation, which is cumbersome and inefficient, resulting in waste of resources and increased error costs.

[0049] Therefore, the present application provides a solution. When the mobile storage device successfully mounts the diagnostic device, it receives the voice command of the vehicle software development staff, and the voice command recognition model deployed in the diagnostic device recognizes the voice command to obtain the command parsing result. The diagnostic device performs corresponding operations according to the command parsing result, obtains the corresponding diagnostic data, and copies the diagnostic data into the mobile storage device. Through the voice command recognition model, automatic recognition of voice commands is achieved, thereby reducing the need for manual operation. The diagnostic device automatically obtains diagnostic data and copies it to the mobile storage device, avoiding data acquisition errors that may be caused by manual operation. Through the automated process, the time from receiving the command to obtaining the diagnostic data is shortened, the operational efficiency of diagnosing vehicle software during the vehicle software development process is improved, and the dependence on manual labor is reduced.

[0050] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a diagnostic device, etc. The following takes the diagnostic device as an example to illustrate this embodiment and the following embodiments.

[0051] Based on this, the present application embodiment provides a voice diagnosis method, which is applied to a diagnostic device, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the speech diagnosis method of the present application.

[0052] In this embodiment, the voice diagnosis method includes steps S10 to S40:

[0053] Step S10, detecting whether the mobile storage device is mounted successfully;

[0054] The diagnostic device periodically scans all USB ports to detect whether a new USB mobile storage device is inserted. Once the diagnostic device detects a new USB device, it will enumerate (identify) the new device. The enumeration process includes reading the Vendor ID and Product ID (VID and PID) of the new USB mobile storage device, as well as the description information of the new USB mobile storage device.

[0055] Step S20, when the mobile storage device is successfully mounted, receiving a voice command;

[0056] When the diagnostic device successfully mounts the mobile storage device, the diagnostic device will return a mount point, usually " / mnt / usb-disk". By detecting this mount point, it can be determined whether the mobile storage device is mounted successfully.

[0057] When the mobile storage device is mounted successfully, the diagnostic device notifies the user through a visual mark, and the user then issues a voice command, such as "start diagnosis", "read log", etc. The diagnostic device receives the voice command sent by the user and converts the voice signal into an electrical signal.

[0058] Step S30, inputting the voice command into a pre-built voice command recognition model for recognition and analysis to obtain a command analysis result;

[0059] The diagnostic device needs to detect the noise in the voice command. The noise contained in the voice command may come from the external environment, such as wind noise, or from the device itself, such as electronic noise. When the noise in the voice command is detected, the diagnostic device uses a noise reduction algorithm to suppress the noise in the voice command.

[0060] A signal that is too strong may cause signal distortion, while a signal that is too weak may contain noise, affecting the accuracy of recognition. Furthermore, the diagnostic device performs gain control on the voice command after noise reduction. For example, if the electrical signal of the voice command is too weak or too strong, the diagnostic device will adjust the gain so that the electrical signal of the voice command is within an appropriate range.

[0061] By performing a series of preprocessing on voice commands, such as denoising and gain control, the quality of the electrical signal of the voice command is significantly improved. The preprocessed voice command is input into the pre-built voice command recognition model. The preprocessed voice command electrical signal is converted into a digital signal, and features such as MFCC (Mel Frequency Cepstral Coefficient) are extracted. These feature data are input into the voice recognition model.

[0062] The voice command recognition model is usually a deep learning model. After receiving the feature data, the neural network in the model begins to perform complex calculations. The neural network may include input layers, convolutional layers, pooling layers, fully connected layers, etc. The voice command recognition model recognizes the vocabulary and grammatical structure in the speech from the feature data through calculations, and understands the vocabulary and grammatical structure in the speech to obtain the command parsing results.

[0063] Step S40, copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device.

[0064] The software system of the diagnostic device matches the received instruction parsing results with the predefined instruction list. The instruction list contains all the operations and functions supported by the device. Once the match is successful, the diagnostic device determines the specific operation to be performed. For example, if the voice instruction is "read log", the diagnostic device will determine that the operation of reading the log file needs to be performed. Based on the determined operation, the diagnostic device starts to perform the corresponding operation, which may include accessing storage devices, reading files, executing system commands, etc. The diagnostic device provides feedback to the user on the operation results in a visual way. For example, when the log file is successfully read, the message "Log has been read" is displayed.

[0065] The software system of the diagnostic device receives the operation results and stores them in the data storage unit, and further copies the operation results into the corresponding directory of the mobile storage device for subsequent analysis, reporting and debugging.

[0066] The diagnostic equipment automatically performs a series of operations, from receiving voice commands to obtaining diagnostic data, which greatly improves efficiency and convenience and reduces the need for manual operations. This automated process helps speed up problem diagnosis and resolution and improves the quality and efficiency of vehicle software development.

[0067] This embodiment provides a voice diagnosis method. When the mobile storage device successfully mounts the diagnostic device, it receives the voice command of the vehicle software development staff, and the voice command recognition model deployed in the diagnostic device recognizes the voice command to obtain the command parsing result. The diagnostic device performs corresponding operations according to the command parsing result, obtains the corresponding diagnostic data, and copies the diagnostic data into the mobile storage device. Through the voice command recognition model, automatic recognition of voice commands is achieved, thereby reducing the need for manual operation. The diagnostic device automatically obtains the diagnostic data and copies it to the mobile storage device, avoiding data acquisition errors that may be caused by manual operation. Through the automated process, the time from receiving the command to obtaining the diagnostic data is shortened, the operational efficiency of diagnosing vehicle software during the vehicle software development process is improved, and the dependence on manual labor is reduced.

[0068] Based on the first embodiment of the present application, the second embodiment of the present application is proposed. In the second embodiment of the present application, the same or similar contents as those of the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following.

[0069] On this basis, please refer to Figure 2 , Figure 2 A flowchart of a second embodiment of the speech diagnosis method of the present application is provided.

[0070] In this embodiment, the voice command is input into a pre-built voice command recognition model for recognition and analysis, and before step S30 of obtaining the command analysis result, step S231 is also included:

[0071] Step S231, training a deep convolutional neural network framework according to a preset sample data set to obtain a voice command recognition model.

[0072] The deep convolutional neural network framework is selected as the basic framework. The parameters of the basic deep convolutional neural network framework are modified and trained by presetting sample data sets to achieve better voice command recognition results.

[0073] Specifically, in a feasible implementation manner, step S231 of training a deep convolutional neural network framework according to a preset sample data set to obtain a voice command recognition model may include steps S1 to S5:

[0074] Step S1, obtaining a sample data set and preprocessing the sample data set;

[0075] The preset sample data set is the original speech data. After obtaining the sample data set, features such as MFCCs (Mel-frequency cepstral coefficients), PLP (perceptual linear prediction), FBANK (filter bank) and other features are extracted from the original speech data. The diversity of the data is increased by adding noise, time expansion, frequency conversion, etc., so as to improve the generalization ability of the final model. The sample data set is divided into a training set, a validation set and a test set for model training, validation and testing.

[0076] Step S2, building a deep convolutional neural network framework by designing input layer, convolution layer, pooling layer, and fully connected layer;

[0077] The deep convolutional neural network framework is further constructed by designing the input layer, convolution layer, pooling layer, and fully connected layer. The input layer is used to receive feature data; the convolution layer extracts local features of speech through multiple convolution layers; the pooling layer reduces the size of the feature map through pooling operations (such as maximum pooling and average pooling), reduces the number of parameters, reduces the number of parameters and the amount of calculation of the model, while retaining important feature information; the fully connected layer integrates the features extracted by the convolution layer and the pooling layer.

[0078] First, when designing the input layer, we need to select parameters that can reflect the essential characteristics of the speech signal. In speech recognition, commonly used features include MFCC (Mel-frequency cepstral coefficients), PLP (perceptual linear prediction), FBANK (filter bank), etc. These features can capture the time domain and frequency domain information of the speech. Based on the above features extracted from the sample data set, determine the dimension of the feature and the length of the feature sequence. Construct the input layer based on the feature dimension and feature sequence length.

[0079] Furthermore, the number and size of convolution kernels in the convolution layer are determined. The number of convolution kernels determines the number of feature maps, while the size of the convolution kernel determines the range of local features that each feature map can capture. In the recognition of voice commands, multiple convolution kernels are usually used to capture features of different scales. Specific weights and biases are designed for each convolution kernel. The weight of the convolution kernel determines the strength and direction of feature extraction, while the bias is used to adjust the offset of the output feature. Slide the convolution kernel on the feature map and calculate the convolution result at each position. The convolution operation performs a dot product operation on the input feature map and the weight of the convolution kernel, and adds the bias to generate a new feature map, thereby completing the construction of the convolution layer.

[0080] Furthermore, after the convolution layer, one or more pooling layers are added. The feature points on the feature map are connected using a fully connected layer to form a one-dimensional vector, and the features are further integrated and processed. The deep neural network framework is built through the previously designed input layer, convolution layer, pooling layer, and fully connected layer communication.

[0081] Step S3, initializing the weights and biases of the deep convolutional neural network framework;

[0082] The weights and biases of the deep convolutional neural network framework are initialized by He initialization technology. He initialization is an improved version of Xavier initialization for the ReLU activation function, which can make the weight distribution more uniform and reduce the gradient vanishing problem of the neural network. He initialization initializes the weights to a normally distributed random value with a mean of 0 and a variance of 2 / fan_in, where fan_in is the dimension of the input feature.

[0083] Step S4, selecting a cross entropy loss function as a loss function for initializing a deep convolutional neural network framework;

[0084] During the training process of the voice command recognition model, the cross entropy loss function is defined as the difference between the predicted probability distribution and the true label distribution. In each training iteration, the cross entropy loss function is used to calculate the difference between the model prediction and the true label. The calculated loss value reflects the inaccuracy of the model prediction. After calculating the loss, the back-propagation algorithm is used to calculate the gradient of each weight and bias. The gradient is the derivative of the loss function with respect to the weights and biases, and is used to guide the update of the weights and biases.

[0085] Step S5, training a deep convolutional neural network framework with a loss function based on the preprocessed sample data set to obtain a voice command recognition model.

[0086] According to the gradient value and the selected optimizer (such as SGD, Adam, etc.), the weights and biases of the network are updated. The purpose of weight update is to minimize the loss function and thus improve the performance of the model. Repeat the above steps, including forward propagation, loss calculation, back propagation, and weight update, until the preset number of iterations is reached or the early stopping condition is met.

[0087] The above is only a feasible implementation of step S231 provided in this embodiment. This embodiment does not specifically limit the specific implementation of step S231.

[0088] In this embodiment, a deep convolutional neural network framework is built by constructing an input layer, a convolutional layer, a pooling layer, and a fully connected layer, the weights and biases of the deep convolutional neural network framework are initialized, the cross entropy loss function is selected as the loss function for initializing the deep convolutional neural network framework, and the deep convolutional neural network framework with the loss function is trained to obtain a voice command recognition model that can accurately recognize a variety of voice commands, thereby improving the operational efficiency of diagnosing vehicle software during vehicle software development.

[0089] Based on the first embodiment and / or the second embodiment of the present application, the third embodiment of the present application is proposed. In the third embodiment of the present application, the same or similar contents as those of the above-mentioned first and second embodiments can be referred to the above introduction, and will not be repeated in the following.

[0090] On this basis, please refer to Figure 3 , Figure 3 This is a flow chart of the third embodiment of the speech diagnosis method of the present application.

[0091] In this embodiment, after step S231 of training a deep convolutional neural network framework according to a preset sample data set to obtain a voice command recognition model, steps S232 to S234 are also included:

[0092] Step S232, optimizing the voice command recognition model using regularization technology;

[0093] Use weight decay regularization technology to optimize the voice command recognition model. Add the sum of the squares of the weight vector to the loss function to encourage the regularization of the weight vector. At the same time, when compiling the model, you can set the weight decay coefficient in the optimizer parameters, such as using the weight_decay parameter for regularization in TensorFlow.

[0094] Step S233, evaluating the performance of the instruction recognition model according to the preprocessed sample data set to obtain a performance evaluation result;

[0095] Evaluate the performance of the voice command recognition model based on the validation set included in the preprocessed sample dataset, such as accuracy, recall, etc. Adjust hyperparameters such as learning rate, batch size, etc. based on the evaluation results.

[0096] Step S234: regularly adjusting the structure and parameters of the instruction recognition model according to the performance evaluation result to maintain the operation performance.

[0097] Based on the analysis results, consider whether the structure of the instruction recognition model needs to be adjusted, and regularly adjust the model's hyperparameters, such as learning rate, batch size, regularization strength, etc. Find the best parameter settings through experiments to improve the performance of the model. Retrain the model using the adjusted model structure and parameters, and make sure to use a new training dataset for training to avoid overfitting.

[0098] In this embodiment, regularization technology is used to optimize the voice command recognition model, and the structure and parameters of the command recognition model are further adjusted regularly based on the performance evaluation results to maintain the operating performance, so that the voice command model can better adapt to different voice commands and environmental conditions, thereby customizing specific operations through different voice commands to achieve a high degree of automation, improve the operational efficiency of diagnosing vehicle software during vehicle software development, and further reduce dependence on manual labor.

[0099] Based on the above embodiments of the present application, a fourth embodiment of the present application is proposed. In the fourth embodiment of the present application, the same or similar contents as those of the above embodiments can be referred to the above introduction, and will not be described in detail later.

[0100] In a feasible implementation of this embodiment, the step S40 of copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device may include steps S401 to S402:

[0101] Step S401, when the instruction parsing result is a fault code reading, a corresponding fault summary is retrieved according to the instruction parsing result;

[0102] First, parse the fault code in the instruction parsing result. The fault code is a combination of numbers and letters, representing a specific fault or abnormality. Use the parsed fault code as a keyword to query the fault database. The fault database is a database that stores various fault codes and their meanings. Retrieve fault information related to the parsed fault code in the fault database. This information may include a description of the fault, possible causes, solutions, etc. Summarize the retrieved fault information for easy understanding and processing. The fault information summary includes the severity of the fault, the impact on the system, repair suggestions, etc.

[0103] Step S402: copy the corresponding fault summary into the corresponding directory of the mobile storage device.

[0104] First, determine the directory on the mobile storage device for storing fault summaries, convert the fault summary information into a format suitable for storage and transmission, such as a text file, PDF document, database file, etc., and write the fault summary data into the corresponding directory of the mobile storage device. If the amount of fault summary data is large, write it in batches or use a data stream.

[0105] In this embodiment, the corresponding fault summary is retrieved according to the instruction parsing result to help the user or system quickly identify and handle the fault, thereby improving the efficiency of fault diagnosis and reducing the time of fault handling, thereby improving the operational efficiency of diagnosing vehicle software during vehicle software development and reducing dependence on manual labor.

[0106] In another feasible implementation of this embodiment, the step S40 of copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device may include steps A401 to A402:

[0107] Step A401, when the instruction parsing result is system log export, packaging the system log according to the instruction parsing result;

[0108] Determine the type and scope of the system logs to be exported based on the information in the command parsing results. The type and scope of the system logs to be exported may include specific log files, log levels, time periods, etc. Further, read the system log file or log data storage to obtain log records of the specified type. According to the filtering conditions in the command parsing results, filter out relevant log records, filter specific log messages, exclude certain log levels, etc. Sort the filtered log records for easy analysis and understanding. The sorting may be based on timestamps, log levels, message content, etc.

[0109] Furthermore, the filtered and sorted log records are packaged into a unified file or data structure. For example, log records are written to a file, or log data is stored in a database table. Finally, the packaged log file is named for easy identification and archiving. The file name may contain information such as log type, log level, and packaging time.

[0110] Step A402, copying the system log into the corresponding directory of the mobile storage device.

[0111] After ensuring that the mobile storage device has been successfully mounted and has sufficient space to store the packaged log files, copy the packaged log files from the computer or diagnostic device to the specified directory on the mobile storage device.

[0112] In this embodiment, the system log is packaged according to the instruction parsing result, and the packaged log file is copied to a mobile storage device for subsequent analysis and processing. This method can easily transfer log data from the diagnostic device to the mobile storage device so that users or technicians can view and analyze it on an external device. The operation is simple and efficient, which improves the operational efficiency of diagnosing vehicle software during vehicle software development and reduces dependence on manual labor.

[0113] In summary, the voice command recognition model is used to recognize and parse the voice command. Since the voice command processing model can quickly process a variety of different voice commands, including natural language and keywords, different voice commands are recognized and parsed according to the voice command recognition model, which improves the efficiency of diagnosing vehicle software during vehicle software development, reduces dependence on manual labor, and can also customize specific operations according to voice commands to adapt to changing user needs and environmental conditions, providing highly flexible, efficient and convenient operations.

[0114] For example, in order to help understand the implementation process of the voice diagnosis method obtained by combining this embodiment with the fourth embodiment, please refer to Figure 4 , Figure 4 A brief flow chart of a speech diagnosis method is provided, specifically:

[0115] First, a U disk is mounted on the diagnostic device, and the diagnostic device detects whether the U disk is mounted successfully. If the U disk is mounted successfully, the diagnostic device receives the voice command.

[0116] Furthermore, voice commands are recognized. For example, when the voice command recognition result is to read a fault code, the diagnostic device retrieves the corresponding fault summary based on the fault code and copies it to the corresponding directory of the USB flash drive. For another example, when the voice command recognition result is to export a system log, the diagnostic device packages all system logs and copies them to the corresponding directory of the USB flash drive. Different specific operations are recognized according to different voice commands. The highly flexible diagnostic vehicle software operation further improves the operating efficiency and reduces the reliance on manual labor.

[0117] Finally, further analysis is performed based on the log files or fault summary data copied to the USB drive to determine the root cause of the fault and develop a corresponding repair plan, or identify problems in the system so that measures can be taken to prevent future failures.

[0118] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the speech diagnosis method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0119] This application also provides a voice diagnosis device, please refer to Figure 5, the speech diagnosis device comprises:

[0120] The detection module 10 is used to detect whether the mobile storage device is successfully mounted;

[0121] The diagnostic device periodically scans all USB ports to detect whether a new USB mobile storage device is inserted. Once the diagnostic device detects a new USB device, it will enumerate (identify) the new device. The enumeration process includes reading the Vendor ID and Product ID (VID and PID) of the new USB mobile storage device, as well as the description information of the new USB mobile storage device.

[0122] A voice command receiving module 20 receives a voice command when the mobile storage device is successfully mounted;

[0123] When the diagnostic device successfully mounts the mobile storage device, the diagnostic device will return a mount point, usually " / mnt / usb-disk". By detecting this mount point, it can be determined whether the mobile storage device is successfully mounted. When the mobile storage device is successfully mounted, the diagnostic device notifies the user through a visual logo, and the user then issues a voice command, such as "start diagnosis", "read log", etc. The diagnostic device receives the voice command sent by the user and converts the voice signal into an electrical signal.

[0124] The instruction parsing module 30 inputs the voice instruction into a pre-built voice instruction recognition model for recognition and parsing to obtain an instruction parsing result;

[0125] The voice command recognition model is usually a deep learning model. After receiving the feature data, the neural network in the model begins to perform complex calculations. The neural network may include input layers, convolutional layers, pooling layers, fully connected layers, etc. The voice command recognition model recognizes the vocabulary and grammatical structure in the speech from the feature data through calculations, and understands the vocabulary and grammatical structure in the speech to obtain the command parsing results.

[0126] The operating module 40 copies the diagnostic data obtained according to the instruction parsing result into a corresponding directory of the mobile storage device.

[0127] The speech diagnosis device provided by the present application adopts the speech diagnosis method in the above embodiment to solve the technical problem of speech diagnosis. Compared with the prior art, the beneficial effects of the speech diagnosis device provided by the present application are the same as the beneficial effects of the speech diagnosis method provided by the above embodiment, and other technical features in the speech diagnosis device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0128] The present application provides a speech diagnosis device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the speech diagnosis method in the above-mentioned embodiment 1.

[0129] Reference below Figure 6 , which shows a schematic diagram of the structure of a voice diagnosis device suitable for implementing the embodiment of the present application. The voice diagnosis device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The voice diagnosis device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0130] like Figure 6 As shown, the speech diagnosis device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the speech diagnosis device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the voice diagnosis device to communicate wirelessly or wired with other devices to exchange data. Although the voice diagnosis device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0131] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0132] The speech diagnosis device provided by the present application adopts the speech diagnosis method in the above embodiment to solve the technical problem of speech diagnosis. Compared with the prior art, the beneficial effects of the speech diagnosis device provided by the present application are the same as the beneficial effects of the speech diagnosis method provided by the above embodiment, and the other technical features in the speech diagnosis device are the same as the features disclosed in the method of the previous embodiment, which will not be described in detail here.

[0133] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0134] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0135] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the speech diagnosis method in the above-mentioned embodiment.

[0136] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0137] The computer-readable storage medium may be included in the speech diagnosis device; or may exist independently without being assembled into the speech diagnosis device.

[0138] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the voice diagnosis device, the voice diagnosis device: detects whether the mobile storage device is successfully mounted; when the mobile storage device is successfully mounted, receives a voice instruction; inputs the voice instruction into a pre-built voice instruction recognition model for recognition and analysis to obtain an instruction analysis result; and copies the diagnostic data obtained according to the instruction analysis result into a corresponding directory of the mobile storage device.

[0139] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0140] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0141] The modules involved in the embodiments described in the present application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0142] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned voice diagnosis method, and can solve the technical problems of voice diagnosis. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the voice diagnosis method provided in the above-mentioned embodiment, and will not be repeated here.

[0143] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned speech diagnosis method when executed by a processor.

[0144] The computer program product provided in this application can solve the technical problem of speech diagnosis. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the speech diagnosis method provided in the above embodiment, which will not be repeated here.

[0145] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A speech diagnosis method, characterized in that: Applied to diagnostic equipment, the method comprises: Check whether the mobile storage device is mounted successfully; When the mobile storage device is successfully mounted, receiving a voice command; Input the voice command into a pre-built voice command recognition model for recognition and analysis to obtain a command analysis result; The diagnostic data obtained according to the instruction parsing result is copied into the corresponding directory of the mobile storage device.

2. The method according to claim 1, characterized in that The step of inputting the voice command into a pre-built voice command recognition model for recognition and analysis to obtain a command analysis result also includes: The deep convolutional neural network framework is trained according to the preset sample data set to obtain the voice command recognition model.

3. The method according to claim 2, characterized in that The step of training a deep convolutional neural network framework according to a preset sample data set to obtain a voice command recognition model comprises: Acquire a sample data set and preprocess the sample data set; Build a deep convolutional neural network framework by designing input layer, convolution layer, pooling layer, and fully connected layer; Initializing the weights and biases of the deep convolutional neural network framework; Select the cross entropy loss function as the loss function for initializing the deep convolutional neural network framework; Based on the preprocessed sample data set, a deep convolutional neural network framework with a loss function is trained to obtain a voice command recognition model.

4. The method according to claim 3, characterized in that After the step of training the deep convolutional neural network framework according to the preset sample data set to obtain the voice command recognition model, the following step is further included: Optimizing the voice command recognition model using regularization technology; Evaluate the performance of the instruction recognition model according to the preprocessed sample data set to obtain a performance evaluation result; The structure and parameters of the instruction recognition model are regularly adjusted according to the performance evaluation results to maintain the operation performance.

5. The method according to claim 1, characterized in that After the step of detecting whether the mobile storage device is successfully mounted, the following steps are further included: When the mobile storage device fails to mount, the user is notified of the mounting failure through a visual mark.

6. The method according to claim 1, characterized in that The step of copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device includes: When the instruction parsing result is a fault code reading, a corresponding fault summary is retrieved according to the instruction parsing result; The corresponding fault summary is copied into the corresponding directory of the mobile storage device.

7. The method according to claim 1, characterized in that The step of copying the diagnostic data obtained according to the instruction parsing result into the corresponding directory of the mobile storage device includes: When the instruction parsing result is system log export, packaging the system log according to the instruction parsing result; The system log is copied into the corresponding directory of the mobile storage device.

8. A speech diagnosis device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the speech diagnosis method according to any one of claims 1 to 7.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the speech diagnosis method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the speech diagnosis method according to any one of claims 1 to 7 are implemented.

Images