Guidance video generation method and device, electronic equipment and storage medium

Abstract

The invention provides a guide video generation method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence. The guide video generation method comprises the following steps: receiving target inquiry information in an interaction scene; obtaining target object information of an object to which the target inquiry information belongs; performing feature extraction on the target inquiry information to obtain target inquiry semantics; based on the target inquiry semantics, determining an operation guidance text corresponding to the target inquiry information; obtaining initial video data corresponding to the operation guidance text, and obtaining object sub-information corresponding to the operation guidance text from the target object information; and enabling the video generation model to generate target video data according to the operation guidance text, the object sub-information and initial video data corresponding to the operation guidance text, and synthesizing to obtain a guidance video conforming to the target object information according to the target video data and the time sequence information. The technical problem that the interaction effect is affected due to the fact that operation guidance is not visual enough in the prior art is solved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a guidance video generation method, device, electronic device and storage medium. Background Art

[0002] In order to provide users with a more considerate customer service experience, many financial services have launched intelligent customer service services. Intelligent customer service services can not only replace manual customer service, but also have the advantages of low consumption and uninterrupted service. In the related technology, the existing intelligent customer service services are not intuitive enough in the face of complex interaction scenarios, which affects the interaction effect. Summary of the invention

[0003] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, the first purpose of the present disclosure is to propose a guidance video generation method that can improve the intuitiveness of operation guidance and thus improve the interaction effect.

[0005] A second objective of the present disclosure is to provide a guidance video generating device.

[0006] A third objective of the present disclosure is to provide an electronic device.

[0007] A fourth objective of the present disclosure is to provide a computer-readable storage medium.

[0008] A fifth object of the present disclosure is to provide a computer program product.

[0009] To achieve the above-mentioned purpose, the first aspect of the present disclosure proposes a method for generating a guidance video, comprising: receiving a request message in an interactive scene, wherein the request message includes target query information; obtaining target object information of an object to which the target query information belongs; performing feature extraction on the target query information to obtain target query semantics; based on the target query semantics, determining target feedback information corresponding to the target query information, wherein the target feedback information includes at least one operation guidance text; obtaining initial video data corresponding to the operation guidance text, wherein the initial video data has corresponding timing information; obtaining object sub-information corresponding to the operation guidance text from the target object information; and inputting at least one operation guidance text, at least one object sub-information, and at least one frame of initial video data into a video generation model, so that the video generation model generates target video data according to the operation guidance text, the object sub-information, and the initial video data corresponding to the operation guidance text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data.

[0010] To achieve the above-mentioned purpose, the second aspect of the present disclosure proposes a guidance video generation device, comprising: a receiving module, used to receive a request message in an interactive scene, wherein the request message includes target query information; a first acquisition module, used to acquire target object information of an object to which the target query information belongs; an extraction module, used to perform feature extraction on the target query information to obtain target query semantics; a determination module, used to determine target feedback information corresponding to the target query information based on the target query semantics, wherein the target feedback information includes at least one operation guidance text; and a second acquisition module, used to acquire initial video data corresponding to the operation guidance text, wherein the initial video data has corresponding timing information; a third acquisition module, used to acquire object sub-information corresponding to the operation guidance text from the target object information; a generation module, used to input at least one operation guidance text, at least one object sub-information, and at least one frame of initial video data into a video generation model, so that the video generation model generates target video data according to the operation guidance text, the object sub-information, and the initial video data corresponding to the operation guidance text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data.

[0011] To achieve the above-mentioned purpose, the third aspect embodiment of the present disclosure proposes an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement a guidance video generation method proposed in the first aspect embodiment of the present disclosure.

[0012] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present disclosure proposes a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed by the processor, they are used to implement the guidance video generation method proposed in the first aspect embodiment of the present disclosure.

[0013] To achieve the above-mentioned purpose, the fifth aspect of the present disclosure proposes a computer program product, including a computer program, which, when executed by a processor, implements the guidance video generation method proposed in the first aspect of the present disclosure.

[0014] The present disclosure provides a guidance video generation method, device, electronic device, non-transitory computer-readable storage medium storing computer instructions, and computer program product. The method receives a request message in an interactive scene, wherein the request message includes target query information, obtains target object information of an object to which the target query information belongs, performs feature extraction on the target query information, obtains target query semantics, determines target feedback information corresponding to the target query information based on the target query semantics, wherein the target feedback information includes at least one operation guidance text, obtains initial video data corresponding to the operation guidance text, wherein the initial video data has corresponding timing information, obtains object sub-information corresponding to the operation guidance text from the target object information, and inputs at least one operation guidance text, at least one object sub-information, and at least one frame of initial video data into a video generation model, so that the video generation model generates target video data according to the operation guidance text, the object sub-information, and the initial video data corresponding to the operation guidance text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data. This supports the generation of a guidance video that conforms to the target object information, and because the guidance video can be adapted to the target object information of the object to which the target query information belongs, when operation guidance is performed based on the guidance video that conforms to the target object information, the intuitiveness of the operation guidance can be improved, thereby improving the interaction effect.

[0015] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0017] Figure 1 A flowchart of a method for generating a guidance video provided by an embodiment of the present disclosure;

[0018] Figure 2 A flowchart of another method for generating a guidance video provided by an embodiment of the present disclosure;

[0019] Figure 3 is a schematic diagram of the model training process in an embodiment of the present disclosure;

[0020] Figure 4 is a schematic diagram of a video generation process in an embodiment of the present disclosure;

[0021] Figure 5 A schematic diagram of the structure of a guidance video generating device provided by an embodiment of the present disclosure;

[0022] Figure 6 It is a schematic structural diagram of an electronic device shown in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0024] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this disclosure are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions. The acquisition, transmission, storage, use and processing of data in the technical solution of this disclosure are in compliance with the relevant provisions of national laws and regulations.

[0025] It should be noted that in the embodiments of the present disclosure, certain software, components, models and other existing solutions in the industry may be mentioned, which should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present disclosure, but it does not mean that the applicant has or will necessarily use the solution.

[0026] The guidance video generation method provided in the embodiments of the present disclosure can be applied in intelligent customer service scenarios, using intelligent customer service to provide customer service to users instead of manual customer service.

[0027] Figure 1 A flowchart of a method for generating a guidance video provided in an embodiment of the present disclosure.

[0028] This embodiment takes the guidance video generation method being configured in a guidance video generation device as an example. In this embodiment, the guidance video generation method can be configured in the guidance video generation device, and the guidance video generation device can be set in a server or in an electronic device, without limitation.

[0029] This embodiment takes the guidance video generation method configured in an electronic device as an example, wherein the electronic device is a hardware device such as a smart phone, a tablet computer, a personal digital assistant, an e-book, etc. having various operating systems.

[0030] It should be noted that the execution subject of the embodiment of the present disclosure may be, for example, a central processing unit (CPU) in a server or electronic device in terms of hardware, or may be, for example, a related background service in a server or electronic device in terms of software, without limitation.

[0031] like Figure 1 As shown, the guidance video generation method includes:

[0032] S101: receiving a request message in an interaction scenario, wherein the request message includes target query information.

[0033] In some embodiments, the interaction scenario may be, for example, a scenario in which a question-and-answer service is provided to a user based on an intelligent customer service service, and this is not limited.

[0034] The request message is used to trigger a request for providing a question-and-answer service, and the request message may include target query information, which is used to trigger obtaining a query result provided by the question-and-answer service. For example, the intelligent customer service service may perform semantic recognition on the request message input by the user, obtain the target query information from the result of the semantic recognition, and respond to the target query information to provide a query result that meets the target query information.

[0035] In some embodiments, a request message can be received by an electronic device. For example, the electronic device can provide an intelligent customer service interface and provide an interface for request messages in the intelligent customer service interface. The electronic device can receive a request message input by a user based on the interface and parse the request message to obtain target inquiry information. There is no limitation on this.

[0036] S102: Acquire target object information of the object to which the target query information belongs.

[0037] In some embodiments, after receiving a request message in an interaction scenario and parsing the request message to obtain target query information, the target object information of the object to which the target query information belongs can be determined. The target object information is used to describe the object to which the target query information belongs. For example, the target object information can be feature information, attribute information, etc. of the object. The object can be, for example, a customer, a user, etc. In other words, the target query information can be a query initiated by the object to obtain feedback information provided by the intelligent customer service service.

[0038] In the embodiments of the present disclosure, the target object information of the object to which the target query information belongs can be determined, such as by reversely deducing the characteristic information that the object to which it belongs may have based on the target query information, and using the characteristic information as the target object information; or, the object that triggers the sending of the request message can be determined, and the characteristic information, attribute information, etc. of the object can be determined, and the characteristic information, attribute information, etc. of the object can be used as the target object information; or, any other possible method can be used to obtain the target object information of the object to which the target query information belongs, such as a query-based method, a prediction method, etc., without limitation.

[0039] In some embodiments, the determined target object information can be used as reference information when generating a guidance video, so that the generated guidance video can be adapted to the target object information, thereby improving the reference value of the guidance video.

[0040] S103: Extract features of the target query information to obtain the target query semantics.

[0041] After obtaining the target query information based on the request message parsing, the target query information can be feature extracted to obtain the target query semantics. The target query semantics is used to characterize the semantic content carried by the target query information, and the target query semantics can accurately reflect the query intention and support obtaining accurate feedback information.

[0042] In some embodiments, feature extraction may be performed on the target query information to obtain query features, and the query features may be vectorized to obtain a query feature vector, and then the query feature vector may be processed based on a semantic analysis model to identify the target query semantics. Alternatively, field decomposition may be performed on the target query information to obtain multiple query fields, sub-semantics of each query field may be determined, and contextual semantic features of different query fields may be combined to obtain the target query semantics. Alternatively, any other possible method may be used to extract features from the target query information to obtain the target query semantics, and there is no limitation on this.

[0043] S104: Determine target feedback information corresponding to the target query information based on the target query semantics, wherein the target feedback information includes at least one operation guidance text.

[0044] After identifying the target query semantics of the target query information, target feedback information corresponding to the target query information can be obtained based on the target query semantics, wherein the target feedback information is information used to provide question-answer feedback on the target query information. The target feedback information can represent the feedback result for the target query information.

[0045] In the embodiments of the present disclosure, the feedback information may include one or more operation guide texts. The operation guide text refers to the text used to guide the customer or user to perform the operation. The operation guide text can be used as the target feedback information of the feedback target inquiry information. Alternatively, the operation guide text can be a part of the target feedback information, and the target feedback information can also include content for explaining the operation guide text, or can also include any other content that may be used to feedback the target inquiry information, without limitation.

[0046] In an embodiment of the present disclosure, one or more adapted operation guidance texts can be identified based on the target query semantics, and the one or more operation guidance texts can be carried in the target feedback information, or the target query semantics can be intelligently parsed, and one or more operation guidance texts can be generated based on the parsed results, and the one or more operation guidance texts can be carried in the target feedback information, or the target query semantics can be processed based on artificial intelligence to obtain one or more adapted operation guidance texts, and the one or more operation guidance texts can be carried in the target feedback information, without limitation.

[0047] S105: Acquire initial video data corresponding to the operation guide text, wherein the initial video data has corresponding timing information.

[0048] The above-mentioned initial video data may be a frame of video data corresponding to the operation guide text. The initial video data may be understood as a video data template. "Corresponding" means that the video semantics of the initial video data is consistent with the semantics of the operation guide text. For example, if the operation guide text is "Click on the credit card, and click on installment repayment after the page jumps", then the initial video data corresponding to the operation guide text may be a video data that includes a credit card and presents a "click" control to support clicking on the credit card to jump, and there is no restriction on this.

[0049] The number of frames of the initial video data corresponding to the operation guide text may be one frame or multiple frames. For example, each operation guide text may correspond to one frame of initial video data, or each operation guide text may correspond to multiple frames of initial video data. Each initial video data may be pre-configured and loaded according to usage requirements, and may support dynamic supplementation and modification of the initial video data, without limitation.

[0050] The above-mentioned initial video data has corresponding timing information, and the timing information may have a one-to-one correspondence with the initial video data. For example, one frame of initial video data corresponds to one timing information, and the timing information is used to describe the presentation time and / or presentation order of the initial video data in the complete video. When the initial video data is processed into target video data, the timing information of the initial video data can be used as the timing information of the corresponding target video data, and there is no limitation on this.

[0051] S106: Acquire object sub-information corresponding to the operation guide text from the target object information.

[0052] After obtaining the target object information, object sub-information corresponding to the operation guide text may also be obtained from the target object information. Part of the information included in the target object information that matches the text semantics of the operation guide text may be referred to as object sub-information corresponding to the operation guide text.

[0053] For example, if the operation guide text is "click on the credit card, and click on installment repayment after the page jumps", the object sub-information corresponding to the operation guide text can be the credit card number of the object, and there is no limitation on this.

[0054] Optionally, when identifying object sub-information corresponding to the operation guide text, the operation guide text can be semantically parsed to obtain a semantic label, and partial information matching the semantic label can be identified from the target object information, and the partial information can be used as the object sub-information corresponding to the operation guide text. There is no limitation on this.

[0055] S107: Input at least one operation instruction text, at least one object sub-information, and at least one frame of initial video data into a video generation model, so that the video generation model generates target video data according to the operation instruction text, the object sub-information, and the initial video data corresponding to the operation instruction text, and synthesizes a instruction video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data.

[0056] In some embodiments, after the target feedback information corresponding to the target query information is determined based on the target query semantics, wherein the target feedback information includes at least one operation guide text, the initial video data corresponding to the operation guide text can be obtained, wherein the initial video data has corresponding timing information, and the object sub-information corresponding to the operation guide text is obtained from the target object information, and then a guidance video that conforms to the target object information is generated based on the operation guide text, the object sub-information, the initial video data, and the timing information. That is to say, the embodiments of the present disclosure support the generation of a guidance video that conforms to the target object information, because the guidance video can be adapted to the target object information of the object to which the target query information belongs, thus, when the operation guidance is performed based on the guidance video that conforms to the target object information, the intuitiveness of the operation guidance can be improved, thereby improving the interaction effect.

[0057] In some embodiments, the initial video data corresponding to each operation instruction text can be obtained. The initial video data can be pre-configured, and each initial video data can effectively conform to the service type provided by the intelligent customer service service. The initial video data can include a part in which object sub-information can be embedded, and then the object sub-information can be embedded into the part of the initial video data in which the object sub-information can be embedded to obtain the target video data. According to this method, multiple target video data can be obtained in sequence, and then based on the timing information of each target video data, the multiple target video data can be integrated into a guidance video, and there is no restriction on this.

[0058] In this embodiment, the operation guidance text, object sub-information, initial video data, and time sequence information can be processed in combination with artificial intelligence to generate a guidance video that conforms to the target object information, and there is no limitation to this.

[0059] The video generation model may be an artificial intelligence model, such as a neural network model or a machine learning model. The video generation model may be used to process input operation instruction text, object sub-information, initial video data, and time sequence information, and output an instruction video that conforms to the target object information.

[0060] In some embodiments, at least one operation guidance text, at least one object sub-information, and at least one frame of initial video data can be input into a video generation model, so that the video generation model generates target video data based on the operation guidance text, the object sub-information, and the initial video data corresponding to the operation guidance text, and synthesizes a guidance video that conforms to the target object information based on at least one frame of target video data and the timing information corresponding to each frame of target video data.

[0061] That is, the initial video data, the operation guide text and the object sub-information can be synthesized based on the video generation model to obtain the target video data, and then multiple frames of the target video data can be synthesized into the guide video based on the various timing information. Optionally, in some embodiments, the video generation model can be pre-trained, and the training process of the video generation model can be specifically referred to in the following embodiments.

[0062] In this embodiment, by receiving a request message in an interactive scene, wherein the request message includes target query information, and obtaining target object information of an object to which the target query information belongs, feature extraction is performed on the target query information to obtain target query semantics, and based on the target query semantics, target feedback information corresponding to the target query information is determined, wherein the target feedback information includes at least one operation guide text, and initial video data corresponding to the operation guide text is obtained, wherein the initial video data has corresponding timing information, object sub-information corresponding to the operation guide text is obtained from the target object information, and at least one operation guide text, at least one object sub-information, and at least one frame of initial video data are input into a video generation model, so that the video generation model generates target video data according to the operation guide text, the object sub-information, and the initial video data corresponding to the operation guide text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data. This supports the generation of a guidance video that conforms to the target object information, and because the guidance video can be adapted to the target object information of the object to which the target query information belongs, when operation guidance is performed based on the guidance video that conforms to the target object information, the intuitiveness of the operation guidance can be improved, thereby improving the interaction effect.

[0063] Figure 2 A flowchart of another guidance video generation method provided by an embodiment of the present disclosure.

[0064] like Figure 2 As shown, the guidance video generation method includes:

[0065] S201: Receive a request message in an interaction scenario, wherein the request message includes target query information.

[0066] In some embodiments, the interaction scenario may be, for example, a scenario in which a question-and-answer service is provided to a user based on an intelligent customer service service, and this is not limited.

[0067] The request message is used to trigger a request for providing a question-and-answer service, and the request message may include target query information, which is used to trigger obtaining a query result provided by the question-and-answer service. For example, the intelligent customer service service may perform semantic recognition on the request message input by the user, obtain the target query information from the result of the semantic recognition, and respond to the target query information to provide a query result that meets the target query information.

[0068] In some embodiments, a request message can be received by an electronic device. For example, the electronic device can provide an intelligent customer service interface and provide an interface for request messages in the intelligent customer service interface. The electronic device can receive a request message input by a user based on the interface and parse the request message to obtain target inquiry information. There is no limitation on this.

[0069] S202: Acquire target object information of the object to which the target query information belongs.

[0070] In some embodiments, after receiving a request message in an interaction scenario and parsing the request message to obtain target query information, the target object information of the object to which the target query information belongs can be determined. The target object information is used to describe the object to which the target query information belongs. For example, the target object information can be feature information, attribute information, etc. of the object. The object can be, for example, a customer, a user, etc. In other words, the target query information can be a query initiated by the object to obtain feedback information provided by the intelligent customer service service.

[0071] In the embodiments of the present disclosure, the target object information of the object to which the target query information belongs can be determined, such as by reversely deducing the characteristic information that the object to which it belongs may have based on the target query information, and using the characteristic information as the target object information; or, the object that triggers the sending of the request message can be determined, and the characteristic information, attribute information, etc. of the object can be determined, and the characteristic information, attribute information, etc. of the object can be used as the target object information; or, any other possible method can be used to obtain the target object information of the object to which the target query information belongs, such as a query-based method, a prediction method, etc., without limitation.

[0072] In some embodiments, the determined target object information can be used as reference information when generating a guidance video, so that the generated guidance video can be adapted to the target object information, thereby improving the reference value of the guidance video.

[0073] S203: Extract features of the target query information to obtain target query semantics.

[0074] After obtaining the target query information based on the request message parsing, the target query information can be feature extracted to obtain the target query semantics. The target query semantics is used to characterize the semantic content carried by the target query information, and the target query semantics can accurately reflect the query intention and support obtaining accurate feedback information.

[0075] In some embodiments, feature extraction may be performed on the target query information to obtain query features, and the query features may be vectorized to obtain a query feature vector, and then the query feature vector may be processed based on a semantic analysis model to identify the target query semantics. Alternatively, field decomposition may be performed on the target query information to obtain multiple query fields, sub-semantics of each query field may be determined, and contextual semantic features of different query fields may be combined to obtain the target query semantics. Alternatively, any other possible method may be used to extract features from the target query information to obtain the target query semantics, and there is no limitation on this.

[0076] S204: Inputting the query semantic features into the operation guidance text generation model to obtain target feedback information output by the operation guidance text generation model, wherein the target feedback information includes at least one operation guidance text, and the operation guidance text generation model has modeled and learned the mapping relationship between the query semantic features and at least one operation guidance text in the target feedback information.

[0077] In this embodiment, the semantic features of the query can be processed in combination with artificial intelligence to generate one or more operation guidance texts that meet the semantic features of the query, and the one or more operation guidance texts can be carried in the target feedback information, without limitation.

[0078] The operation guide text generation model may be an artificial intelligence model, such as a neural network model or a machine learning model. The operation guide text generation model may be used to process the input query semantic features and parse the query semantic features to determine the content that can be used as feedback of the target query information, and then match one or more operation guide texts from the feedback content.

[0079] The operation guide text generation model has modeled and learned the mapping relationship between the query semantic features and at least one operation guide text in the target feedback information. The modeling and learning process of the operation guide text generation model can be seen in the following description.

[0080] In some embodiments, the operation guidance text generation model is trained based on the following method: historical customer service data is obtained, wherein the historical customer service data includes: sample inquiry information, and annotated feedback information corresponding to the sample inquiry information, wherein the annotated feedback information includes at least one annotated guidance text; feature extraction is performed on the sample inquiry information to obtain sample inquiry semantics; the initial text generation model is iteratively trained at least once according to the sample inquiry semantics until at least one predicted guidance text in the predicted feedback information obtained based on the trained text generation model and at least one annotated guidance text meet the loss condition, and the trained text generation model is used as the operation guidance text generation model. In this way, the modeling accuracy of the operation guidance text generation model can be effectively ensured, and when the inquiry semantic features are processed based on the operation guidance text generation model to obtain at least one operation guidance text, the acquisition efficiency and accuracy of the target feedback information can be effectively improved, and the interaction effect can be improved.

[0081] The sample query information and the annotation feedback information corresponding to the sample query information are both used to train the operation guide text generation model. The annotation feedback information is used as feedback information of the sample query information. The annotation feedback information may be pre-annotated. The annotation feedback information may include one or more annotation guidance texts, which may also be pre-annotated. The one or more annotation guidance texts in the annotation feedback information are used as a reference for judging whether the operation guide text generation model converges. For example, one or more annotation guidance texts and the loss value between one or more predicted guidance texts output by the operation guide text generation model based on the sample query information during the training process may be used. When the loss value is less than or equal to the loss threshold, it is determined that at least one predicted guidance text in the prediction feedback information obtained by the trained text generation model and at least one annotation guidance text meet the loss condition. When the loss value is greater than the loss threshold, it is determined that at least one predicted guidance text in the prediction feedback information obtained by the trained text generation model and at least one annotation guidance text do not meet the loss condition. Then, when the loss condition is not met, the next round of training of the operation guide text generation model is triggered until the operation guide text generation model converges.

[0082] S205: Acquire initial video data corresponding to the operation guide text, wherein the initial video data has corresponding timing information.

[0083] S206: Acquire object sub-information corresponding to the operation guide text from the target object information.

[0084] For the description of S205 - S206 , please refer to the above embodiment for details, which will not be repeated here.

[0085] S207: Input at least one operation instruction text, at least one object sub-information, and at least one frame of initial video data into the video generation model, so that the video generation model generates target video data according to the operation instruction text, the object sub-information, and the initial video data corresponding to the operation instruction text, and synthesizes a instruction video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data.

[0086] In some embodiments, after inputting the query semantic features into the operation guidance text generation model, obtaining the target feedback information output by the operation guidance text generation model, and clarifying one or more operation guidance texts contained in the target feedback information, it is possible to further obtain initial video data corresponding to the operation guidance text, wherein the initial video data has corresponding timing information, and obtain object sub-information corresponding to the operation guidance text from the target object information, and generate a guidance video that conforms to the target object information based on the operation guidance text, the object sub-information, the initial video data, and the timing information.

[0087] In this embodiment, the operation guidance text, object sub-information, initial video data, and timing information can be processed in combination with artificial intelligence to generate a guidance video that conforms to the target object information, and there is no limitation to this.

[0088] The video generation model may be an artificial intelligence model, such as a neural network model or a machine learning model. The video generation model may be used to process input operation guide text, object sub-information, initial video data, and timing information, and output a guide video that conforms to the target object information.

[0089] In some embodiments, all operation instruction texts, object sub-information, initial video data, and timing information may be input into a video generation model, and a instruction video output by the video generation model that conforms to the target object information may be obtained.

[0090] In some embodiments, the video generation model includes: a text feature extraction submodule, an information feature extraction submodule, and a video generation submodule; wherein, the text feature vector corresponding to each operation guide text can be determined based on the text feature extraction submodule, and the information feature vector of the object sub-information can be determined based on the information feature extraction submodule, and the text feature vector, the information feature vector, and the initial video data corresponding to the operation guide text can be processed based on the video generation submodule to obtain the target video data, and the guidance video that conforms to the target object information can be synthesized based on the timing information corresponding to each frame of the target video data and multiple frames of target video data. In this way, the video generation quality of the video generation model can be effectively improved, and the degree of adaptation of the guidance video and the target object information can be further supported.

[0091] The video generation model has modeled and learned the mapping relationship between the operation guide text, object sub-information, initial video data, timing information, and the guide video that meets the target object information. The modeling and learning process of the video generation model can be found in the following description.

[0092] In some embodiments, the video generation model is trained based on the following method: obtaining at least one sample guidance text and sample object information, wherein at least one sample guidance text and sample object information have a corresponding annotated guidance video, and obtaining sample video data corresponding to the sample guidance text, wherein the sample video data has corresponding timing information, obtaining sample sub-information corresponding to the sample guidance text from the sample object information, and performing at least one iterative training on the initial generation model based on at least one sample guidance text, at least one frame of sample video data, and at least one sample sub-information, until the loss value between the predicted guidance video and the annotated guidance video output by the training generation model meets the conditions, and the training generation model is used as the video generation model. In this way, the modeling accuracy of the video generation model can be effectively ensured. When the sample guidance text, sample video data, sample sub-information, and timing information are processed based on the video generation model to obtain the guidance video, the efficiency and accuracy of obtaining the guidance video can be effectively improved, and the interaction effect can be improved.

[0093] Among them, the sample guidance text, sample video data, sample sub-information, timing information, and annotated guidance video are all used to train the video generation model. The annotated guidance video is used as a guidance video that conforms to the sample object information. The annotated guidance video can be pre-annotated. The number of sample guidance texts can be one or more, and the annotated guidance video can be used as a reference for judging whether the video generation model has converged. For example, the annotated guidance video and the loss value between the predicted guidance video output by the video generation model based on the sample guidance text, sample video data, sample sub-information, and timing information during the training process can be used, and when the loss value is less than or equal to the loss threshold, it is determined that the loss value between the predicted guidance video output by the trained generation model and the annotated guidance video meets the condition, and when the loss value is greater than the loss threshold, it is determined that the loss value between the predicted guidance video output by the trained generation model and the annotated guidance video does not meet the condition, and then, when the loss condition is not met, the next round of training of the video generation model is triggered until the video generation model converges.

[0094] S208: Generate a response message corresponding to the request message according to the guidance video.

[0095] After the operation guidance text, object sub-information, initial video data, and timing information are input into the video generation model, and the guidance video output by the video generation model that conforms to the target object information is obtained, a response message corresponding to the request message can be generated based on the guidance video, and the response message can be fed back to the target object to provide intelligent customer service.

[0096] In some embodiments, in the process of generating a response message corresponding to a request message according to a guidance video, the guidance video may be directly included in the response message. That is, when the response message is fed back, the guidance video is intuitively fed back to the target object to intuitively guide the target object to complete the corresponding operation based on the operation instructions in the guidance video.

[0097] In some embodiments, in the process of generating a response message corresponding to a request message according to a guidance video, an access link of the guidance video may be generated and included in the response message. That is, the access link of the guidance video may be fed back to the target object through the response message, and if the target object clicks on the access link, the guidance video may be played to the target object to intuitively guide the target object to complete the corresponding operation based on the operation instructions in the guidance video.

[0098] In some embodiments, in the process of generating a response message corresponding to a request message based on a guidance video, prompt information may be generated and the guidance video and / or an access link to the guidance video may be included in the response message, wherein the prompt information is used to prompt the object to watch the guidance video.

[0099] S209: Send a response message.

[0100] After generating a response message corresponding to the request message according to the guidance video, the response message may be sent to respond to the target query information in the request message.

[0101] In this embodiment, by receiving a request message in an interactive scene, wherein the request message includes target query information, and obtaining target object information of an object to which the target query information belongs, feature extraction is performed on the target query information to obtain target query semantics, and based on the target query semantics, target feedback information corresponding to the target query information is determined, wherein the target feedback information includes at least one operation guide text, and initial video data corresponding to the operation guide text is obtained, wherein the initial video data has corresponding timing information, object sub-information corresponding to the operation guide text is obtained from the target object information, and at least one operation guide text, at least one object sub-information, and at least one frame of initial video data are input into a video generation model, so that the video generation model generates target video data according to the operation guide text, the object sub-information, and the initial video data corresponding to the operation guide text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data. Thus, it supports the generation of a guidance video that conforms to the target object information, and because the guidance video can be adapted to the target object information of the object to which the target query information belongs, when the operation guidance is performed based on the guidance video that conforms to the target object information, the intuitiveness of the operation guidance can be improved, thereby improving the interaction effect. The modeling accuracy of the operation guidance text generation model can be effectively ensured. When the query semantic features are processed based on the operation guidance text generation model to obtain at least one operation guidance text, the acquisition efficiency and accuracy of the target feedback information can be effectively improved, and the interaction effect can be improved. In addition, the video generation quality of the video generation model can be effectively improved, and the adaptation degree of the guidance video and the target object information can be further improved. The modeling accuracy of the video generation model can be effectively ensured. When the operation guidance text, object sub-information, initial video data, and timing information are processed based on the video generation model to obtain the guidance video, the acquisition efficiency and accuracy of the guidance video can be effectively improved, and the interaction effect can be improved.

[0102] The examples for the above embodiment are as follows:

[0103] It can be applied in the scenario of intelligent customer service of financial services, with customers (users) as examples. Thus, intelligent customer service of financial services can provide customers with intuitive video operation guidance (an optional example of guidance video) generated based on the actual situation of the customers.

[0104] In the disclosed embodiment, two models may be used.

[0105] Model 1 (an optional example of an operation guide text generation model):

[0106] Through a large amount of user historical data and training samples (sample data used to train the operation guide text generation model), supervised learning and other methods can be used to train a model that takes customer questions (an optional example of the target inquiry information) as input and text responses (one or more optional examples of operation guide texts) as output.

[0107] Model 2 (an optional example of a video generation model):

[0108] A model can be trained to generate a guide video based on the text response, the customer's actual situation information (an optional example of target object information).

[0109] A deep generative model (such as a diffusion model) can be used. The diffusion model is divided into a forward process and a reverse process. The former converts the data into a simple prior distribution, and the latter uses a trained neural network to simulate a fractional equation to generate data.

[0110] It contains the following three parts:

[0111] A text reply of the operation instructions (one or more optional examples of the operation instruction text), such as "Click on the credit card, and after the page jumps, click on installment repayment."

[0112] The customer's actual situation information (an optional example of target object information), such as the customer's account status, the number of cards, card numbers, etc.

[0113] An operation instruction video that matches the actual situation of the customer (an optional example of an instruction video that conforms to the target object information).

[0114] Take a sample as an example:

[0115] A text reply to the operation instructions (one or more optional examples of operation instruction texts): "Click on the credit card, and after the page jumps, click on installment repayment."

[0116] The customer’s actual information: one credit card with the card number 662*******5588.

[0117] Operation guide video: The video first shows the account status. There is a credit card in the interface with the card number 662*******5588. After clicking the credit card with the mouse, the page jumps to the credit card details page. The "Installment Repayment" button appears in the interface, and the mouse clicks the button.

[0118] like Figure 3 As shown, Figure 3: is a schematic diagram of the model training process in the embodiment of the present disclosure. Among them, the text-based financial service question-and-answer model is an optional example of the operation guide text generation model. The video generation model based on the operation guide and customer information is an optional example of the above video generation model. Customer information is an optional example of target object information.

[0119] like Figure 4 As shown, Figure 4 It is a schematic diagram of the video generation process in the embodiment of the present disclosure.

[0120] The guidance video generation method provided by the disclosed embodiment makes the video response form richer and more intuitive. It can solve the situation where customers cannot understand the customer service response due to the complexity of the financial service scenario or lack of relevant knowledge. The financial service intelligent customer service can generate a corresponding guidance video based on the customer's question, the customer's actual situation and the answer to the question (an optional example of target feedback information), so that customers can more easily understand the answer to the question.

[0121] Figure 5 A schematic diagram of the structure of a guidance video generating device provided in an embodiment of the present disclosure.

[0122] like Figure 5 As shown, the guidance video generating device 50 includes:

[0123] The receiving module 501 is used to receive a request message in an interaction scenario, wherein the request message includes target query information.

[0124] The first acquisition module 502 is used to acquire target object information of the object to which the target query information belongs.

[0125] The extraction module 503 is used to extract features from the target query information to obtain the target query semantics.

[0126] The determination module 504 is configured to determine target feedback information corresponding to the target query information based on the target query semantics, wherein the target feedback information includes at least one operation guidance text.

[0127] The second acquisition module 505 is used to acquire initial video data corresponding to the operation guide text, wherein the initial video data has corresponding timing information.

[0128] The third acquisition module 506 is used to acquire object sub-information corresponding to the operation guide text from the target object information.

[0129] The generation module 507 is used to input at least one operation instruction text, at least one object sub-information, and at least one frame of initial video data into the video generation model, so that the video generation model generates target video data according to the operation instruction text, the object sub-information, and the initial video data corresponding to the operation instruction text, and synthesizes the instruction video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data.

[0130] It should be noted that the above explanation of the guidance video generating method is also applicable to the guidance video generating device of this embodiment, and will not be repeated here.

[0131] In this embodiment, by receiving a request message in an interactive scene, wherein the request message includes target query information, and obtaining target object information of an object to which the target query information belongs, feature extraction is performed on the target query information to obtain target query semantics, and based on the target query semantics, target feedback information corresponding to the target query information is determined, wherein the target feedback information includes at least one operation guide text, and initial video data corresponding to the operation guide text is obtained, wherein the initial video data has corresponding timing information, object sub-information corresponding to the operation guide text is obtained from the target object information, and at least one operation guide text, at least one object sub-information, and at least one frame of initial video data are input into a video generation model, so that the video generation model generates target video data according to the operation guide text, the object sub-information, and the initial video data corresponding to the operation guide text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of target video data and the timing information corresponding to each frame of target video data. This supports the generation of a guidance video that conforms to the target object information, and because the guidance video can be adapted to the target object information of the object to which the target query information belongs, when operation guidance is performed based on the guidance video that conforms to the target object information, the intuitiveness of the operation guidance can be improved, thereby improving the interaction effect.

[0132] As an example, Figure 6 is a schematic diagram of the structure of an electronic device shown in an exemplary embodiment of the present disclosure, such as Figure 6 As shown, the electronic device 700 may further include:

[0133] The memory 710 and the processor 720, and the bus 730 connecting different components (including the memory 710 and the processor 720), the memory 710 stores a computer program, and when the processor 720 executes the program, the guidance video generation method of the embodiment of the present disclosure is implemented.

[0134] Bus 730 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

[0135] The electronic device 700 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by the electronic device 700, including volatile and non-volatile media, removable and non-removable media.

[0136] The memory 710 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 740 and / or cache memory 750. The electronic device 700 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 760 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 730 via one or more data medium interfaces. The memory 710 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.

[0137] A program / utility 780 having a set (at least one) of program modules 770 may be stored, for example, in the memory 710, such program modules 770 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 770 generally perform the functions and / or methods of the embodiments described in the present disclosure.

[0138] The electronic device 700 may also communicate with one or more external devices 790 (e.g., keyboard, pointing device, display 791, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 700, and / or communicate with any device that enables the electronic device 700 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 792. Furthermore, the electronic device 700 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) through a network adapter 793. As shown, the network adapter 793 communicates with other modules of the electronic device 700 through a bus 730. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0139] The processor 720 executes various functional applications and data processing by running the programs stored in the memory 710 .

[0140] It should be noted that the implementation process and technical principles of the electronic device of this embodiment refer to the aforementioned explanation of the guidance video generation method of the embodiment of the present disclosure, and will not be repeated here.

[0141] In order to implement the above embodiments, the present disclosure also proposes an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0142] In order to implement the above embodiments, the present disclosure further proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0143] In order to implement the above embodiments, the present disclosure also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0144] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this disclosure shall comply with the relevant laws and regulations and shall not violate public order and good morals.

[0145] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0146] The present disclosure anticipates providing implementation schemes for users to selectively block the use or access of personal information data. That is, the present disclosure anticipates providing hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, risks can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0147] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0148] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0149] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present disclosure belong.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0151] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0152] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0153] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0154] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present disclosure. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present disclosure.

Claims

1. A method for generating a guidance video, characterized in that: The following steps are involved: Receiving a request message in an interaction scenario, wherein the request message includes target query information; Acquire target object information of the object to which the target query information belongs; Extracting features of the target query information to obtain target query semantics; Based on the target query semantics, determining target feedback information corresponding to the target query information, wherein the target feedback information includes at least one operation guidance text; Acquire initial video data corresponding to the operation guide text, wherein the initial video data has corresponding timing information; Acquiring object sub-information corresponding to the operation guide text from the target object information; and The at least one operation instruction text, at least one object sub-information, and at least one frame of the initial video data are input into a video generation model, so that the video generation model generates target video data according to the operation instruction text, the object sub-information, and the initial video data corresponding to the operation instruction text, and synthesizes a instruction video that conforms to the target object information according to at least one frame of the target video data and the timing information corresponding to each frame of the target video data.

2. The method according to claim 1, characterized in that The determining, based on the target query semantics, target feedback information corresponding to the target query information includes: The query semantic feature is input into an operation instruction text generation model to obtain the target feedback information output by the operation instruction text generation model, wherein the operation instruction text generation model has modeled and learned the mapping relationship between the query semantic feature and at least one operation instruction text in the target feedback information.

3. The method according to claim 2, characterized in that The operation guide text generation model is trained based on the following method: Acquire historical customer service data, wherein the historical customer service data includes: sample inquiry information, and annotation feedback information corresponding to the sample inquiry information, wherein the annotation feedback information includes at least one annotation guidance text; Extracting features from the sample query information to obtain sample query semantics; The initial text generation model is iteratively trained at least once according to the sample query semantics until a loss condition is satisfied between at least one predicted guidance text in the predicted feedback information obtained based on the trained text generation model and the at least one labeled guidance text, and the trained text generation model is used as the operation guidance text generation model.

4. The method according to claim 1, characterized in that: The video generation model includes: a text feature extraction submodule, an information feature extraction submodule and a video generation submodule; The video generation model generates target video data according to the operation guide text, the object sub-information and the initial video data corresponding to the operation guide text, and synthesizes the guide video that conforms to the target object information according to at least one frame of the target video data and the timing information corresponding to each frame of the target video data, including: Determine a text feature vector corresponding to each of the operation guide texts based on the text feature extraction submodule; Determine the information feature vector of the object sub-information based on the information feature extraction submodule; Based on the video generation submodule, the text feature vector, the information feature vector and the initial video data corresponding to the operation guidance text are processed to obtain the target video data, and based on the timing information corresponding to each frame of the target video data and multiple frames of the target video data, a guidance video that conforms to the target object information is synthesized.

5. The method according to claim 1 or 4, characterized in that: The video generation model is trained based on the following method: Acquire at least one sample guidance text and sample object information, wherein the at least one sample guidance text and the sample object information have corresponding annotated guidance videos; Acquire sample video data corresponding to the sample guide text, wherein the sample video data has corresponding timing information; Acquire sample sub-information corresponding to the sample guidance text from the sample object information; The initial generation model is iteratively trained at least once according to the at least one sample guidance text, at least one frame of the sample video data, and at least one sample sub-information, until the loss value between the predicted guidance video output by the trained generation model and the labeled guidance video meets a condition, and the trained generation model is used as the video generation model.

6. The method according to claim 1, characterized in that The method further comprises: Generating a response message corresponding to the request message according to the guidance video; The response message is sent.

7. The method according to claim 6, characterized in that The step of generating a response message corresponding to the request message according to the guidance video includes any one of the following: Including the guidance video in the response message; Generate an access link for the guidance video, and include the access link in the response message; Generate prompt information, and include the guidance video and / or the access link of the guidance video in the response message, wherein the prompt information is used to prompt the object to watch the guidance video.

8. A guidance video generating device, characterized in that: include: A receiving module, used to receive a request message in an interaction scenario, wherein the request message includes target query information; A first acquisition module, used to acquire target object information of the object to which the target query information belongs; An extraction module, used for extracting features from the target query information to obtain target query semantics; a determination module, configured to determine target feedback information corresponding to the target query information based on the target query semantics, wherein the target feedback information includes at least one operation guidance text; and A second acquisition module is used to acquire initial video data corresponding to the operation guide text, wherein the initial video data has corresponding timing information; A third acquisition module is used to acquire object sub-information corresponding to the operation guide text from the target object information; A generation module is used to input the at least one operation instruction text, at least one object sub-information, and at least one frame of the initial video data into a video generation model, so that the video generation model generates target video data according to the operation instruction text, the object sub-information, and the initial video data corresponding to the operation instruction text, and synthesizes a guidance video that conforms to the target object information according to at least one frame of the target video data and the timing information corresponding to each frame of the target video data.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when being executed by a processor.

Images