CN115757718A

CN115757718A - Text generation method and device

Info

Publication number: CN115757718A
Application number: CN202211385349.2A
Authority: CN
Inventors: 李宜昌; 王杰; 金天龙; 周倩; 许阳; 左敬超; 王立鹏
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2022-11-07
Filing date: 2022-11-07
Publication date: 2023-03-07

Abstract

The embodiment of the specification provides a text generation method and a text generation device, wherein the text generation method comprises the following steps: obtaining a dialog text of the associated project participation behavior, and determining a target problem text corresponding to the dialog text; determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with a project participation behavior; determining a target answer text associated with the target question text based on the text to be processed; and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text. The method comprises the steps of determining a target question text according to a dialog text, introducing an auxiliary text rich information source corresponding to the dialog text, determining a text to be processed according to the auxiliary text and the dialog text, obtaining a target answer text with higher accuracy, enabling the fine granularity of the finally generated dialog summary text to be higher, enabling project personnel to know the intention of a project participating user more quickly based on the dialog summary text, and accordingly providing more convenient service for the user.

Description

Text generation method and device

Technical Field

The embodiment of the specification relates to the technical field of natural languages, in particular to a text generation method. One or more embodiments of the present specification relate to a text generation apparatus, a text generation system, a computing device, and a computer-readable storage medium.

Background

With the rapid development of technologies such as computers, networks and the like, a customer service center based on telephones and networks becomes an important way for interaction between enterprises and users, at present, after the customer service and the users complete conversation, conversation records are carried out according to preset conversation knots in a classified mode, the problems of the user in the consultation are summarized, the problems do not need to be described repeatedly when the user calls the customer service again subsequently, and the customer service can know the problems which the user wants to consult according to the last conversation knots. However, the above summary user dialog process often presents the following problems: after the consultation is finished, the client personnel spend a great deal of time filling out the summary records; the summary records inaccurate content and cannot play a role in the next consultation experience of the user. Therefore, how to quickly and accurately summarize the client communication content is a problem that needs to be solved urgently at present.

Disclosure of Invention

In view of this, the embodiments of the present specification provide a text generation method. One or more embodiments of the present specification also relate to a text generation apparatus, a text generation system, a computing device, a computer-readable storage medium, and a computer program, so as to solve technical deficiencies in the prior art.

According to a first aspect of embodiments of the present specification, there is provided a text generation method including:

obtaining a dialog text of a related project participation behavior, and determining a target problem text corresponding to the dialog text;

determining a text to be processed based on the conversation text and an auxiliary text corresponding to the conversation text, wherein the auxiliary text is associated with the project participation behavior;

determining a target answer text associated with the target question text based on the text to be processed;

and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text.

According to a second aspect of embodiments herein, there is provided a text generation method including:

receiving a text generation instruction submitted for a dialog text through a summary page, wherein the summary page is associated with a project participation action;

responding to the text generation instruction, determining a target question text corresponding to the conversation text, and determining a text to be processed based on the conversation text and an auxiliary text corresponding to the conversation text, wherein the auxiliary text is associated with the project participation behavior;

and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and displaying the dialog summary text through the summary page.

According to a third aspect of embodiments herein, there is provided a text generation method including:

acquiring a communication session text between a user and a customer service, and determining a target problem text corresponding to the communication session text;

determining a text to be processed based on the communication session text and an auxiliary text corresponding to the communication session text, wherein the auxiliary text is a communication template text used by the customer service;

and generating a customer service communication summary text corresponding to the communication dialogue text based on the target question text and the target answer text.

According to a fourth aspect of embodiments herein, there is provided a text generation apparatus including:

the acquisition module is configured to acquire a dialog text of a related project participation behavior and determine a target problem text corresponding to the dialog text;

a first determination module configured to determine a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior;

a second determination module configured to determine a target answer text associated with the target question text based on the text to be processed;

a generating module configured to generate a dialog summary text corresponding to the dialog text based on the target question text and the target answer text.

According to a fifth aspect of embodiments herein, there is provided a text generation apparatus including:

a receiving module configured to receive a text generation instruction submitted for a dialog text through a summary page, wherein the summary page is associated with a project participation action;

a first determining module, configured to determine, in response to the text generation instruction, a target question text corresponding to the dialog text, and determine a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, where the auxiliary text is associated with the project participation behavior;

and the generating module is configured to generate a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and display the dialog summary text through the summary page.

According to a sixth aspect of embodiments herein, there is provided a text generation apparatus including:

the system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is configured to acquire a communication session text between a user and a customer service and determine a target question text corresponding to the communication session text;

a first determining module, configured to determine a text to be processed based on the communication session text and an auxiliary text corresponding to the communication session text, where the auxiliary text is a communication template text used by the customer service;

and the generating module is configured to generate a customer service communication summary text corresponding to the communication conversation text based on the target question text and the target answer text.

Practice in accordance with the present specification in a seventh aspect of the present invention, there is provided a text generation system including:

the client is used for storing the text display executable instruction, and the server is used for storing the text generation executable instruction; the text display executable instructions when executed by the client and the text generation executable instructions when executed by the server implement the steps of the text generation method described above.

According to an eighth aspect of embodiments herein, there is provided a computing device comprising a memory, a processor and computer instructions stored on the memory and executable on the processor, the processor implementing the steps of the text generation method when executing the computer instructions.

According to a ninth aspect of embodiments herein, there is provided a computer readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the text generation method.

According to a tenth aspect of embodiments of the present specification, there is provided a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-described text generation method.

The text generation method provided by the specification comprises the steps of obtaining a conversation text of a related project participation behavior, and determining a target problem text corresponding to the conversation text; determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior; determining a target answer text associated with the target question text based on the text to be processed; and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text.

According to the embodiment of the description, the corresponding target question text is determined according to the conversation text, the auxiliary text corresponding to the conversation text is introduced to enrich the information source, the text to be processed is determined according to the auxiliary text and the conversation text, the target answer text with higher accuracy can be obtained, the fine granularity of the finally generated conversation summary text is higher, a project worker can quickly know the intention of a project participating user based on the conversation summary text, and therefore more convenient service is provided for the user.

Drawings

Fig. 1 is an effect schematic diagram of a text generation method provided in an embodiment of the present specification

FIG. 2 is a flow chart of a method for generating text provided in one embodiment of the present specification;

FIG. 3 is a flow chart of a method for generating text that is provided in another embodiment of the present description;

FIG. 4 is a flowchart illustrating a process of a text generation method according to an embodiment of the present disclosure;

FIG. 5 is a flow chart of a text generation method provided in another embodiment of the present description;

fig. 6 is a schematic structural diagram of a text generation apparatus according to an embodiment of the present specification;

fig. 7 is a schematic structural diagram of a text generating apparatus according to another embodiment of the present specification;

fig. 8 is a schematic structural diagram of a text generation apparatus according to another embodiment of the present specification;

fig. 9 is a block diagram of a computing device according to an embodiment of the present disclosure.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present description. This description may be implemented in many ways other than those specifically set forth herein, and those skilled in the art will appreciate that the present description is susceptible to similar generalizations without departing from the scope of the description, and thus is not limited to the specific implementations disclosed below.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification is intended to encompass any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if," as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination," depending on the context.

First, the noun terms to which one or more embodiments of the present specification relate are explained.

Service summary: after the customer service finishes the customer incoming line consultation, the customer service needs to summarize and describe the current consultation, and after the customer enters the line again, the new customer service can quickly position the user problem and process the progress, so that the repeated description of the user is reduced, and the user experience is improved.

Service abstraction: when the user enters the line, the service abstract describes the general problems and appeal of the current user, and if the processing progress exists, the service abstract also comprises the processing progress of the current user.

And (4) SOP: the standard operation program, the SOP refers to the problem solving process operated by customer service personnel.

ISO: the digital solution is based on SOP's upgrading, more intelligent more convenient operation procedure.

In the service field, after a user communicates with a customer service, the customer service can summarize and describe the communication to generate a service summary, and when the service summary is used for the next consultation of the user, the next customer service providing the consultation service for the user can quickly know the historical problems and the processing progress of the current user based on the service summary, so that the repeated description of the user is reduced, and the user experience is improved. The generation of service summaries currently requires overcoming two types of problems. Firstly, the service bar filling takes long time, service personnel need to summarize and precipitate the communication record after the service is finished, and the user can assist the next connection customer service to quickly locate and solve the problem when inquiring again. Secondly, the accuracy of filling in the service summary is low, and the prior service summary is manually filled or summarized by a corpus identification model, so that the filling information of the service summary is inaccurate, the deviation of the customer service intention and appeal understanding of the user is easily caused, and the consultation experience of the user is damaged. A service summary filling system in the existing customer service system adopts a corpus identification model to identify enumerated values, the method cannot cover rich actual service scenes, the identification problem is inaccurate, the filling content is coarse in granularity, and accurate information cannot be provided for subsequent customer service personnel.

Based on this, in the present specification, a text generation method is provided for quickly generating a summary and abstract of a service with higher accuracy, thereby providing support for customer service to quickly locate and solve a problem, so that a user can be better served.

Fig. 1 is an effect schematic diagram of a text generation method according to an embodiment of the present disclosure, in fig. 1, a project person may be understood as a customer service person, and after the customer service person completes a communication service with a user, the customer service person may check a communication record, that is, a dialog text, through a client, such as a computer terminal in fig. 1, where in an actual application, the dialog text may be obtained by communication audio conversion, and a customer service system on the client queries, on the basis of obtaining the dialog text, whether the customer service performs a service according to a standard operation flow in the communication process, if so, an auxiliary text corresponding to the dialog text may be obtained, where information such as a customer problem, a processing mode, a processing progress and the like is recorded in the auxiliary text in detail, and then a dialog summary text may be directly generated according to contents in the auxiliary text, that is a service summary; if not, corpus recognition can be carried out based on the conversation text, and a conversation summary text is generated according to the recognition result. As shown in fig. 1, if the dialog summary text is generated based on the dialog text, the question text "merchant does not return goods" related to the original speech of the user in the dialog text may be recognized, so that the target question text "quality question" may be generated, the merchant does not return goods ", and the target answer text" customer service intervention processing, waiting for patience processing "may be recognized. In practical applications, the dialog summary text may further include other information, such as basic account information of the user, communication information between the customer service and the merchant, and the like. By the text generation method provided by the specification, the dialogue summary text with higher retention rate can be generated under the condition of introducing the auxiliary text with higher accuracy, so that a project worker does not need to adjust the dialogue summary text greatly subsequently, and the service efficiency of the project worker and the use experience of a user are improved.

Fig. 2 shows a flowchart of a text generation method provided according to an embodiment of the present specification, including step 202 to step 208.

Step 202: and acquiring a dialog text of the participation behavior of the associated project, and determining a target problem text corresponding to the dialog text.

The project participation behavior may be understood as a behavior of a user participating in a project provided by a project party, such as a behavior of a user performing online shopping and a behavior of a user purchasing tickets online, and when the participation behavior of the user is questioned after the user participates in the project, a consultation may be performed to a project staff, where the project staff may be a customer service staff of a project platform providing the project, or a service staff of each service party in the project platform. Taking online shopping as an example, a user purchases a piece of clothes on the platform a, but the clothes are not delivered, the item participation behavior is that the user purchases a commodity on the platform a, and the user wants to ask about the problem of commodity delivery at this time, and can ask about a shop attendant selling the commodity, or directly ask about a customer attendant on the platform a, and after the two parties communicate, a communication record is left. When the user communicates with the customer service staff on line through the terminal, the communication record can be directly in a text form, when the user communicates with the customer service staff through the terminal in a telephone mode, the communication record can be in an audio form, and voice recognition can be carried out on the communication audio subsequently to convert the communication audio into the text form. Whatever the communication mode, the conversation text related to the project participation behavior can be generated finally, so that the subsequent service summary, namely the conversation summary text, can be generated continuously. The dialog summary text may include a target question text, and the target question text may be understood as a question that the user wants to ask this time, such as that no goods are shipped.

In specific implementation, in order to make the filling content fine-grained and the content accurate, the target problem text includes a problem type and an original problem text, that is, the original words of the user. In practical applications, the purpose of the user to communicate with the customer service is to ask about related questions, such as asking about specific details about the product before purchasing the product, such as the number of pixels of the mobile phone; or after purchasing the product, ask a question about the product after sale, such as how the mobile phone cannot be turned on. Therefore, a plurality of dialog texts may exist in the project participation behavior of one user, and before a customer service person communicates with the user, in order to improve the communication efficiency, the customer service person can know the intention in advance based on the service summary, quickly locate the user problem, and know the information such as the processing progress of the problem before the user. Generating the dialog summary text based on the dialog text can therefore improve the user experience.

In one embodiment of the present description, a user who wants to purchase a ticket for an amusement park asks the relevant customer service personnel specific questions about the ticket, such as how the ticket can be used during the work day, how the ticket can play all items in the park, and the like. After the user and the customer service staff finish the communication, the customer service system can automatically acquire the conversation text of the communication and determine the target problem text corresponding to the conversation text, wherein the target problem text is ' ticket use ', and the ticket can be used in working days '.

Further, if the user's intention is judged only according to the enumerated questions, the intention of the user may not be accurately recognized, for example, when the user wants to ask about the goods for shipment, but the enumerated questions do not have an enumerated value related to the issue time problem, the recognized question may only be the shipment time of the goods, which results in recognition errors, and subsequent customer service staff may misunderstand the intention of the user. Therefore, the question text should include a question text described in the original speech of the user, so as to accurately describe the question consulted by the user, and specifically, determining the target question text corresponding to the dialog text includes: selecting an initial question text in a question text database based on the dialog text; determining an original question text in the dialog text according to the initial question text; and merging the initial problem text and the original problem text, and obtaining a target problem text according to a merging result.

The initial problem text can be understood as a problem set in advance, a problem of historical inquiry is stored in the problem text database, namely, the enumeration value is exhausted manually, the initial problem text is a problem category, but the granularity of the initial problem text is coarse, and the real problem of a user may not be accurately described. If the initial question text is determined to be returned, but what the user wants to be returned and what the user returns cannot be known according to the initial question text, the original question text can be supplemented, and fine-grained question text is generated based on the initial question text and the original question text, so that the filling content is more accurate. The original question text can be understood as the text describing the question in the user's original language, e.g. i want to go back to the phone because i buy the wrong model.

In practical application, in order to enable the filling content to be more accurate and restore key information in the service process, the original words of the user can be added into the problem text, and the customer service is helped to improve the working efficiency. After the initial question text is determined, a description text which best meets the question can be determined in the whole dialog text based on the initial question text, namely the initial question text is determined, and then the subsequent target question text generation is carried out.

In a specific embodiment of this specification, an initial problem text is selected from a problem text database based on a dialog text between a user and a customer service, the initial problem text is "ticket usage", an original problem text is selected from the dialog text according to the determined initial problem text, it is determined that the ticket of the original problem text "can be used on holidays", the initial problem text and the original problem text are combined, a target problem text is obtained according to a combination result, and the target problem text is "ticket usage: this ticket can be used on holidays.

In practical application, in order to solve the problem of low corpus recognition accuracy, the participation state information may be introduced to determine the initial problem text, and specifically, selecting the initial problem text in the problem text database based on the dialog text includes: acquiring participation state information corresponding to the conversation text; performing text screening on a problem text database based on the participation state information, and determining a problem text to be selected according to a screening result; and calculating the matching degree of each question text to be selected and the dialog text, and selecting the initial question text according to each matching degree.

The participation state information can be understood as state information of project participation behaviors, if a user consults before purchasing commodities, the participation state information is in a pre-sale state, pre-sale problem texts can be screened out in a problem text database based on the participation state information, the screened problem texts are to-be-selected problem texts, and then further selection can be carried out based on conversation texts, so that an initial problem text is determined.

In specific implementation, when the user makes a consultation after purchasing a commodity, the participation state information not only includes after-sale state information, but also includes specific order information, such as a text summarized by historical conversation or a text summarized by the user after the user has paid for and waits for delivery, and a text summarized by the user after the user communicates with customer service last time, so that a text of a problem to be selected can be screened out more accurately. After the problem text to be selected is determined, the matching degree between each problem text to be selected and the dialog text can be calculated, the matching degree can be understood as the association degree between the problem text and the dialog text, the higher the association degree is, the more the problem text is likely to meet the user intention, namely, the more the dialog text is associated, and the initial problem text can be determined according to the matching degree between each problem text to be selected and the dialog text.

In a specific embodiment of the present specification, the participation state information corresponding to the dialog text is obtained, and if the participation state information is a pre-sale state, the pre-sale problem may be first screened out from the problem text database, the problem text to be selected is determined according to the screening result, the matching degree between each problem text to be selected and the dialog text is calculated, the problem text to be selected with the highest matching degree is selected as the initial problem text, and the initial problem text is "ticket used".

Based on the method, the problem filling is carried out by introducing the original sound field of the user, and the combined field of the initial problem text and the original problem text is adopted for filling, so that the problem of the user can be accurately and specifically described.

Step 204: determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior.

The text to be processed can be understood as a text which needs to be identified and processed subsequently, the auxiliary text can be understood as an operation flow text of an SOP and ISO system, the auxiliary text comprises different flow nodes selected by the customer service in the service process, each flow node corresponds to a description text, and a processing scheme which is finally given to the user can be determined.

In practical application, a customer service has an ISO or SOP guidance system in the service process, which can be understood as a long tree structure, and according to different requirements and descriptions of users, the customer service can select different branch nodes to finally determine leaf nodes, namely, a solution provided for the users. The auxiliary text is the description text selected by the customer service staff in the process of communicating with the user through the user description step by step clicking the selection processing step, so that the auxiliary text is equivalent to a data source with higher priority relative to the conversation text, and the determined answer text can be directly determined according to the auxiliary text subsequently, so that the answer text does not need to be recognized from the conversation text, and the generation efficiency of the conversation summary text is improved.

In specific implementation, because a plurality of dialog texts may exist in a project participation behavior, and each dialog text has a corresponding auxiliary text, when the auxiliary text of the current dialog text is determined, the determination needs to be determined according to the current dialog text, and the specific determination of the auxiliary text includes: determining dialog identification information of the dialog text; and determining an auxiliary text corresponding to the dialog text according to the dialog identification information.

The dialog identification information can be understood as a unique identification of the dialog text, such as a unique ID generated according to the communication time and the user information, and the auxiliary text corresponding to the dialog text can be determined according to the dialog identification information.

In practical applications, since a customer service staff may not use an ISO operation flow or an SOP operation flow to communicate in order to better provide a service for a user, in this case, the dialog text has no auxiliary text, and it may also be understood that the dialog text is empty for moving the auxiliary text, but since the auxiliary text has a higher priority than the dialog text, when determining the answer text, it is first required to identify whether an answer text exists in the auxiliary text, and specifically, a text to be processed is determined based on the dialog text and the auxiliary text corresponding to the dialog text, including: and identifying the auxiliary text, and taking the auxiliary text as a text to be processed under the condition that answer texts exist in the auxiliary text according to the identification result.

In specific implementation, after the dialog text and the auxiliary text corresponding to the dialog text are determined, whether the answer text exists in the auxiliary text is identified, and since the auxiliary text is generated by a customer service worker in a process step of clicking, if the answer text exists in the auxiliary text, the answer text is directly used as a target answer text and is filled in the dialog summary text, and corpus identification is not required to be performed from the dialog text. Therefore, the auxiliary text is recognized, the auxiliary text is used as a text to be processed under the condition that the answer text exists in the auxiliary text according to the recognition result, and then the target answer text related to the target question text is determined based on the auxiliary text.

In another case, when there is no answer text in the auxiliary text, continuing to identify the dialog text, specifically, determining a text to be processed based on the dialog text and the auxiliary text corresponding to the dialog text, including: and identifying the auxiliary text, and taking the dialog text as a text to be processed under the condition that the auxiliary text is determined to have no answer text according to the identification result.

In specific implementation, when the customer service staff does not communicate with the user according to the standard operation flow and selects to input the reply content by himself, the auxiliary text of the dialog text is empty, and at this time, in order to determine the answer text, corpus recognition needs to be performed on the dialog text, and the dialog text is used as a text to be processed for subsequent recognition processing.

Step 206: and determining a target answer text associated with the target question text based on the text to be processed.

And determining a target answer text associated with the target question text according to the text to be processed determined under two different conditions.

In practical application, in the case that answer text exists in the auxiliary text, determining a target answer text associated with the target question text based on the text to be processed includes: and taking the answer text in the auxiliary text as the target answer text associated with the target question text. Namely, the answer text in the auxiliary text is directly used as the target question text.

In another case, in a case that no answer text exists in the auxiliary text, determining a target answer text associated with the target question text based on the text to be processed includes: and identifying the dialog text, determining an answer text in the dialog text according to an identification result, and taking the answer text in the dialog text as a target answer text associated with the target question text. Namely, the dialog text is used as a text to be processed, corpus recognition processing is carried out on the dialog text, an answer text is determined according to a recognition result, and therefore the answer text recognized in the dialog text is subsequently used as a target answer text related to the target question text.

In order to solve the problem of coarse granularity of past recognition, a preset recognition rule is adopted for recognition, specifically, recognition processing is carried out on the dialog text, an answer text in the dialog text is determined according to a recognition result, and the answer text in the dialog text is used as a target answer text related to the target question text, and the method comprises the following steps: inputting the conversation text into a guide model to perform guide recognition processing to obtain a guide answer text, and inputting the conversation text into an action model to perform action recognition processing to obtain an action answer text; and determining an answer text in a dialog text according to the guide answer text and the action answer text, and taking the answer text in the dialog text as a target answer text associated with the target question text.

The identification objects of the guide model and the action model are dialog texts, but the identification emphasis points of the guide model and the action model are different, the guide model is more inclined to identify what the customer service guide user does, if the customer service guide user inquires the logistics information, the action model is more inclined to identify what the customer service directly helps the user do, if the customer service directly inquires the logistics information for the user, the models in the river are combined, namely two identification rules are combined for combined identification, more service scenes can be combined, and the identification result is more accurate. In practical application, the result recognized by the two methods can be further processed to be used as an answer text in the dialog text, or a more accurate result can be selected to be used as the answer text in the dialog text according to the recognition result.

In a specific embodiment of the present specification, a dialog text is respectively input into a guidance model and an action model, the guidance recognition processing is performed on the dialog text by the guidance model, the action recognition processing is performed on the dialog text by the action model, a guidance answer text output by the guidance model and an action answer text output by the action model are obtained, the guidance answer text and the action answer text are combined to generate an answer text of the dialog text, the answer text is "the answer text is usable for any time period of the ticket", and the answer text is used as a target answer text associated with a target question text.

Step 208: and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text.

After the target question text and the target answer text are determined by the method, a dialog summary text corresponding to the current dialog text can be generated. In practical application, namely after the problem asked by the user and the solution provided by the customer service are determined, a service summary of the communication service can be generated based on the problem asked by the user and the solution provided by the customer service.

In order to enable the customer service to quickly locate and solve the problem, the problem text and the answer text can be combined to generate a conversation summary text, and in order to ensure the normalization of the conversation summary text, the problem text and the answer text can be filled in a preset conversation summary template. Specifically, generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text includes: determining participation progress information of the project participation behavior, determining a conversation summary template corresponding to the conversation text based on the participation progress information, writing the target question text and the target answer text into the conversation summary template, and generating a conversation summary text corresponding to the conversation text.

The participation progress information can be understood as the progress information of the current project participation behavior of the user, for example, the order progress information during online shopping includes information such as whether the transaction is completed, whether the user evaluates, and evaluation content.

In practical application, when a user consults a question at different participation schedules, the corresponding conversation summary templates may be different, and if the user consults the question when the transaction is not completed, the corresponding conversation summary template may include a transaction state text, a target question text, a target answer text, a merchant state text, and the like, and if the user consults the question after the transaction is completed, the corresponding conversation summary model may include a user evaluation text, a target question text, a target answer text, and the like. For different participation schedules, a dialog summary template for determining the current schedule of the current dialog text is needed, so that the problems encountered by the current schedule of the user can be described in more detail.

In a specific embodiment of the present specification, order information of a user online shopping behavior is determined, it is determined that a user has completed an order and made an evaluation, a conversation summary template of a conversation text is determined based on participation progress information, a target question text and a target answer text are written into the conversation summary template, and the conversation summary template further includes a satisfaction evaluation of the user for the current online shopping, so that subsequent customer service can quickly learn about a last-consulted problem and a processing scheme of the user according to the conversation summary text generated based on the conversation summary template, and can accurately learn about the satisfaction of the user for the current online shopping, thereby better providing customer service for the user, and improving user experience.

In another case, a summary utterance may be automatically generated based on the target question text and the target answer text, specifically, the target question text and the target answer text are fused, and a dialog summary text corresponding to the dialog text is generated according to a fusion result.

In specific implementation, the target question text and the target answer text can be fused based on the pre-training model, and a dialogue summary text output by the model is obtained. Taking the target question text as 'ticket use, whether the ticket can be used on holidays' and the target answer text 'the ticket can be used at any time period' as an example, the target question text and the target answer text are input into a pre-training model, the pre-training model performs coding and decoding processing on the target question text and the target answer text to generate a fused conversation summary text, the conversation summary text is 'a problem that a user asks for the use timeliness of the ticket, and the ticket can be used at any time period after replying the user', and by means of the method, a more convenient service summary function can be provided for the customer service, so that the customer service can quickly and accurately know whether the historical problem and the problem of the user are solved, and better consultation service is provided for the user.

In practical application, the text generation method provided by the present specification may also be implemented by a pre-training model, and the specific method further includes: inputting the dialogue text of the associated project participation behavior into a text generation model; determining a target question text corresponding to the dialog text through the text generation model, determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior, determining a target answer text associated with the target question text based on the text to be processed, generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and outputting the text generation model.

The method comprises the steps of inputting a communication conversation text of a user and customer service into a trained text generation model, automatically determining a target question text corresponding to the conversation text by the text generation model, determining a text to be processed based on the conversation text and an auxiliary text, identifying the text to be processed to determine a target answer text, and finally directly outputting the generated conversation summary text. In practical application, order features can be input into the text generation model for prediction and evaluation, and accuracy of prediction results is improved.

The text generation method provided by the specification comprises the steps of obtaining a conversation text of a related project participation behavior, and determining a target problem text corresponding to the conversation text; determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior; determining a target answer text associated with the target question text based on the text to be processed; and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text. The method comprises the steps of determining a corresponding target question text according to a conversation text, introducing an auxiliary text corresponding to the conversation text to enrich information sources, determining a text to be processed according to the auxiliary text and the conversation text, obtaining a target answer text with higher accuracy, and enabling the finally generated conversation summary text to be higher in fine granularity, so that a project worker can know the intention of a project participating in a user more quickly based on the conversation summary text, and therefore more convenient service is provided for the user.

Fig. 3 shows a flowchart of a text generation method provided in accordance with another embodiment of the present specification, including steps 302 to 308.

Step 302: receiving a text generation instruction submitted for the dialog text through a summary page, wherein the summary page is associated with a project participation action.

The summary page can be understood as a page displayed by the customer service system for the customer service staff, when the customer service staff is connected with the user, the text generation of the service summary of the communication can be performed on the summary page, and after the customer service staff submits a text generation instruction for the dialog text through the summary page, the customer service system automatically starts to generate the service summary of the communication for the customer service staff.

Step 304: and responding to the text generation instruction, determining a target question text corresponding to the conversation text, and determining a text to be processed based on the conversation text and an auxiliary text corresponding to the conversation text, wherein the auxiliary text is associated with the project participation behavior.

In practical application, a target question text corresponding to a dialog text is determined in response to a text generation instruction, the target question text is a question to be consulted for the user in the communication, the target question text comprises a question type and a user acoustic field, and when an answer text exists in an auxiliary text corresponding to the dialog text, the auxiliary text is directly used as a text to be processed to perform subsequent text generation operation; and under the condition that the answer text does not exist in the auxiliary text, taking the dialog text as the text to be processed, and continuing to perform the subsequent text generation operation.

Step 306: and determining a target answer text associated with the target question text based on the text to be processed.

In practical application, when the text to be processed is the auxiliary text, the answer text in the auxiliary text is directly used as the target answer text associated with the target question text. And under the condition that the text to be processed is the dialogue text, performing corpus recognition on the dialogue text, determining an answer text in the dialogue text according to a recognition result, and taking the answer text in the dialogue text as a target answer text.

Step 308: and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and displaying the dialog summary text through the summary page.

In practical application, after the target question text and the target answer text are combined to generate a dialogue summary text corresponding to the dialogue text, the dialogue summary text, namely a service summary, is displayed to customer service staff through a summary page.

The text generation method provided by the specification comprises the steps of receiving a text generation instruction submitted by a summary page aiming at a dialog text, wherein the summary page is associated with a project participation behavior; responding to the text generation instruction, determining a target question text corresponding to the conversation text, and determining a text to be processed based on the conversation text and an auxiliary text corresponding to the conversation text, wherein the auxiliary text is associated with the project participation behavior; determining a target answer text associated with the target question text based on the text to be processed; and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and displaying the dialog summary text through the summary page. The method comprises the steps of determining a corresponding target question text according to a conversation text, introducing an auxiliary text corresponding to the conversation text to enrich information sources, determining a text to be processed according to the auxiliary text and the conversation text, obtaining a target answer text with higher accuracy, and enabling the finally generated conversation summary text to be higher in fine granularity, so that a project worker can know the intention of a project participating in a user more quickly based on the conversation summary text, and therefore more convenient service is provided for the user.

The following further describes the text generation method by taking an application of the text generation method provided in this specification to an online shopping item as an example, with reference to fig. 4. Fig. 4 shows a flowchart of a processing procedure of a text generation method provided in an embodiment of the present specification, and specific steps include steps 402 to 410.

Step 402: and acquiring a dialog text of the participation behavior of the associated project, and selecting an initial question text in a question text database based on the dialog text.

In an implementation mode, the item participation behavior is a user online shopping behavior, a conversation text of the item participation behavior is associated, participation state information corresponding to the conversation text is acquired for a conversation text for a user to communicate the online shopping behavior with customer service, the participation state information is in an after-sales state, screening is carried out in a problem text database according to the after-sales state, candidate problem texts are screened out, the matching degree of each candidate problem text and the conversation text is calculated, the text with the highest matching degree is selected as an initial problem text, and the initial problem text is a quality problem.

Step 404: and determining an original problem text in the dialog text according to the initial problem text, combining the initial problem text and the original problem text, and obtaining a target problem text according to a combination result.

In an implementation mode, an original problem text describing the initial problem text is selected from a conversation text according to the determined initial problem text, the original problem text is 'commodity with problem, a merchant does not return goods', the initial problem text and the original problem text are combined, a target problem text is obtained according to a combination result, the target problem text is 'quality problem, the commodity has problem, and the merchant does not return goods'.

Step 406: and determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior.

In an implementation mode, an auxiliary text corresponding to the dialog text is determined based on the dialog identifier of the dialog text, the auxiliary text is a description text of an ISO standard operation process and an SOP standard operation process, the auxiliary text is identified, and the auxiliary text is used as a text to be processed under the condition that an answer text exists in the auxiliary text.

In another implementation manner, an auxiliary text corresponding to the dialog text is determined based on the dialog identifier of the dialog text, the auxiliary text is recognized, and the dialog text is used as the text to be processed when the answer text does not exist in the auxiliary text.

Step 408: and determining target answer text associated with the target question text based on the text to be processed.

In an implementable manner, in the case that the text to be processed is the auxiliary text, the answer text in the auxiliary text, namely the question solution, is directly used as the target answer text associated with the target question text.

In another realizable mode, when the text to be processed is the dialogue text, the dialogue text is respectively input into the guide model and the action model for combined recognition processing, the answer text in the dialogue text, namely the problem solution, is determined according to the recognition result, and the answer text in the dialogue text is used as the target answer text associated with the target problem text.

Step 410: and generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text.

In an implementation mode, a dialog summary template corresponding to the dialog text is determined, the target question text and the target answer text are written into the dialog summary template, and the dialog summary text, namely the service summary, is generated.

In another realizable mode, fusion processing is carried out according to the target question text and the target answer text, and a conversation summary text corresponding to the conversation text is generated according to a fusion result, wherein the conversation summary text is 'the user wants to return goods due to the commodity quality problem and provides return service for the user'.

According to the text generation method provided by the specification, the corresponding target question text is determined according to the dialog text, the auxiliary text corresponding to the dialog text is introduced to enrich the information source, the text to be processed is determined according to the auxiliary text and the dialog text, the target answer text with higher accuracy can be obtained, the fine granularity of the finally generated dialog summary text is higher, a project worker can know the intention of a project participating user more quickly based on the dialog summary text, and therefore more convenient service is provided for the user.

Fig. 5 shows a flowchart of a text generation method provided in accordance with another embodiment of the present specification, which includes steps 502 to 508.

Step 502: the method comprises the steps of obtaining a communication dialogue text between a user and customer service, and determining a target question text corresponding to the communication dialogue text.

The communication dialogue text can be understood as a dialogue text for communicating with customer service aiming at online shopping behaviors, the target problem text can be understood as a problem which a user mainly wants to consult in the communication dialogue, and the target problem text comprises a problem type and a user voice field. The target problem text can be filled in the service summary, so that the customer service providing service for the user later can know the intention of the user more quickly, the communication efficiency is improved, and the use experience of the user is improved.

In an embodiment of the present specification, a user purchases a piece of clothing, inquires of a customer service about an after-sale question of the clothing, obtains a communication session text of this time after the customer service communicates with the user, and determines a target question text of the communication session text, where the target question text is a question that the user mainly wants to consult in this communication.

Step 504: determining a text to be processed based on the communication session text and an auxiliary text corresponding to the communication session text, wherein the auxiliary text is a communication template text used by the customer service.

The auxiliary text can be understood as a dialog text replied according to a communication template of the customer service in the process of communication between the customer service and the user, so that the solution of the user problem recorded in the auxiliary text is the most accurate, when the answer text exists in the auxiliary text, the auxiliary text can be directly used as a text to be processed, and then the target answer text is determined according to the auxiliary text. And under the condition that the answer text does not exist in the auxiliary text, performing corpus recognition on the dialogue text, and determining the target answer text according to the recognition result.

In an embodiment of the present specification, the customer service replies with the user according to the communication template, acquires an auxiliary text corresponding to the communication session text, uses the auxiliary text as a text to be processed, and then can determine a target answer text from the auxiliary text.

Step 506: and determining a target answer text associated with the target question text based on the text to be processed.

In an embodiment of the present specification, following the above example, in the case where the text to be processed is an auxiliary text, the target answer text associated with the target question text may be directly determined from the auxiliary text.

Step 508: and generating a customer service communication summary text corresponding to the communication dialogue text based on the target question text and the target answer text.

In an embodiment of the present specification, the target question text and the target answer text are combined to generate a customer service communication summary text corresponding to the communication session text, where the customer service communication summary text is a service summary of the customer service providing service for the user, and subsequent customer services can quickly know the user intention based on the service summary, so as to provide more efficient consultation service for the user.

The text generation method provided by the present specification includes: acquiring a communication session text between a user and a customer service, and determining a target problem text corresponding to the communication session text; determining a text to be processed based on the communication session text and an auxiliary text corresponding to the communication session text, wherein the auxiliary text is a communication template text used by the customer service; determining a target answer text associated with the target question text based on the text to be processed; and generating a customer service communication summary text corresponding to the communication dialogue text based on the target question text and the target answer text. The method and the device have the advantages that the corresponding target problem text is determined according to the communication conversation text, the auxiliary text corresponding to the conversation text is introduced to enrich the information source, the text to be processed is determined according to the auxiliary text and the conversation text, the target answer text with higher accuracy can be obtained, the fine granularity of the finally generated customer service communication summary text is higher, project personnel can quickly know the intention of a project participating user based on the conversation summary text, and therefore more convenient service is provided for the user.

Corresponding to the above method embodiment, this specification further provides a text generation apparatus embodiment, and fig. 6 shows a schematic structural diagram of a text generation apparatus provided in an embodiment of this specification. As shown in fig. 6, the apparatus includes:

an obtaining module 602, configured to obtain a dialog text of a participation action of an associated project, and determine a target question text corresponding to the dialog text;

a first determining module 604 configured to determine a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior;

a second determining module 606 configured to determine a target answer text associated with the target question text based on the text to be processed;

a generating module 608 configured to generate a dialog summary text corresponding to the dialog text based on the target question text and the target answer text.

Optionally, the obtaining module 602 is further configured to:

selecting an initial question text in a question text database based on the dialog text;

determining an original problem text in the dialog text according to the initial problem text;

and merging the initial problem text and the original problem text, and obtaining a target problem text according to a merging result.

Optionally, the obtaining module 602 is further configured to:

acquiring participation state information corresponding to the conversation text;

performing text screening on a problem text database based on the participation state information, and determining a problem text to be selected according to a screening result;

and calculating the matching degree of each question text to be selected and the dialog text, and selecting the initial question text according to each matching degree.

Optionally, the first determining module 604 is further configured to:

identifying the auxiliary text, and taking the auxiliary text as a text to be processed under the condition that answer texts exist in the auxiliary text according to the identification result;

correspondingly, the step of determining a target answer text associated with the target question text based on the text to be processed comprises the following steps:

and taking the answer text in the auxiliary text as the target answer text associated with the target question text.

Optionally, the first determining module 604 is further configured to:

identifying the auxiliary text, and taking the dialog text as a text to be processed under the condition that the auxiliary text is determined not to have an answer text according to an identification result;

correspondingly, determining a target answer text associated with the target question text based on the text to be processed comprises:

and identifying the dialog text, determining an answer text in the dialog text according to an identification result, and taking the answer text in the dialog text as a target answer text associated with the target question text.

Optionally, the first determining module 604 is further configured to:

inputting the conversation text into a guide model to perform guide recognition processing to obtain a guide answer text, and inputting the conversation text into an action model to perform action recognition processing to obtain an action answer text;

and determining an answer text in a dialog text according to the guide answer text and the action answer text, and taking the answer text in the dialog text as a target answer text associated with the target question text.

Optionally, the generating module 608 is further configured to:

determining participation progress information of the project participation behavior, determining a conversation summary template corresponding to the conversation text based on the participation progress information, writing the target question text and the target answer text into the conversation summary template, and generating a conversation summary text corresponding to the conversation text; alternatively, the first and second electrodes may be,

and fusing the target question text and the target answer text, and generating a dialog summary text corresponding to the dialog text according to a fusion result.

Optionally, the apparatus further comprises an input module configured to:

inputting the dialogue text of the associated project participation behavior into a text generation model;

determining a target question text corresponding to the dialog text through the text generation model, determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior, determining a target answer text associated with the target question text based on the text to be processed, generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and outputting the text generation model.

Optionally, the first determining module 604 is further configured to:

determining dialog identification information of the dialog text;

and determining an auxiliary text corresponding to the dialog text according to the dialog identification information.

The text generation device provided by the specification comprises an acquisition module, a processing module and a display module, wherein the acquisition module is configured to acquire a conversation text of a related project participation behavior and determine a target problem text corresponding to the conversation text; a first determination module configured to determine a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior; a second determination module configured to determine a target answer text associated with the target question text based on the text to be processed; a generating module configured to generate a dialog summary text corresponding to the dialog text based on the target question text and the target answer text. By determining the corresponding target question text according to the dialog text, introducing the auxiliary text corresponding to the dialog text to enrich the information source, and determining the text to be processed according to the auxiliary text and the dialog text, the target answer text with higher accuracy can be obtained, the fine granularity of the finally generated dialog summary text is higher, and a project worker can know the intention of the project participating in the user more quickly based on the dialog summary text, so that more convenient service is provided for the user.

The above is a schematic scheme of a text generating apparatus of the present embodiment. It should be noted that the technical solution of the text generation apparatus and the technical solution of the text generation method belong to the same concept, and details that are not described in detail in the technical solution of the text generation apparatus can be referred to the description of the technical solution of the text generation method.

Corresponding to the above method embodiment, this specification further provides a text generation apparatus embodiment, and fig. 7 shows a schematic structural diagram of a text generation apparatus provided in another embodiment of this specification. As shown in fig. 7, the apparatus includes:

a receiving module 702 configured to receive a text generation instruction submitted for a dialog text through a summary page, wherein the summary page is associated with a project participation action;

a first determining module 704, configured to determine, in response to the text generation instruction, a target question text corresponding to the dialog text, and determine a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, where the auxiliary text is associated with the project participation behavior;

a second determination module 706 configured to determine a target answer text associated with the target question text based on the text to be processed;

a generating module 708 configured to generate a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and display the dialog summary text through the summary page.

Optionally, the first determining module 702 is further configured to:

Optionally, the generating module 708 is further configured to:

Optionally, the apparatus further comprises an input module configured to:

Optionally, the first determining module 702 is further configured to:

determining dialog identification information of the dialog text;

The text generation device comprises a receiving module, a processing module and a display module, wherein the receiving module is configured to receive a text generation instruction submitted for a dialog text through a summary page, and the summary page is associated with a project participation behavior; a first determining module, configured to determine, in response to the text generation instruction, a target question text corresponding to the dialog text, and determine a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, where the auxiliary text is associated with the project participation behavior; a second determination module configured to determine a target answer text associated with the target question text based on the text to be processed; and the generating module is configured to generate a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, and display the dialog summary text through the summary page. By determining the corresponding target question text according to the dialog text, introducing the auxiliary text corresponding to the dialog text to enrich the information source, and determining the text to be processed according to the auxiliary text and the dialog text, the target answer text with higher accuracy can be obtained, the fine granularity of the finally generated dialog summary text is higher, and a project worker can know the intention of the project participating in the user more quickly based on the dialog summary text, so that more convenient service is provided for the user.

The foregoing is a schematic scheme of a text generating apparatus of the present embodiment. It should be noted that the technical solution of the text generation apparatus and the technical solution of the text generation method belong to the same concept, and details that are not described in detail in the technical solution of the text generation apparatus can be referred to the description of the technical solution of the text generation method.

Corresponding to the above method embodiment, this specification further provides a text generation apparatus embodiment, and fig. 8 shows a schematic structural diagram of a text generation apparatus provided in another embodiment of this specification. As shown in fig. 8, the apparatus includes:

an obtaining module 802, configured to obtain a communication session text between a user and a customer service, and determine a target question text corresponding to the communication session text;

a first determining module 804, configured to determine a text to be processed based on the communication session text and an auxiliary text corresponding to the communication session text, where the auxiliary text is a communication template text used by the customer service;

a second determining module 806 configured to determine a target answer text associated with the target question text based on the text to be processed;

a generating module 808, configured to generate a customer service communication summary text corresponding to the communication session text based on the target question text and the target answer text.

Optionally, the obtaining module 802 is further configured to:

determining an original question text in the dialog text according to the initial question text;

Optionally, the obtaining module 802 is further configured to:

Optionally, the first determining module 804 is further configured to:

Optionally, the generating module 808 is further configured to:

Optionally, the apparatus further comprises an input module configured to:

inputting the dialogue text of the participation behavior of the associated project into a text generation model;

Optionally, the first determining module 804 is further configured to:

determining dialog identification information of the dialog text;

The text generation device provided by the specification comprises an acquisition module, a processing module and a display module, wherein the acquisition module is configured to acquire a communication session text between a user and a customer service and determine a target question text corresponding to the communication session text; a first determining module, configured to determine a text to be processed based on the communication session text and an auxiliary text corresponding to the communication session text, where the auxiliary text is a communication template text used by the customer service; a second determination module configured to determine a target answer text associated with the target question text based on the text to be processed; and the generating module is configured to generate a customer service communication summary text corresponding to the communication conversation text based on the target question text and the target answer text. The corresponding target question text is determined according to the communication conversation text, the auxiliary text corresponding to the conversation text is introduced to enrich the information source, and the text to be processed is determined according to the auxiliary text and the conversation text, so that the target answer text with higher accuracy can be obtained, the fine granularity of the finally generated customer service communication summary text is higher, a project worker can quickly know the intention of the project participating in the user based on the customer service communication summary text, and more convenient service is provided for the user.

Fig. 9 illustrates a block diagram of a computing device 900 provided in accordance with an embodiment of the present description. Components of the computing device 900 include, but are not limited to, a memory 910 and a processor 920. The processor 920 is coupled to the memory 910 via a bus 930, and a database 950 is used to store data.

Computing device 900 also includes access device 940, access device 940 enabling computing device 900 to communicate via one or more networks 960. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. Access device 940 may include one or more of any type of network interface (e.g., a Network Interface Card (NIC)) whether wired or wireless, such as an IEEE902.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.

In one embodiment of the present description, the above-described components of computing device 900, as well as other components not shown in FIG. 9, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device architecture shown in FIG. 9 is for purposes of example only and is not limiting as to the scope of the description. Those skilled in the art may add or replace other components as desired.

Computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet, personal digital assistant, laptop, notebook, netbook, etc.), a mobile phone (e.g., smartphone), a wearable computing device (e.g., smartwatch, smartglasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or PC. Computing device 900 may also be a mobile or stationary server.

Wherein the steps of the text generation method are implemented by processor 920 when executing the computer instructions.

The above is an illustrative scheme of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the technical solution of the text generation method belong to the same concept, and details that are not described in detail in the technical solution of the computing device can be referred to the description of the technical solution of the text generation method.

An embodiment of the present specification further provides a text generation system, in which computer instructions are stored, where the system includes a server and a client; the client is used for storing the text display executable instruction, and the server is used for storing the text generation executable instruction; the text presentation executable instructions when executed by the client and the text generation executable instructions when executed by the server implement the steps of the text generation method as previously described.

An embodiment of the present specification further provides a computer readable storage medium storing computer instructions, which when executed by a processor implement the steps of the text generation method as described above.

The above is an illustrative scheme of a computer-readable storage medium of the embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the text generation method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the text generation method.

An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the text generation method.

The above is an illustrative scheme of a computer program of the present embodiment. It should be noted that the technical solution of the computer program and the technical solution of the text generation method belong to the same concept, and details that are not described in detail in the technical solution of the computer program can be referred to the description of the technical solution of the text generation method.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying said computer program code, a recording medium, a usb-disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a Random Access Memory (RAM), an electrical carrier signal, a telecommunications signal, a software distribution medium, etc. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

It should be noted that, for the sake of simplicity, the foregoing method embodiments are described as a series of acts, but those skilled in the art should understand that the present embodiment is not limited by the described acts, because some steps may be performed in other sequences or simultaneously according to the present embodiment. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that acts and modules referred to are not necessarily required for an embodiment of the specification.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are intended only to aid in the description of the specification. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the embodiments. The specification is limited only by the claims and their full scope and equivalents.

Claims

1. A text generation method, comprising:

determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, wherein the auxiliary text is associated with the project participation behavior;

2. The method of claim 1, determining a target question text corresponding to the dialog text, comprising:

3. The method of claim 2, selecting an initial question text in a question text database based on the dialog text, comprising:

4. The method of any one of claims 1-3, determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, comprising:

5. The method of any one of claims 1-3, determining a text to be processed based on the dialog text and an auxiliary text corresponding to the dialog text, comprising:

6. The method of claim 5, wherein the recognizing the dialog text, determining an answer text in the dialog text according to the recognition result, and using the answer text in the dialog text as a target answer text associated with the target question text comprises:

7. The method of claim 1, generating a dialog summary text corresponding to the dialog text based on the target question text and the target answer text, comprising:

8. The method of claim 1, further comprising:

9. The method of claim 1, the determining of the auxiliary text comprising:

determining dialog identification information of the dialog text;

10. A text generation method, comprising:

11. A text generation method, comprising:

12. A text generation system comprises a server and a client;

the client is used for storing the text display executable instruction, and the server is used for storing the text generation executable instruction; the text presentation executable instructions when executed by the client and the text generation executable instructions when executed by the server implement the steps of the method of any one of claims 1 to 9 or 10 or 11.

13. A computing device comprising a memory, a processor and computer instructions stored on the memory and executable on the processor, the processor implementing the steps of the method of any one of claims 1 to 9 or 10 or 11 when executing the computer instructions.

14. A computer readable storage medium storing computer executable instructions which, when executed by a processor, carry out the steps of the method of any one of claims 1 to 9 or 10 or 11.