Voice question and answer processing method and device

Abstract

The invention discloses a voice question and answer processing method and device, relates to the technical field of data processing, and mainly aims to solve the problem that the existing voice question and answer processing effect is poor. Comprising the following steps: acquiring voice data collected for a power grid dispatching service, and converting the voice data to obtain voice text data; vector conversion is carried out on the voice text data to obtain problem vector data, at least one service dimension of the problem vector data is determined, and the service dimension is obtained by analyzing collected power grid service data based on a power grid dispatching dimension; and generating a voice feedback result based on the dimension relevance integration of the problem vector data and the service dimension, and outputting the voice feedback result based on a pre-configured virtual playing object.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for processing voice question and answer. Background Art

[0002] As the use of artificial intelligence technology in various industries continues to deepen, heavy industry companies have also begun to use artificial intelligence technology to assist production operations. For example, in the power grid dispatching business, it is necessary to perform corresponding operations for power grid production and operation, power grid dispatching systems, and personnel operations. At this time, the corresponding information needs to be entered.

[0003] At present, the input of information such as power grid dispatching usually relies on manual review, confirmation, planning and arrangement. However, manual operation will lead to slow response of power grid dispatching business operations, and it is impossible to quickly execute relevant business of the power grid according to demand, which makes the power grid dispatching work cycle longer. The processing of relevant information occupies a lot of power grid dispatching time. Therefore, a voice question and answer processing method is urgently needed to solve the above problems. Summary of the invention

[0004] In view of this, the present invention provides a method and device for processing voice question and answer, the main purpose of which is to solve the problem of poor processing effect of existing voice question and answer.

[0005] According to one aspect of the present invention, a method for processing voice question and answer is provided, comprising:

[0006] Acquire voice data collected for power grid dispatching services, and convert the voice data to obtain voice text data;

[0007] Performing vector conversion on the speech text data to obtain question vector data, and determining at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension;

[0008] A voice feedback result is generated based on the dimensional correlation integration between the question vector data and the business dimension, and the voice feedback result is output based on a pre-configured virtual playback object.

[0009] Furthermore, before acquiring the voice data collected for the power grid dispatching service, the method further includes:

[0010] Retrieving power grid business data collected from different collection sources, and associating the power grid business data with corresponding business dimensions according to the power grid dispatching dimensions matched by the collection sources;

[0011] The power grid business data is stored according to the business dimension.

[0012] Further, storing the power grid service data according to the service dimension includes:

[0013] The power grid business data is converted into business vector data and stored as the business vector data, wherein the business vector data tv is determined based on the functional relationship between the number of occurrences of words in the power grid business data under different business dimensions, the total number of words, the total amount of data, and the quantized value of each power grid business data.

[0014] Furthermore, the method further comprises:

[0015] Establishing correlations between the multiple business dimensions, wherein the correlations are expressed as:

[0016] D=(∑(if(|tv i -tv j |≤α,1,0))) / S;

[0017] Among them, tv i ,tv j is each business vector data under two different business dimensions i and j, α is a preset vector threshold parameter, and S is the total number of business vectors in the business vector data dimension of two different business dimensions i and j; if D≥β, there is an association between the two different business dimensions i and j, otherwise, there is no association; wherein β is a preset association threshold parameter.

[0018] Furthermore, the step of performing vector conversion on the speech text data to obtain question vector data includes:

[0019] Segmenting the speech text information into words, and counting the total number of key words obtained after segmentation;

[0020] Perform vector conversion based on the total number of times to obtain the problem vector data;

[0021] Wherein, the problem vector data qv is expressed as:

[0022] N k is the number of times each word k appears in the speech text information, tv f,n The business vector data of each business dimension n, X is the total number of business vector data under each business dimension, B is the total number of dimensions of the business dimension, and M is the total number of key words in the voice text information.

[0023] Further, the generating of the voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension includes:

[0024] The problem vector data is compared with the business vector data corresponding to the target business dimension in the business dimension according to the constraint condition, and the constraint condition is:

[0025] tv n is each business vector data under each business dimension n, u is a preset vector threshold parameter, and X is the total number of business vector data under each business dimension;

[0026] When the constraint condition is matched, the words corresponding to the business vector data are integrated to obtain a voice feedback result;

[0027] When the constraint condition is not matched, the problem vector data is re-compared with the business vector data corresponding to other business dimensions in the business dimension to determine the business vector data matching the constraint condition.

[0028] Furthermore, the outputting the voice feedback result based on the pre-configured virtual playback object includes:

[0029] At least one configured virtual playback object is retrieved, and the voice feedback result is played according to at least one of a playback voice mode and a playback scene mode corresponding to a target virtual playback object selected by a user.

[0030] According to another aspect of the present invention, a voice question and answer processing device is provided, comprising:

[0031] An acquisition module, used to acquire voice data collected for power grid dispatching services, and convert the voice data to obtain voice text data;

[0032] A determination module, configured to perform vector conversion on the speech text data to obtain question vector data, and determine at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension;

[0033] An output module is used to generate a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension, and output the voice feedback result based on a pre-configured virtual playback object.

[0034] Furthermore, the device also includes:

[0035] A retrieval module, used to retrieve power grid business data collected from different collection sources, and associate the power grid business data with corresponding business dimensions according to the power grid dispatching dimensions matched by the collection sources;

[0036] A storage module is used to store the power grid business data according to the business dimension.

[0037] Furthermore, the storage module is specifically used to convert the power grid business data into business vector data and store it as the business vector data, wherein the business vector data tv is determined based on the functional relationship between the number of occurrences of words in the power grid business data under different business dimensions, the total number of words, the total amount of data, and the quantized value of each power grid business data.

[0038] Furthermore, the device also includes:

[0039] An establishing module is used to establish the correlation between the plurality of business dimensions, wherein the correlation is expressed as:

[0040] D=(∑(if(|tv i -tv j |≤α,1,0))) / S;

[0041] Among them, tv i ,tv j is each business vector data under two different business dimensions i and j, α is a preset vector threshold parameter, and S is the total number of business vectors in the business vector data dimension of two different business dimensions i and j; if D≥β, there is an association between the two different business dimensions i and j, otherwise, there is no association; wherein β is a preset association threshold parameter.

[0042] Furthermore, the determination module is specifically used to perform word segmentation on the speech text information, and to count the total number of key words obtained after segmentation; perform vector conversion based on the total number of times to obtain the question vector data;

[0043] Wherein, the problem vector data qv is expressed as:

[0044] N k is the number of times each word k appears in the speech text information, tv f,n The business vector data of each business dimension n, X is the total number of business vector data under each business dimension, B is the total number of dimensions of the business dimension, and M is the total number of key words in the voice text information.

[0045] Furthermore, the output module is specifically used to compare the problem vector data with the business vector data corresponding to the target business dimension in the business dimension according to the constraint condition, and the constraint condition is: tv nis each business vector data under each business dimension n, u is a preset vector threshold parameter, and X is the total number of business vector data under each business dimension; when the constraint condition is matched, the words corresponding to the business vector data are integrated to obtain a voice feedback result; when the constraint condition is not matched, the problem vector data is re-compared with the business vector data corresponding to other business dimensions in the business dimension to determine the business vector data that matches the constraint condition.

[0046] Furthermore, the output module is also specifically used to call at least one configured virtual playback object, and play the voice feedback result according to at least one of the playback voice mode and playback scene mode corresponding to the target virtual playback object selected by the user.

[0047] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to perform operations corresponding to the above-mentioned voice question and answer processing method.

[0048] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0049] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned voice question and answer processing method.

[0050] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0051] The present invention provides a method and device for processing voice question and answer. Compared with the prior art, the embodiment of the present invention obtains voice data collected for power grid dispatching business, and converts the voice data to obtain voice text data; performs vector conversion on the voice text data to obtain question vector data, and determines at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension; generates a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension, and outputs the voice feedback result based on a pre-configured virtual playback object, thereby achieving a more accurate power grid dispatching question and answer effect, increasing the rapid question and answer matching of power grid dispatching related businesses, shortening the working cycle time, and reducing the time for processing voice data during power grid dispatching, thereby meeting the processing requirements of different power grid dispatches based on voice question and answer.

[0052] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0054] Figure 1 A flow chart of a method for processing voice question and answer provided by an embodiment of the present invention is shown;

[0055] Figure 2 A schematic diagram of the structure of a power grid dispatching question-and-answer system provided by an embodiment of the present invention is shown;

[0056] Figure 3 A block diagram showing the composition of a voice question and answer processing device provided by an embodiment of the present invention;

[0057] Figure 4 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0058] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0059] The embodiment of the present invention provides a method for processing voice question and answer, such as Figure 1 As shown, the method includes:

[0060] 101. Acquire voice data collected for power grid dispatching services, and convert the voice data to obtain voice text data.

[0061] In the embodiment of the present invention, the current execution end serves as a server end for executing voice question and answer processing in the power grid dispatching service. It can be a cloud server or a terminal server. The power grid dispatching service is a service for dispatching electricity for different power transmission networks in the power system. Correspondingly, when the user consults about the power grid dispatching service, he can access the current execution end through a communication network, such as a telephone network or an Internet network. At this time, the current execution end processes the input voice data as a server end of the intelligent voice question and answer robot. The embodiment of the present invention does not make specific limitations. Among them, the voice data is a segment of voice or multiple segments of voice entered by the user when consulting the power grid dispatching service. Furthermore, the current execution end converts the voice data into voice text data based on the voice conversion technology, thereby processing the voice text data. The embodiment of the present invention does not make specific limitations.

[0062] 102. Perform vector conversion on the speech text data to obtain question vector data, and determine at least one business dimension of the question vector data.

[0063] In an embodiment of the present invention, after the current execution end obtains the voice text data, it converts the text data into a vector based on natural language processing technology to obtain question vector data. At this time, the question vector data is used to characterize the numerical vector of the problem generated by the user's consultation on the power grid dispatching service, so as to match the corresponding business dimension based on this question vector data. Among them, the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension, that is, the power grid dispatching dimension is used to characterize the dimension that can be used to divide the types of power grid business, including but not limited to the equipment type dimension, the jurisdiction part dimension, the task urgency dimension, etc., and then the current execution end divides the power grid business data according to the power grid dispatching dimension in advance, and obtains the business dimensions associated with the power grid business data of different classification dimensions, including but not limited to electricity charges, electricity usage duration, power outage distribution, etc., which are not specifically limited in the embodiment of the present invention.

[0064] 103. Generate a voice feedback result based on the dimensional correlation integration between the question vector data and the business dimension, and output the voice feedback result based on a pre-configured virtual playback object.

[0065] In an embodiment of the present invention, after determining the business dimension, the current execution end generates a voice feedback result based on the dimensional correlation between the question vector data and the business dimension. Among them, since the question vector data can be matched to at least one business dimension, in order to prepare to provide an effective reply, the current execution end calculates the dimensional correlation between the question vector data and multiple business dimensions so as to integrate and obtain the voice feedback result. In addition, in order to show a variety of output effects, the current execution end schedules a pre-configured virtual playback object and outputs the voice feedback result according to the playback mode of this virtual playback object. At this time, the virtual playback object includes but is not limited to a virtual voice playback object, a virtual visual playback object, etc. Different virtual playback objects can be configured with different playback modes, such as different voices, intonations, etc., which are not specifically limited in the embodiment of the present invention.

[0066] In another embodiment of the present invention, for further definition and explanation, before the step of acquiring the voice data collected for the power grid dispatching service, the method further includes:

[0067] Retrieving power grid business data collected from different collection sources, and associating the power grid business data with corresponding business dimensions according to the power grid dispatching dimensions matched by the collection sources;

[0068] The power grid business data is stored according to the business dimension.

[0069] In order to meet the multi-dimensional division of voice data to match accurate reply content, the current execution end first retrieves power grid business data from different collection sources. At this time, the collection source includes but is not limited to the processing system, transmission channel, etc., which is not specifically limited in the embodiment of the present invention. Among them, power grid business data includes but is not limited to all data generated in the power grid dispatching business, such as electricity, power supply time, etc., and then, the power grid dispatching dimension is matched according to the collection source, that is, different collection sources can be pre-configured with different power grid dispatching dimensions, such as equipment type dimension, jurisdiction dimension, task urgency dimension, etc., which is not specifically limited in the embodiment of the present invention. In addition, business dimensions include but are not limited to electricity charges, power consumption time, power outage distribution, etc., which can be matched based on the mapping relationship between the pre-configured power grid dispatching dimension and the business dimension to determine that different power grid dispatching dimensions are associated with the business data of the network, and the business dimension is determined to store the power grid business data.

[0070] In another embodiment of the present invention, for further definition and explanation, the step of storing the power grid service data according to the service dimension includes:

[0071] The power grid service data is converted into service vector data, and stored as the service vector data.

[0072] In order to improve storage efficiency, when storing power grid business data according to business dimensions, specifically, the power grid business data needs to be converted into business vector data. At this time, the business vector data tv is determined based on the functional relationship between the number of occurrences of words in the power grid business data under different business dimensions, the total number of words, the total amount of data, and the quantized value of each power grid business data. In a specific implementation scenario, the business vector data tv is expressed as:

[0073]

[0074] Among them, C w is the number of times each word w appears in each power grid dispatching data under each business dimension, tx is each power grid dispatching data in all business dimensions, A is the total number of power grid dispatching data in all business dimensions, Z is the total number of words in each power grid dispatching data under each business dimension, if(x,y,z) means that if condition x is met, result x is returned, otherwise result z is returned.

[0075] In another embodiment of the present invention, for further definition and description, the steps further include:

[0076] Establishing correlations between the plurality of business dimensions.

[0077] In order to reflect whether the business dimensions between different power grid dispatching data are related, the reply content matching the voice data is retrieved based on the correlation. Therefore, after the current execution end determines multiple business dimensions, the correlation between the multiple business dimensions is established. The correlation is expressed as:

[0078] D=(∑(if(|tv i -tv j |≤α,1,0))) / S;

[0079] Among them, tv i ,tv j is each business vector data under two different business dimensions i and j, α is a preset vector threshold parameter, and S is the total number of business vectors in the business vector data dimension of two different business dimensions i and j; if D≥β, there is an association between the two different business dimensions i and j, otherwise, there is no association; wherein β is a preset association threshold parameter.

[0080] In another embodiment of the present invention, for further definition and explanation, the step of performing vector conversion on the speech text data to obtain question vector data includes:

[0081] Segmenting the speech text information into words, and counting the total number of key words obtained after segmentation;

[0082] Vector conversion is performed based on the total number of times to obtain the problem vector data.

[0083] Because in the application process, in order to accurately extract useful consulting questions from the voice content, the current execution end converts the voice text data into vectors to obtain the question vector data. Specifically, the voice text information is first segmented into words. At this time, the word segmentation can be based on the meaning of the word in the natural language processing technology to obtain words such as modal particles, conjunctions, names, etc., which are not specifically limited in the embodiment of the present invention. At the same time, in order to avoid useless recognition, the current execution end can also delete meaningless words such as modal particles and conjunctions, retain key words, and then count the total number of key words after segmentation to perform vector conversion according to the total number of times. Specifically, the question vector data qv is expressed as:

[0084] N k is the number of times each word k appears in the speech text information, tv f,n The business vector data of each business dimension n, X is the total number of business vector data under each business dimension, B is the total number of dimensions of the business dimension, and M is the total number of key words in the voice text information.

[0085] In another embodiment of the present invention, for further definition and explanation, the step of generating a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension includes:

[0086] Comparing the problem vector data with the business vector data corresponding to the target business dimension in the business dimension according to the constraint condition;

[0087] When the constraint condition is matched, the words corresponding to the business vector data are integrated to obtain a voice feedback result;

[0088] When the constraint condition is not matched, the problem vector data is re-compared with the business vector data corresponding to other business dimensions in the business dimension to determine the business vector data matching the constraint condition.

[0089] In order to improve the matching between question-answer vector data and business dimensions, and thus improve the accuracy and effectiveness of matching reply content for voice data, the current execution end, when performing dimension correlation integration to generate voice feedback results, specifically, compares the question vector data with the business vector data corresponding to the target business dimension in the business dimension according to the constraint conditions. Among them, the constraint conditions are:

[0090] tv nis each business vector data under each business dimension n, u is a preset vector threshold parameter, and X is the total number of business vector data under each business dimension. In a specific implementation scenario, when the constraint condition is matched, the words corresponding to the business vector data are integrated to obtain a voice feedback result. When the constraint condition is not matched, the problem vector data is re-compared with the business vector data corresponding to other business dimensions in the business dimension to determine the business vector data that matches the constraint condition. At this time, the other business dimensions are the uncompared business dimensions relative to the compared business dimensions. After all businesses are compared or the constraint condition is matched, the words corresponding to the matched business vector data are integrated to obtain a voice feedback result. Among them, when integrating, the words can be combined and spliced ​​according to the language logical order of subject, predicate, object, etc. to obtain a complete sentence for feedback, which is not specifically limited in the embodiment of the present invention.

[0091] In another embodiment of the present invention, for further definition and explanation, the step of outputting the voice feedback result based on the pre-configured virtual playback object includes:

[0092] At least one configured virtual playback object is retrieved, and the voice feedback result is played according to at least one of a playback voice mode and a playback scene mode corresponding to a target virtual playback object selected by a user.

[0093] In order to meet the diversified display efficiency of voice questions and answers, thereby enhancing the playback effect of voice questions and answers, when the current execution end outputs the voice feedback result, specifically, first call at least one configured virtual playback object. At this time, the virtual playback object can be a virtual playback sound constructed by three-dimensional virtual technology, or a virtual playback sound with different voice intonations can be configured based on intelligent voice technology, thereby forming a realistic and flexible playback sound effect. After calling the virtual playback object, the user can output it to the user for selection to play as the target virtual playback object. Among them, different virtual playback objects are pre-configured with playback voice mode, playback image mode and playback scene mode, so that after determining the target virtual playback object, the voice feedback result is played according to at least one of the corresponding playback voice mode and playback scene mode. Among them, the playback voice mode is used to characterize the voice playback speed and intonation of the voice effect, including but not limited to the child's voice playback mode, the woman's voice playback mode, etc., and the playback scene mode is used to characterize the scene mode of voice playback, including but not limited to positive scenes, happy scenes, sad scenes, etc., which are not specifically limited in the embodiments of the present invention. In addition, if the voice feedback result is output through an image in the embodiment of the present invention, the virtual playback object may also be a virtual portrait, such as a virtual anchor outputting a corresponding voice feedback result, which is not specifically limited in the embodiment of the present invention.

[0094] In another embodiment of the present invention, the voice question and answer processing method of the present invention is embedded in a computer system and may include: Figure 2 Each virtual unit in the system structure diagram shown in the figure is used to execute the corresponding method. For example, the system includes: a power grid dispatching data acquisition unit, a database unit, a data preprocessing unit, a voice acquisition and recognition unit, a voice data processing unit, a comprehensive analysis and processing unit, a voice synthesis unit, and a virtual image construction unit; wherein:

[0095] Power grid dispatching data collection unit: used to collect business information data related to power grid dispatching;

[0096] Database unit: used to associate and store the business information data related to power grid dispatching collected by the power grid dispatching data collection unit according to preset dimensions;

[0097] Data preprocessing unit: used for preprocessing operations of converting and mapping the business information data stored in the database unit;

[0098] Voice collection and recognition unit: used to collect the user's voice and convert the collected user's voice recognition into text information data;

[0099] Speech data processing unit: used to convert the speech text information data converted by the speech acquisition and recognition unit, thereby converting it into question vector data;

[0100] Comprehensive analysis processing unit: used to match dimensions with high conformity according to the question vector data converted by the voice data processing unit, and integrate several dimensions with high conformity to form feedback results;

[0101] Speech synthesis unit: used for synthesizing speech according to the feedback result formed by the comprehensive analysis and processing unit;

[0102] Virtual image construction unit: used to construct a virtual image that meets the requirements as needed, and broadcast the voice synthesized by the voice synthesis unit through the virtual image.

[0103] Furthermore, the database unit includes: a dimension presetting module, a corresponding association module and a data storage module;

[0104] Dimension presetting module: used to set different grid dispatch dimensions according to the source of grid dispatch related business data, including but not limited to the equipment type targeted by grid dispatch related business data, the jurisdiction department of grid dispatch related business, and the urgency of dispatch tasks of grid dispatch related business;

[0105] Corresponding association module: used to associate the collected power grid dispatching data with different business dimensions;

[0106] Data storage module: used to store different business dimensions and associated power grid dispatching data under different business dimensions.

[0107] The data preprocessing unit is composed of a data retrieval module, a data conversion module and a return mapping module;

[0108] Data retrieval module: used to retrieve power grid dispatching data associated with different business dimensions from the database unit;

[0109] Data conversion module: used to convert the power grid dispatching data associated with different business dimensions retrieved by the data retrieval module into business vector data;

[0110] Feedback mapping module: used to map the business vector data converted by the data conversion module to the power grid dispatching data under the corresponding dimension.

[0111] When converting the power grid dispatching data into the business vector data, the data conversion module converts each power grid dispatching data in each dimension retrieved by the data retrieval module into the business vector data tv.

[0112] After the data conversion module converts each power grid dispatching data under each dimension into business vector data, it establishes the correlation degree D between several business dimensions.

[0113] When the speech data processing unit converts the speech text information data into question vector data, it first divides the speech text information data into several words, removes the modal particles, conjunctions and meaningless words, then obtains the number of occurrences of each retained word, and then converts the speech text information data into question vector data qv.

[0114] The comprehensive analysis and processing unit includes: a data acquisition and matching module, a feedback result generation module, and a feedback result forwarding module;

[0115] Data acquisition and matching module: used to match dimensions with high consistency with the question vector data from the database unit according to the question vector data;

[0116] Feedback result generation module: used to integrate several dimensions matched by the data acquisition and matching module with high consistency with the problem vector data to form feedback results;

[0117] Feedback result forwarding module: used to forward the feedback result generated by the feedback result generation module to the speech synthesis unit.

[0118] In the process of matching dimensions with high consistency with the problem vector data from the database unit, the data acquisition and matching module compares the problem vector data with each business vector data under each dimension, and then determines that the current dimension has a high degree of consistency with the problem vector data. At this time, the comparison with the business vector data under other dimensions is stopped, and other dimensions associated with the current dimension are obtained based on the correlation between the dimensions.

[0119] In a specific scenario, the speech synthesis unit selects a voice that is suitable for the style and personality of the virtual anchor of the power grid dispatching to perform speech synthesis, thereby generating the answer audio of the virtual anchor of the power grid dispatching. The virtual image construction unit uses the audio results synthesized by the speech synthesis unit, lip prediction technology, face recognition technology, and virtual image power grid dispatching virtual anchor generation technology to drive the power grid dispatching virtual anchor to speak and generate picture frames and audio frames, thereby completing the construction of the virtual image.

[0120] The embodiment of the present invention provides a method for processing voice question and answer. Compared with the prior art, the embodiment of the present invention obtains voice data collected for power grid dispatching business and converts the voice data to obtain voice text data; performs vector conversion on the voice text data to obtain question vector data, and determines at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension; generates a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension, and outputs the voice feedback result based on a pre-configured virtual playback object, thereby achieving a more accurate power grid dispatching question and answer effect, increasing the rapid question and answer matching of power grid dispatching related businesses, shortening the working cycle time, and reducing the time for processing voice data during power grid dispatching, thereby meeting the processing requirements of different power grid dispatches based on voice question and answer.

[0121] Furthermore, as a response to the above Figure 1 The embodiment of the present invention provides a voice question and answer processing device, such as Figure 3 As shown, the device comprises:

[0122] The acquisition module 21 is used to acquire voice data collected for power grid dispatching services and convert the voice data to obtain voice text data;

[0123] A determination module 22 is used to perform vector conversion on the speech text data to obtain question vector data, and determine at least one business dimension of the question vector data, where the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension;

[0124] The output module 23 is used to generate a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension, and output the voice feedback result based on a pre-configured virtual playback object.

[0125] Furthermore, the device also includes:

[0126] A retrieval module, used to retrieve power grid business data collected from different collection sources, and associate the power grid business data with corresponding business dimensions according to the power grid dispatching dimensions matched by the collection sources;

[0127] A storage module is used to store the power grid business data according to the business dimension.

[0128] Furthermore, the storage module is specifically used to convert the power grid business data into business vector data and store it as the business vector data, wherein the business vector data tv is determined based on the functional relationship between the number of occurrences of words in the power grid business data under different business dimensions, the total number of words, the total amount of data, and the quantized value of each power grid business data.

[0129] Furthermore, the device also includes:

[0130] An establishing module is used to establish the correlation between the plurality of business dimensions, wherein the correlation is expressed as:

[0131] D=(∑(if(|tv i -tv j |≤α,1,0))) / S;

[0132] Among them, tv i ,tv j is each business vector data under two different business dimensions i and j, α is a preset vector threshold parameter, and S is the total number of business vectors in the business vector data dimension of two different business dimensions i and j; if D≥β, there is an association between the two different business dimensions i and j, otherwise, there is no association; wherein β is a preset association threshold parameter.

[0133] Furthermore, the determination module is specifically used to perform word segmentation on the speech text information, and to count the total number of key words obtained after segmentation; perform vector conversion based on the total number of times to obtain the question vector data;

[0134] Wherein, the problem vector data qv is expressed as:

[0135] N k is the number of times each word k appears in the speech text information, tv f,nThe business vector data of each business dimension n, X is the total number of business vector data under each business dimension, B is the total number of dimensions of the business dimension, and M is the total number of key words in the voice text information.

[0136] Furthermore, the output module is specifically used to compare the problem vector data with the business vector data corresponding to the target business dimension in the business dimension according to the constraint condition, and the constraint condition is: tv n is each business vector data under each business dimension n, u is a preset vector threshold parameter, and X is the total number of business vector data under each business dimension; when the constraint condition is matched, the words corresponding to the business vector data are integrated to obtain a voice feedback result; when the constraint condition is not matched, the problem vector data is re-compared with the business vector data corresponding to other business dimensions in the business dimension to determine the business vector data that matches the constraint condition.

[0137] Furthermore, the output module is also specifically used to call at least one configured virtual playback object, and play the voice feedback result according to at least one of the playback voice mode and playback scene mode corresponding to the target virtual playback object selected by the user.

[0138] The embodiment of the present invention provides a voice question and answer processing device. Compared with the prior art, the embodiment of the present invention obtains voice data collected for power grid dispatching business and converts the voice data to obtain voice text data; performs vector conversion on the voice text data to obtain question vector data, and determines at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension; generates a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension, and outputs the voice feedback result based on a pre-configured virtual playback object, thereby achieving a more accurate power grid dispatching question and answer effect, increasing the rapid question and answer matching of power grid dispatching related businesses, shortening the working cycle time, and reducing the time for processing voice data during power grid dispatching, thereby meeting the processing requirements of different power grid dispatches based on voice question and answer.

[0139] According to an embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the voice question and answer processing method in any of the above method embodiments.

[0140] Figure 4 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0141] like Figure 4 As shown, the terminal may include: a processor (processor) 302 , a communication interface (Communications Interface) 304 , a memory (memory) 306 , and a communication bus 308 .

[0142] The processor 302 , the communication interface 304 , and the memory 306 communicate with each other via the communication bus 308 .

[0143] The communication interface 304 is used to communicate with other devices such as clients or other servers.

[0144] The processor 302 is used to execute the program 310, and specifically can execute the relevant steps in the above-mentioned voice question and answer processing method embodiment.

[0145] Specifically, the program 310 may include program codes, which include computer operation instructions.

[0146] The processor 302 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0147] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0148] The program 310 may be specifically configured to enable the processor 302 to perform the following operations:

[0149] Acquire voice data collected for power grid dispatching services, and convert the voice data to obtain voice text data;

[0150] Performing vector conversion on the speech text data to obtain question vector data, and determining at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension;

[0151] A voice feedback result is generated based on the dimensional correlation integration between the question vector data and the business dimension, and the voice feedback result is output based on a pre-configured virtual playback object.

[0152] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0153] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing voice question and answer, characterized in that: include: Acquire voice data collected for power grid dispatching services, and convert the voice data to obtain voice text data; Performing vector conversion on the speech text data to obtain question vector data, and determining at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension; A voice feedback result is generated based on the dimensional correlation integration between the question vector data and the business dimension, and the voice feedback result is output based on a pre-configured virtual playback object.

2. The method according to claim 1, characterized in that Before acquiring the voice data collected for the power grid dispatching service, the method further includes: Retrieving power grid business data collected from different collection sources, and associating the power grid business data with corresponding business dimensions according to the power grid dispatching dimensions matched by the collection sources; The power grid business data is stored according to the business dimension.

3. The method according to claim 2, characterized in that The storing of the power grid service data according to the service dimension comprises: The power grid business data is converted into business vector data and stored as the business vector data, wherein the business vector data tv is determined based on the functional relationship between the number of occurrences of words in the power grid business data under different business dimensions, the total number of words, the total amount of data, and the quantized value of each power grid business data.

4. The method according to claim 3, characterized in that The method further comprises: Establishing correlations between the multiple business dimensions, wherein the correlations are expressed as: D=(∑(if(|tv i -tv j |≤α,1,0))) / S; Among them, tv i ,tv j is each business vector data under two different business dimensions i and j, α is a preset vector threshold parameter, and S is the total number of business vectors in the business vector data dimension of two different business dimensions i and j; if D≥β, there is an association between the two different business dimensions i and j, otherwise, there is no association; wherein β is a preset association threshold parameter.

5. The method according to claim 1, characterized in that The performing vector conversion on the speech text data to obtain question vector data comprises: Segmenting the speech text information into words, and counting the total number of key words obtained after segmentation; Perform vector conversion based on the total number of times to obtain the problem vector data; Wherein, the problem vector data qv is expressed as: N k is the number of times each word k appears in the speech text information, tv f,n The business vector data of each business dimension n, X is the total number of business vector data under each business dimension, B is the total number of dimensions of the business dimension, and M is the total number of key words in the voice text information.

6. The method according to claim 1, characterized in that The generating of the voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension includes: The problem vector data is compared with the business vector data corresponding to the target business dimension in the business dimension according to the constraint condition, and the constraint condition is: tv n is each business vector data under each business dimension n, u is a preset vector threshold parameter, and X is the total number of business vector data under each business dimension; When the constraint condition is matched, the words corresponding to the business vector data are integrated to obtain a voice feedback result; When the constraint condition is not matched, the problem vector data is re-compared with the business vector data corresponding to other business dimensions in the business dimension to determine the business vector data matching the constraint condition.

7. The method according to any one of claims 1 to 6, characterized in that: The outputting the voice feedback result based on the pre-configured virtual playback object includes: At least one configured virtual playback object is retrieved, and the voice feedback result is played according to at least one of a playback voice mode and a playback scene mode corresponding to a target virtual playback object selected by a user.

8. A voice question and answer processing device, characterized in that: include: An acquisition module, used to acquire voice data collected for power grid dispatching services, and convert the voice data to obtain voice text data; A determination module, configured to perform vector conversion on the speech text data to obtain question vector data, and determine at least one business dimension of the question vector data, wherein the business dimension is obtained by parsing the collected power grid business data based on the power grid dispatching dimension; An output module is used to generate a voice feedback result based on the dimensional correlation integration of the question vector data and the business dimension, and output the voice feedback result based on a pre-configured virtual playback object.

9. A storage medium storing at least one executable instruction, wherein the executable instruction enables a processor to execute an operation corresponding to the voice question and answer processing method as described in any one of claims 1 to 7.

10. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform an operation corresponding to the voice question and answer processing method as described in any one of claims 1-7.

Images