A method and device for extracting optimization points of a user voice data extraction system

Abstract

The present invention discloses a method and device for extracting system optimization points based on user voice data. The method includes: obtaining a user voice command; determining whether the user voice command contains negative emotion voice data. If so, obtaining multiple interaction intention data and interaction action data of the user voice conversation before the generation of the user negative emotion voice data, where at least one interaction intention data corresponds to at least one interaction action data; determining whether there are functional points that may cause negative emotions of the user according to the interaction intention data and the interaction action data. If so, extracting the functional points that may cause negative emotions of the user; generating an interactive voice and an interactive interface according to the functional points that may cause negative emotions of the user, and confirming system optimization points through interaction with the user. The present invention can extract system functional points that cause negative emotions of users in voice interaction, facilitate subsequent product improvement, and enhance the accuracy of voice interaction.

Description

Technical Field

[0001] This application relates to the field of voice interaction technology, and in particular, to a method for extracting system optimization points based on user voice data and a device for extracting system optimization points based on user voice data. Background Art

[0002] Voice assistants based on deep learning are widely used in people's daily lives. People can use voice assistants to query the weather, add reminder items, set alarms, etc. However, since the current voice recognition system configured in the voice assistant is trained with limited voice data manually labeled with low efficiency, the limitations of the training data lead to blind spots in the voice recognition system, making it easy for the voice assistant to have recognition errors in daily use. When recognition errors occur, it may cause negative emotions of users, and then cause users to resist using the voice system. To better achieve human-machine voice interaction, there is currently a need for a technical solution to improve voice recognition ability and discover the pain points of users using the voice system, so as to improve the user voice interaction experience. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for extracting system optimization points based on user voice data and a device for extracting system optimization points based on user voice data to at least solve the above technical problems.

[0004] The present invention provides the following solutions:

[0005] According to one aspect of the present invention, there is provided a method for extracting system optimization points based on user voice data, including:

[0006] Obtain a user voice command;

[0007] Judge whether the user voice command contains negative emotion voice data. If so, then

[0008] Obtain multiple interaction intention data and interaction action data of the user voice conversation before the generation of the user negative emotion voice data, where one interaction intention data corresponds to at least one interaction action data;

[0009] Judge whether there is a function point that may cause negative emotions of users according to the interaction intention data and the interaction action data. If so, then extract the function point that may cause negative emotions of users;

[0010] Generate an interaction voice and an interaction interface according to the function point that may cause negative emotions of users, and confirm the system optimization point through interaction with the user.

[0011] Optionally, before judging whether the user voice command contains negative emotion voice data, the method for extracting system optimization points based on user voice data further includes:

[0012] Determine whether the user voice command is a functional command. If so, then

[0013] Execute the function corresponding to the user voice command according to the user voice command;

[0014] Determine whether the user voice command is at an execution node. If so, then

[0015] Determine whether the user voice command contains negative emotion voice data.

[0016] Optionally, the obtaining of multiple interaction intention data and interaction action data of the user voice conversation before the generation of the user negative emotion voice data includes:

[0017] Obtain at least three interaction intention data, where

[0018] One of the interaction intention data includes an association score between an interaction intention and multiple function points.

[0019] Optionally, the obtaining of multiple interaction intention data and interaction action data of the user voice conversation before the generation of the user negative emotion voice data includes:

[0020] Obtain at least three interaction action data, where

[0021] One of the interaction action data includes an association score between an interaction action and multiple function points.

[0022] Optionally, the determining whether there are function points that may cause negative emotions of the user according to the interaction intention data and the interaction action data. If so, then the extracting of the function points includes:

[0023] Obtain a preset weight adjustment strategy;

[0024] Perform weighted calculation on multiple interaction intention data respectively according to the preset weight adjustment strategy to obtain weighted values of the interaction intention data;

[0025] Perform weighted calculation on multiple interaction action data respectively according to the preset weight adjustment strategy to obtain weighted values of the interaction action data;

[0026] Obtain the weighted values of multiple interaction intention data and the weighted values of multiple interaction action data of the same function point;

[0027] Add up the weighted values of multiple interaction intention data and the weighted values of multiple interaction action data of the same function point to obtain possibility scores of multiple function points;

[0028] Obtain the three function points with the highest possibility scores among the multiple function points.

[0029] Optionally, the interaction intention data further includes the interaction intention generation time;

[0030] The weighted calculation of multiple pieces of interaction intention data respectively according to the preset weight adjustment strategy to obtain the weighted value of the interaction intention data includes:

[0031] Sort the multiple pieces of interaction intention data in reverse order according to the interaction intention generation time;

[0032] Multiply the sorted interaction intention data by preset weight values respectively to obtain the weighted value of the association score between the interaction intention and multiple function points, where

[0033] The preset weight values corresponding to the interaction intention data are arranged in a decreasing order according to the arrangement order of the interaction intention data from large to small.

[0034] Optionally, the interaction action data further includes the interaction action generation time;

[0035] The weighted calculation of multiple pieces of interaction action data respectively according to the preset weight adjustment strategy to obtain the weighted value of each piece of interaction action data includes:

[0036] Sort the multiple pieces of interaction action data in reverse order according to the interaction action generation time;

[0037] Multiply the sorted interaction action data by preset weight values respectively to obtain the weighted value of the association score between the interaction action and multiple function points, where

[0038] The preset weight values corresponding to the interaction action data are arranged in a decreasing order according to the arrangement order of the interaction action data from large to small.

[0039] Optionally, generating an interactive voice and an interactive interface according to the function points that may cause negative emotions of the user, and confirming system optimization points through interaction with the user includes:

[0040] Generate a first question voice according to the three function points with the highest possibility scores among the multiple function points;

[0041] Generate a first interactive interface according to the three function points with the highest possibility scores among the multiple function points;

[0042] Broadcast the first question voice;

[0043] Display the first interactive interface;

[0044] Obtain the first reply voice input by the user;

[0045] Determine whether the first replied voice includes any one of the three function points with the highest possibility scores. If so, use this function point as the system optimization point.

[0046] Optionally, generating an interactive voice and an interactive interface according to the function points that may cause negative emotions of the user, and confirming the system optimization point through voice interaction with the user includes:

[0047] If the first replied voice of the user does not include any one of the three function points with the highest possibility scores, generate a second question voice and a second interactive interface;

[0048] Broadcast the second question voice;

[0049] Display the second interactive interface;

[0050] Obtain the second replied voice input by the user;

[0051] Determine whether the second replied voice includes a system function point. If so, use the system function point in the second replied voice as the system optimization point.

[0052] The present invention also provides a device for extracting system optimization points based on user voice data, including:

[0053] A voice command acquisition module for acquiring user voice commands;

[0054] A judgment module for judging whether the user voice command contains negative emotion voice data;

[0055] A data acquisition module for acquiring multiple interactive intention data and interactive action data of the user voice conversation before the generation of the user negative emotion voice data, wherein one interactive intention data corresponds to at least one interactive action data;

[0056] A function point extraction module for judging whether there are function points that may cause negative emotions of the user according to the interactive intention data and the interactive action data. If so, extract the function points that may cause negative emotions of the user;

[0057] An interaction module for generating an interactive voice and an interactive interface according to the function points that may cause negative emotions of the user, and confirming the system optimization point through interaction with the user.

[0058] The present invention has the following advantages compared with the prior art:

[0059] The present invention utilizes the voice data of users' negative emotions to identify the pain points of users' poor voice interaction. By interacting with the users to confirm the system function points that cause the users' negative emotions, these system function points are used as system optimization points, which facilitates subsequent improvement of product functions, simplifies the feedback operation of users, enhances the user experience, and further improves the accuracy of voice interaction and the working efficiency of the system. Description of the Drawings

[0060] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0061] Figure 1 Schematic flowchart of the method for extracting system optimization points based on user voice data according to an embodiment of the present invention;

[0062] Figure 2 Schematic flowchart of the method for extracting system optimization points based on user voice data according to another embodiment of the present invention;

[0063] Figure 3 Schematic structural diagram of the device for extracting system optimization points based on user voice data according to an embodiment of the present invention;

[0064] Figure 4 Structural diagram of the electronic device that can implement the present invention. Detailed Description of the Embodiments

[0065] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0066] Figure 1 Schematic flowchart of the method for extracting system optimization points based on user voice data according to an embodiment of the present invention;

[0067] As Figure 1 shown, the method for extracting system optimization points based on user voice data includes:

[0068] Step 1: Obtain the user voice command;

[0069] Step 2: Determine whether the user voice command contains negative emotion voice data. If so, then

[0070] Step 3: Obtain multiple interactive intent data and interactive action data of the user's voice conversation before the generation of the user's negative emotion voice data, where one interactive intent data corresponds to at least one interactive action data;

[0071] Step 4: Determine whether there are functional points that may cause the user's negative emotion according to the interactive intent data and the interactive action data. If so, extract the functional points that may cause the user's negative emotion;

[0072] Step 5: Generate an interactive voice and an interactive interface according to the functional points that may cause the user's negative emotion, and confirm the system optimization points through interaction with the user.

[0073] The present invention has the following advantages compared with the prior art:

[0074] The present invention uses the voice data of the user's negative emotion to identify the pain points of the user's poor voice interaction. By interacting with the user to confirm the system functional points that may cause the user's negative emotion, these system functional points are extracted as system optimization points, which is convenient for subsequent product function improvement, simplifies the user's feedback operation, improves the user experience, and then enhances the accuracy of voice interaction and improves the working efficiency of the system.

[0075] In this embodiment, before determining whether the user's voice command contains negative emotion voice data, the method for extracting system optimization points based on the user's voice data further includes:

[0076] Determine whether the user's voice command is a functional command. If so,

[0077] Execute the function corresponding to the user's voice command according to the user's voice command;

[0078] Determine whether the user's voice command is at an execution node. If so,

[0079] Determine whether the user's voice command contains negative emotion voice data.

[0080] In this embodiment, obtaining multiple interactive intent data and interactive action data of the user's voice conversation before the generation of the user's negative emotion voice data includes:

[0081] Obtain at least three interactive intent data, where

[0082] One interactive intent data includes the association score between one interactive intent and multiple functional points.

[0083] In this embodiment, obtaining multiple interactive intent data and interactive action data of the user's voice conversation before the generation of the user's negative emotion voice data includes:

[0084] Obtain at least three interactive action data, where

[0085] An interaction action data includes the associated scores between an interaction action and multiple function points.

[0086] In this embodiment, based on the interaction intention data and the interaction action data, it is determined whether there are function points that may cause negative emotions to the user. If so, the extracted function points include:

[0087] Obtain a preset weight adjustment strategy;

[0088] Perform weighted calculations on multiple interaction intention data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction intention data;

[0089] Perform weighted calculations on multiple interaction action data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction action data;

[0090] Obtain the weighted values of multiple interaction intention data and the weighted values of multiple interaction action data for the same function point;

[0091] Add up the weighted values of multiple interaction intention data and the weighted values of multiple interaction action data for the same function point to obtain the possibility scores of multiple function points;

[0092] Obtain the three function points with the highest possibility scores among the multiple function points.

[0093] In this embodiment, the interaction intention data further includes the interaction intention generation time;

[0094] Performing weighted calculations on multiple interaction intention data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction intention data includes:

[0095] Sort multiple interaction intention data in reverse order according to the interaction intention generation time;

[0096] Multiply the sorted interaction intention data by the preset weights respectively to obtain the weighted values of the associated scores between the interaction intention and multiple function points, where

[0097] The preset weights corresponding to the interaction intention data are arranged in decreasing order according to the arrangement order of the interaction intention data from large to small.

[0098] In this embodiment, the interaction action data further includes the interaction action generation time;

[0099] Performing weighted calculations on multiple interaction action data respectively according to the preset weight adjustment strategy to obtain the weighted value of each interaction action data includes:

[0100] Sort multiple interaction action data in reverse order according to the interaction action generation time;

[0101] Multiply the sorted interactive action data by preset weights respectively to obtain the weighted values of the association scores between the interactive actions and multiple function points, where

[0102] The preset weights corresponding to the interactive action data are arranged in decreasing order according to the arrangement order of the interactive action data from large to small.

[0103] In this embodiment, the interactive voice and the interactive interface are generated according to the function points that may cause negative emotions of the user. The system optimization points confirmed by interacting with the user include:

[0104] Generate the first question voice according to the three function points with the highest possibility scores among the multiple function points;

[0105] Generate the first interactive interface according to the three function points with the highest possibility scores among the multiple function points;

[0106] Broadcast the first question voice;

[0107] Display the first interactive interface;

[0108] Obtain the first reply voice input by the user;

[0109] Judge whether the first reply voice includes any one of the three function points with the highest possibility scores. If so, use this function point as the system optimization point.

[0110] In this embodiment, the interactive voice and the interactive interface are generated according to the function points that may cause negative emotions of the user. The system optimization points confirmed by interacting with the user's voice include:

[0111] If the first reply voice of the user does not include any one of the three function points with the highest possibility scores, generate the second question voice and the second interactive interface;

[0112] Broadcast the second question voice;

[0113] Display the second interactive interface;

[0114] Obtain the second reply voice input by the user;

[0115] Judge whether the second reply voice includes system function points. If so, use the system function points in the second reply voice as the system optimization points.

[0116] Figure 2 It is a schematic flowchart of the method for extracting system optimization points based on user voice data according to another embodiment of the present invention;

[0117] As Figure 2 shown, in this embodiment, the method for extracting system optimization points based on user voice data includes:

[0118] The user initiates a voice interaction with the in-vehicle intelligent voice assistant;

[0119] After the in-vehicle unit obtains the user's voice command, it first uses the ASR (Automatic Speech Recognition) technology to convert the voice into text;

[0120] The in-vehicle unit uses an algorithm model locally or in the cloud to perform semantic analysis on the user's voice command to determine whether the voice command belongs to a functional intention or a chitchat intention (this embodiment only focuses on the functional intention part);

[0121] In this embodiment, it is judged whether the current semantics is at an execution node, that is, after executing this voice command, the user does not need to issue the next command according to this function. For example, if the user's voice command is "Exit navigation", after the vehicle executes the exit navigation function according to this user voice command, since the navigation function has been closed, then theoretically, before the user turns on the navigation function again, the interaction between the user and the navigation function has ended, so this user voice command ("Exit navigation") is considered to be a user voice command at the execution node. If not, the user needs to continue to input commands; if at the execution node, after the in-vehicle unit completes the execution of the user voice command, it is judged whether the user voice command contains negative emotion voice data. Among them, negative emotion voice data includes voice commands such as "You can get out when exiting navigation", which will be judged as voice data with negative emotions;

[0122] It can be understood that if the user voice command does not belong to a functional command but contains negative emotion voice data, such as "The navigation is too stupid", at this time, the vehicle can also execute the method of extracting system optimization points according to this negative emotion voice data.

[0123] If the user voice command contains negative emotion voice data, it is judged which interaction function points before the user's negative emotion voice data cause the user's negative emotion. Specifically, multiple interaction intention data and interaction action data of the user's voice conversation before the generation of the user's negative emotion voice data are obtained, where at least one interaction intention data corresponds to one interaction action data; for example, when the user issues a voice command of "Open the window", the system recognizes the interaction intention that the user wants to open the window, and the system executes the interaction action of opening the window according to the interaction intention.

[0124] In this embodiment, three pieces of interactive intention data before the generation of the user's negative emotion voice data are obtained. One piece of interactive intention data includes the association scores between an interactive intention and multiple function points. The three obtained interactive intentions are Id, Ia, and Ib. It can be understood that there are different association scores between each interactive intention and five function points F1, F2, F3, F4, and F5. In the preset association score table (as shown in the following table), the association scores between an interactive intention and multiple function points are obtained respectively according to the three obtained interactive intentions Id, Ia, and Ib;

[0125]

[0126] In this embodiment, three pieces of interactive action data before the generation of the user's negative emotion voice data are obtained. One piece of interactive action data includes the association scores between an interactive action and multiple function points. It can be understood that there are different association scores between each interactive action and five function points F1, F2, F3, F4, and F5. The three obtained interactive actions are Ac, Ae, and Ad. In the preset association score table (as shown in the following table), the association scores between an interactive action and multiple function points are obtained respectively according to the three obtained interactive actions Ac, Ae, and Ad;

[0127]

[0128] In another embodiment, the vehicle-mounted computer can also determine whether there are function points that may cause the user's negative emotion by obtaining multiple pieces of interactive intention data and multiple pieces of interactive action data within 5 minutes before the generation of the user's negative emotion voice data;

[0129] In this embodiment, it is determined whether there are function points that may cause the user's negative emotion according to the three pieces of interactive intention data and the three pieces of interactive action data. If so, the function points that may cause the user's negative emotion are extracted;

[0130] Specifically, weighted calculations are respectively performed on the multiple pieces of interactive intention data according to the preset weight adjustment strategy to obtain the weighted values of the interactive intention data;

[0131] In this embodiment, the interactive intention data further includes the interactive intention generation time;

[0132] The three pieces of interactive intention data are sorted in reverse order according to the interactive intention generation time. The order of the three sorted interactive intentions is Ib, Ia, and Id;

[0133] Multiply the sorted interaction intention data by preset weights respectively to obtain the weighted values of the association scores between the interaction intentions and multiple function points. Among them, the preset weights corresponding to the interaction intention data are arranged in a decreasing order according to the arrangement order of the interaction intention data from large to small. In this embodiment, the preset weights are 3, 2, and 1. Specifically, the weight corresponding to the interaction intention Ib is the largest, and the weight corresponding to the interaction intention Id is the smallest. Multiply the interaction intention data by the corresponding weights respectively to obtain the weighted values of the interaction intention data.

[0134] Perform weighted calculations on multiple interaction action data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction action data;

[0135] In this embodiment, the interaction action data further includes the interaction action generation time;

[0136] Reverse-sort the three interaction action data according to the interaction action generation time, and the order of the three sorted interaction actions becomes Ad, Ae, Ac;

[0137] Multiply the sorted interaction action data by preset weights respectively to obtain the weighted values of the association scores between the interaction actions and multiple function points. Among them, the preset weights corresponding to the interaction action data are arranged in a decreasing order according to the arrangement order of the interaction intention data from large to small. In this embodiment, the preset weight corresponding to the interaction action Ad is 3, the preset weight corresponding to the interaction action Ae is 2, and the preset weight corresponding to the interaction action Ac is 1. Multiply the interaction action data by the corresponding weights respectively to obtain the weighted values of the interaction action data.

[0138] Obtain the weighted values of multiple interaction intention data and multiple interaction action data of the same function point;

[0139] Add up the weighted values of multiple interaction intention data and multiple interaction action data of the same function point to obtain the possibility scores of multiple function points;

[0140] The specific calculation process is as follows:

[0141] S(F1) = S(IbF1) * 3 + S(IaF1) * 2 + S(IdF1) * 1 + S(AdF1) * 3 + S(AeF1) * 2 + S(AcF1) * 1 = 26;

[0142] S(F2) = S(IbF2) * 3 + S(IaF2) * 2 + S(IdF2) * 1 + S(AdF2) * 3 + S(AeF2) * 2 + S(AcF2) * 1 = 14;

[0143] S(F3) = S(IbF3) * 3 + S(IaF3) * 2 + S(IdF3) * 1 + S(AdF3) * 3 + S(AeF3) * 2 + S(AcF3) * 1 = 27;

[0144] S(F4) = S(IbF4) * 3 + S(IaF4) * 2 + S(IdF4) * 1 + S(AdF4) * 3 + S(AeF4) * 2 + S(AcF4) * 1 = 17;

[0145] S(F5) = S(IbF5) * 3 + S(IaF5) * 2 + S(IdF5) * 1 + S(AdF5) * 3 + S(AeF5) * 2 + S(AcF5) * 1 = 3;

[0146] Obtain the three function points with the highest possibility scores among multiple function points. In this embodiment, the function points F3, F1, and F4 have the highest possibility scores. Therefore, extract the function points F3, F1, and F4.

[0147] In this embodiment, generate an interactive voice and an interactive interface according to the three function points with the highest possibility scores among multiple function points. The system optimization points confirmed through interaction with the user include:

[0148] Generate a first question voice according to the function points F3, F1, and F4, such as "Excuse me, do the function points F3, F1, and F4 have a poor experience in use?", and broadcast the first question voice to the user;

[0149] Generate a first interactive interface according to the function points F3, F1, and F4, and display it on the vehicle display device;

[0150] Obtain the first reply voice input by the user. If the first reply voice of the user includes any one of the function points F3, F1, and F4, then use this function point as the system optimization point.

[0151] If the first reply voice of the user does not include any one of the function points F3, F1, and F4, then generate a second question voice, such as "Excuse me, which function points have a poor experience in use?", and broadcast the second question voice to the user and synchronously display the second interactive interface;

[0152] Obtain the second reply voice input by the user;

[0153] Judge whether the second reply voice includes system function points. If so, use the system function points in the second reply voice as the system optimization points for subsequent analysis and optimization of functions.

[0154] It can be understood that this embodiment can collect more detailed information about poor experience from the user through voice interaction, which is convenient for subsequent analysis and optimization of functions.

[0155] Figure 3 Structural schematic diagram of a device based on the optimization points of the user voice data extraction system according to an embodiment of the present invention;

[0156] As Figure 3 shown, the present invention also provides a device based on the optimization points of the user voice data extraction system, including:

[0157] A voice command acquisition module for acquiring user voice commands;

[0158] A judgment module for judging whether the user voice command contains negative emotion voice data;

[0159] A data acquisition module for acquiring multiple interaction intention data and interaction action data of the user voice conversation before the generation of the user negative emotion voice data, wherein one interaction intention data corresponds to at least one interaction action data;

[0160] A function point extraction module for judging whether there are function points that may cause negative emotions of the user according to the interaction intention data and the interaction action data, and if so, extracting the function points that may cause negative emotions of the user;

[0161] An interaction module for generating interaction voices and interaction interfaces according to the function points that may cause negative emotions of the user, and confirming the system optimization points through interaction with the user.

[0162] Figure 4 It is a structural diagram of an electronic device that can implement the present invention.

[0163] As Figure 4 shown, the electronic device includes: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus; a computer program is stored in the memory. When the computer program is executed by the processor, the processor executes the steps of the method for extracting the optimization points of the user voice data extraction system.

[0164] The present application also provides a computer-readable storage medium, which stores a computer program executable by an electronic device. When the computer program runs on the electronic device, the electronic device executes the steps of the method for extracting the optimization points of the user voice data extraction system.

[0165] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0166] The electronic device includes a hardware layer, an operating system layer running on the hardware layer, and an application layer running on the operating system. The hardware layer includes hardware such as a Central Processing Unit (CPU), a Memory Management Unit (MMU), and a memory. The operating system can be any one or more computer operating systems that implement the control of the electronic device through a process. For example, the Linux operating system, the Unix operating system, the Android operating system, the iOS operating system, or the Windows operating system, etc. And in the embodiments of the present invention, the electronic device can be a handheld device such as a smart phone or a tablet computer, or an electronic device such as a desktop computer or a portable computer. The embodiments of the present invention do not particularly limit this.

[0167] The execution subject of the electronic device control in the embodiments of the present invention can be the electronic device, or a functional module in the electronic device that can call and execute a program. The electronic device can obtain the firmware corresponding to the storage medium, and the firmware corresponding to the storage medium is provided by the supplier. The firmware corresponding to different storage media can be the same or different, and this is not limited here. After the electronic device obtains the firmware corresponding to the storage medium, it can write the firmware corresponding to the storage medium into the storage medium. Specifically, it burns the firmware corresponding to the storage medium into the storage medium. The process of burning the firmware into the storage medium can be implemented by the prior art and will not be elaborated in the embodiments of the present invention.

[0168] The electronic device can also obtain the reset command corresponding to the storage medium, and the reset command corresponding to the storage medium is provided by the supplier. The reset commands corresponding to different storage media can be the same or different, and this is not limited here.

[0169] At this time, the storage medium of the electronic device is the storage medium written with the corresponding firmware. The electronic device can respond to the reset command corresponding to the storage medium in the storage medium written with the corresponding firmware, so that the electronic device resets the storage medium written with the corresponding firmware according to the reset command corresponding to the storage medium. The process of resetting the storage medium according to the reset command can be implemented by the prior art and will not be elaborated in the embodiments of the present invention.

[0170] For the convenience of description, when describing the above device, it is divided into various units and modules according to functions for separate description. Of course, when implementing the present application, the functions of each unit and module can be implemented in the same or multiple software and / or hardware.

[0171] Those skilled in the art of this technical field can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the field to which the present invention belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless specifically defined.

[0172] For method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of the present invention are not limited by the described order of actions, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential for the embodiments of the present invention.

[0173] From the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present application.

[0174] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting optimization points of a user voice data extraction system, characterized in that, Including: Obtain the user's voice command; Determine whether the user's voice command contains negative emotion voice data. If so, then Obtain multiple interaction intention data and interaction action data of the user's voice conversation before the generation of the user's negative emotion voice data, where one interaction intention data corresponds to at least one interaction action data; Judge whether there are functional points that may cause the user's negative emotion according to the interaction intention data and the interaction action data. If so, extract the functional points that may cause the user's negative emotion; Generate interaction voice and interaction interface according to the functional points that may cause the user's negative emotion, and confirm the system optimization points through interaction with the user; Among them, before judging whether the user's voice command contains negative emotion voice data, the method for extracting system optimization points based on user voice data further includes: Judge whether the user's voice command is a functional command. If so, then Execute the function corresponding to the user's voice command according to the user's voice command; Judge whether the user's voice command is at an execution node. The execution node refers to a situation where the user does not need to issue the next instruction according to the function corresponding to the user's voice command. If so, then Judge whether the user's voice command contains negative emotion voice data.

2. The method for extracting system optimization points based on user voice data according to claim 1, characterized in that, The obtaining of multiple interaction intention data and interaction action data of the user's voice conversation before the generation of the user's negative emotion voice data includes: Obtain at least three interaction intention data, where One of the interaction intention data includes the association score between one interaction intention and multiple functional points.

3. The method for extracting system optimization points based on user voice data according to claim 2, wherein, The obtaining of multiple interaction intention data and interaction action data of the user's voice conversation before the generation of the user's negative emotion voice data includes: Obtain at least three interaction action data, where One of the interaction action data includes the association score between one interaction action and multiple functional points.

4. The method for extracting system optimization points based on user voice data according to claim 3, characterized in that The judging whether there are functional points that may cause the user's negative emotion according to the interaction intention data and the interaction action data. If so, then the extraction of the functional points includes: Obtain a preset weight adjustment strategy; Perform weighted calculation on multiple interaction intention data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction intention data; Perform weighted calculation on multiple interaction action data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction action data; Obtain the weighted values of multiple interaction intention data and the weighted values of multiple interaction action data of the same functional point; Add up the weighted values of multiple interaction intention data and the weighted values of multiple interaction action data of the same functional point to obtain the possibility scores of multiple functional points; Obtain the three functional points with the highest possibility scores among the multiple functional points.

5. The method for extracting system optimization points based on user voice data according to claim 4, characterized in that The interaction intention data further includes the interaction intention generation time; The performing of weighted calculation on multiple interaction intention data respectively according to the preset weight adjustment strategy to obtain the weighted values of the interaction intention data includes: Sort the multiple interaction intention data in reverse order according to the interaction intention generation time; Multiply the sorted interaction intention data by the preset weight values respectively to obtain the weighted values of the association scores between the interaction intention and multiple functional points, where The preset weight values corresponding to the interaction intention data are arranged in decreasing order according to the arrangement order of the interaction intention data from large to small.

6. The method for extracting system optimization points based on user voice data according to claim 5, characterized in that The interaction action data further includes the generation time of the interaction action; The weighted calculation of multiple interaction action data respectively according to the preset weight adjustment strategy to obtain the weighted value of each interaction action data includes: Sort the multiple interaction action data in reverse chronological order according to the generation time of the interaction action; Multiply the sorted interaction action data by the preset weight values respectively to obtain the weighted values of the association scores between the interaction actions and multiple function points, where The preset weight values corresponding to the interaction action data are arranged in decreasing order according to the arrangement order of the interaction action data from large to small.

7. The method for extracting system optimization points based on user voice data according to claim 6, characterized in that Generating an interaction voice and an interaction interface according to the function points that may cause negative emotions of the user, and confirming system optimization points through interaction with the user includes: Generating a first question voice according to the three function points with the highest possibility scores among the multiple function points; Generating a first interaction interface according to the three function points with the highest possibility scores among the multiple function points; Broadcasting the first question voice; Displaying the first interaction interface; Obtaining the first reply voice input by the user; Judging whether the first reply voice includes any one of the three function points with the highest possibility scores. If so, taking this function point as the system optimization point.

8. The method for extracting system optimization points based on user voice data according to claim 7, characterized in that, Generating an interaction voice and an interaction interface according to the function points that may cause negative emotions of the user, and confirming system optimization points through voice interaction with the user includes: If the first reply voice of the user does not include any one of the three function points with the highest possibility scores, generating a second question voice and a second interaction interface; Broadcasting the second question voice; Displaying the second interaction interface; Obtaining the second reply voice input by the user; Judging whether the second reply voice includes system function points. If so, taking the system function points in the second reply voice as the system optimization points.

9. An apparatus for extracting optimization points of a user voice data extraction system, characterized in that, The device for extracting system optimization points based on user voice data is used to execute the method for extracting system optimization points based on user voice data according to any one of claims 1-8; The device for extracting system optimization points based on user voice data includes: A voice command acquisition module for acquiring user voice commands; A judgment module for judging whether the user voice command contains negative emotion voice data; A data acquisition module for acquiring multiple interaction intention data and interaction action data of the user voice conversation before the generation of the user negative emotion voice data, where one interaction intention data corresponds to at least one interaction action data; A function point extraction module for judging whether there are function points that may cause negative emotions of the user according to the interaction intention data and the interaction action data. If so, extracting the function points that may cause negative emotions of the user; An interaction module for generating an interaction voice and an interaction interface according to the function points that may cause negative emotions of the user, and confirming system optimization points through interaction with the user.

Images